Re: [time-nuts] Strange event on my Z3805A [REVISED to correct links]
Matt, It's not easy to see but if you look at the 72h plot, you can see that the EFC voltage stops at a low value before the event and then continues at a higher voltage after it. Look at the EFC level values before and after. This is the first time I've seen anything of the magnitude on the Z3805A in almost a year. As you can see from the graph showing the 36h prior to the event, there are many changes in the number of sats and there would be many changes of different sats that are tracked over that period but without any events like this. You can see a lot of spikes in the TI graph indicating the switching of the sats. This is the magnitude I expect to see but when you compare those spikes on the 72h graph you can see that this is a very very large and inexplicable event. Steve On 2 February 2010 16:58, Matt Osborn kc0...@msosborn.com wrote: Steve, I'm much more wizened than wise, but my understanding is that crystal jumps do not come back. They stay at their new frequency. I did notice on your graph of the event that there were changes in the sat. constellation at the beginning and end of the anomaly. I've seen similar moves on both my Thunderbolt and Fury GPSDOs. All seem to be related to satellite switching. One was so repetitive I was able to identify the offending satellite. On Tue, 2 Feb 2010 15:00:50 +1300, Steve Rooke sar10...@gmail.com wrote: On 2 February 2010 11:11, Matt Osborn kc0...@msosborn.com wrote: I've seen similar results when the receiver switches satellites. Multipath?? It does that all the time as sats come in and out of view but I've never seen an event of this magnitude before. If you look at the trace for the preceeding 36h you can see peaks where the unit switches stats but those peaks are decades less than this peak. I also see that the EFC voltage changes very significantly, is this perhaps one of the infamous crystal jumps we have spoken about I wonder. The area of interest is that it appears somewhat unstable after the event but is improving now. Multipath, I don't think so as nothing has changed here and I live in an urban area well outside the city. Nothing has been errected or changed as far as I can see. Cheers, Steve -- kc0ukk at msosborn dot com ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. -- Steve Rooke - ZL3TUV G8KVD A man with one clock knows what time it is; A man with two clocks is never quite sure. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Triangle Waves
Hi The DAC will quite nicely produce a trapezoid (or clipped triangle wave). It's certainly on the list. --- I've had a lot of lunch time discussions with the NIST guys about their obsession with input levels. About all I can say is that I don't see the same sensitivities they do. I suspect a lot of the issue is that we're not using the exact same circuits / components. - Matching the isolation amplifier to the mixer for efficient power transfer is something I do plan to look at. I suspect it's only good up to a certain point and then you get into trouble. - Lots of things to check ... Bob On Feb 2, 2010, at 12:06 AM, Bruce Griffiths wrote: Bob Camp wrote: Hi At least from the last time I tried it: If you use a sine wave input source, it's got to be an amazingly good 10 Hz sine wave. A normal audio generator will not produce a 10 Hz output with good enough short term stability / noise to give you useful data. Audio generators may be out there that will do the job, but I certainly don't have one, and have never come across one. Since the output of the mixer is basically a triangle wave, it makes sense to use that as your test source. A triangle wave also has the nice property that it's easy on the math. You don't have any approximation issues with the integers going into the DAC. That shoves the inevitable digital crud higher in frequency. When both the RF and LO ports are saturated, the mixer output waveform depends on how the IF port is terminated. The output is indeed approximately triangular with your IF port termination method when both the RF and LO ports are saturated. With the IF port terminated in a capacitor when both RF and LO ports are saturated the output waveform is quasi trapezoidal. When only the LO port is saturated the IF output is sinusoidal. Another nice thing about a pure digital approach is that it provides a clean trigger for the start channel of the counter you are testing things with. You can even set up the DAC to put out square waves to see just how good various bits of the chain are. Tough to do that with anything other than another arbitrary function generator. I agree that the reference is going to be an issue and that a LED stack may be the way to go. No matter how you generate the test tone, power supply noise will be an issue. The output amplifier on the DAC is my biggest worry. I could go with a current out DAC and something like an OP-27. That won't give me 1nV/Hz either, but it will at least be within shouting distance of it. Sigma deltas might be a third option. I have no idea what their low frequency flicker noise looks like. Producing a high amplitude (eg 20V pp) output and attenuating it down to say 2V pp or so typical of a mixer will significantly reduce the noise due to the output amplifier. So, other than the noise issue (which obviously needs to be analyzed / tested / pounded on) any other issues with the approach? At least from what I have seen in the past, level sensitivity on the inputs shows up pretty fast in the output beat note as you vary the input signals that are supposed to be saturating the mixer. If they are doing their job, a 2 db level change produces a very small change in the output. If you have something amiss in that department, you will see it pretty fast. On that I'm pretty much in agreement with Rubiola's stuff. Yes but NIST used a saturated mixer and still found that the mixer phase shift depended on how hard you drive the diodes. Long term variations in isolation amplifier output due to temperature variations may be significant. Since I intend to mate the isolation amps up directly on the same board as the mixer, there is no real need for a 50 ohm interface between them. If the mixer looks like 18.26 ohms, the amp output can be transformed to that level rather than 50 ohms. Everything is matched (over a 1/8 trace) and you don't burn up power in a bunch of resistors. How well that idea works - time will tell. It's easy to put the resistors in if it flunks out. So many things to try Bob You could also try driving the mixer ports from a highe impedance source (eg transistor collector). One early NIST paper advocated this. Bruce On Feb 1, 2010, at 9:48 PM, Bruce Griffiths wrote: Bob Camp wrote: Hi Ok, next up on the dual mixer stuff is checking the limiter chain. To do that with any chance of the results meaning anything you need a good triangle wave. You certainly can build some pretty complex gizmos to make them. There also appears to be a fairly simple approach. If I take a fairly good 16 bit DAC that will accept a clock a bit above 1 MHz, I can feed a simple count up / count down into it. That should give me a triangle wave at (clock rate) / 2^32. Simply put, 1.3 MHz data gives me a 10 Hz triangle wave. The digital crud
Re: [time-nuts] Test Equipment
Hi The 2.x firmware is the magic that lets it run down to HF type frequencies. You really do not want to upgrade the firmware. It would be very nice to find a back shelf somewhere with a set of original manuals for the 2.x version. Now if it just had a SR-620 counter built into it Bob On Feb 2, 2010, at 1:20 AM, Don Latham wrote: Oh, forgot. My firmware appears to be something like A.02.4 or something like that, and the manuals are A.05.0 or so. another possible problem. Don - Original Message - From: Bob Camp li...@cq.nu To: Discussion of precise time and frequency measurement time-nuts@febo.com Sent: Monday, February 01, 2010 6:42 PM Subject: Re: [time-nuts] Test Equipment Hi The big question in my mind about these is how well they do below 30 MHz. Until I know I can trust them it down there, I'm not selling very much stuff. I have yet to find a data sheet from before 2000 when they shipped with the sub-800 MHz stuff enabled. I get the impression that the E8285's never quite did as well below 30 MHz as the 8920's do. Bob On Feb 1, 2010, at 7:29 PM, Don Latham wrote: Hi Bob. Display shows use, definitely, but I can see info over the whole tube. I have a couple of things to do before I can start the learning curve, but am looking forward to using it. I'll have a bunch of test stuff for sale if this thing works out ;-) Don Bob Camp Hi I *suspect* that any size that was common in 2003 will be ok. I have no basis for that claim. That likely will limit you to 2 gig and down. Each time I called Amtronix, Rick answered on the first ring. It's definitely someplace I would recommend dealing with. How's the display on your unit? That sees to be the weakness of a lot of test gear these days. Can't wait to measure -100 dbc/Hz phase noise with mine :) Bob On Feb 1, 2010, at 5:34 PM, Don Latham wrote: Just bought one last week. As advertised, came with a cal and checkout sheet. BTW, cost another $150 to have manuals printed out. But, I'm old fashioned and have a hard time using manuals onscreen... I also got the feeling (phone order) that I can call Amtronix and at least reach a Real Person who will talk to me. I think the E8285A will replace at least three present instruments with better, once I master Instrument Basic :-). Does anyone know which low-cost PCMCIA memory card will work? They're on Epay for as little as $10 Don Bob Camp Hi I do believe the last (or maybe next to last) of the Amtronix E8285A's is now on it's way to a basement in Pennsylvania. If anybody else here is looking for one, I'd sure call Rick pretty quick. Bob -Original Message- From: time-nuts-boun...@febo.com [mailto:time-nuts-boun...@febo.com] On Behalf Of John Ackermann N8UR Sent: Thursday, January 21, 2010 4:14 PM To: john.fo...@gmail.com; Discussion of precise time and frequency measurement Subject: Re: [time-nuts] Test Equipment If RF measurement is your bag, and you're able to spend a couple of $K (actually, $2K if what I've seen recently holds), consider a communications service monitor like the HP 8920A/8920B/8921/8924/8935/E8285 (all pretty much the same thing). You get an RF generator, RF power meter, RX frequency meter and modulation analyzer, audio generator, audio analyzer, digital o'scope, and in most units a spectrum analyzer (many have a tracking generator, too) in one box. And I've probably forgotten a few things. If you get one with spec analyzer and tracking generator, there's software that lets you do swept insertion/return loss and cable fault finding. None of its capabilities are as good as those of a dedicated box performing a single function, but they're good enough for the vast majority of uses. An 8920 was the first significant piece of test gear I bought, and if I ever have to sell out, it'll be the last one to go. The prices came down a lot when Lucent surplused hundreds (thousands?) of them from their portable and cell phone production lines. I saw an 8935 with spec an, fully functional (as far as I could tell) for about $1500 this summer. A guy who sells and services a lot of these boxes is Rick at http://www.amtronix.com -- that web site will give you lots of info about the various versions and options. (I just noticed he has some Agilent 8285As as a hobbyist special with spec an and tracking generator for $650. That looks like a deal.) John john.fo...@gmail.com said the following on 01/21/2010 03:43 PM: Just that John, I'm looking to setup a general purpose lab. I'd lean towards RF type stuff since I'm a HAM. Sent via BlackBerry by ATT -Original Message- From: J. Forster j...@quik.com Date: Thu, 21 Jan 2010 12:22:11 To: did...@cox.net; Discussion of precise time and frequency measurementtime-nuts@febo.com Subject: Re: [time-nuts] Test Equipment I made the mistake of setting up a purpose-built lab once,
Re: [time-nuts] Test Equipment -Memory cards
Hi Don, Congratulations on the E8285A. I've an 8924C that does me nicely and came with a bunch of other stuff including two 10811A's and a crystal impedance meter (gotta keep on-topic) for £300 (~$500). Another useful instrument in the range that can sometimes be picked up cheaply is the 8922X if you get the 06 or 106 option you get a nice 1GHz digital SA with TG, a CW RF generator and low frequency scope. The GSM test stuff is an unwanted extra. A bit big, but better than a 141T setup. I think you will find that the E8285A is the same as the 8924C and uses non-volatile RAM cards, not flash. These cards are rare now and have CMOS ram and a lithium coin cell. You MIGHT be able to read a flash card, but I'm pretty certain you can't write to them. Robert G8RPI. --- On Mon, 1/2/10, Don Latham d...@montana.com wrote: From: Don Latham d...@montana.com Subject: Re: [time-nuts] Test Equipment To: Discussion of precise time and frequency measurement time-nuts@febo.com Date: Monday, 1 February, 2010, 22:34 Just bought one last week. As advertised, came with a cal and checkout sheet. BTW, cost another $150 to have manuals printed out. But, I'm old fashioned and have a hard time using manuals onscreen... I also got the feeling (phone order) that I can call Amtronix and at least reach a Real Person who will talk to me. I think the E8285A will replace at least three present instruments with better, once I master Instrument Basic :-). Does anyone know which low-cost PCMCIA memory card will work? They're on Epay for as little as $10 Don Bob Camp Hi I do believe the last (or maybe next to last) of the Amtronix E8285A's is now on it's way to a basement in Pennsylvania. If anybody else here is looking for one, I'd sure call Rick pretty quick. Bob -Original Message- From: time-nuts-boun...@febo.com [mailto:time-nuts-boun...@febo.com] On Behalf Of John Ackermann N8UR Sent: Thursday, January 21, 2010 4:14 PM To: john.fo...@gmail.com; Discussion of precise time and frequency measurement Subject: Re: [time-nuts] Test Equipment If RF measurement is your bag, and you're able to spend a couple of $K (actually, $2K if what I've seen recently holds), consider a communications service monitor like the HP 8920A/8920B/8921/8924/8935/E8285 (all pretty much the same thing). You get an RF generator, RF power meter, RX frequency meter and modulation analyzer, audio generator, audio analyzer, digital o'scope, and in most units a spectrum analyzer (many have a tracking generator, too) in one box. And I've probably forgotten a few things. If you get one with spec analyzer and tracking generator, there's software that lets you do swept insertion/return loss and cable fault finding. None of its capabilities are as good as those of a dedicated box performing a single function, but they're good enough for the vast majority of uses. An 8920 was the first significant piece of test gear I bought, and if I ever have to sell out, it'll be the last one to go. The prices came down a lot when Lucent surplused hundreds (thousands?) of them from their portable and cell phone production lines. I saw an 8935 with spec an, fully functional (as far as I could tell) for about $1500 this summer. A guy who sells and services a lot of these boxes is Rick at http://www.amtronix.com -- that web site will give you lots of info about the various versions and options. (I just noticed he has some Agilent 8285As as a hobbyist special with spec an and tracking generator for $650. That looks like a deal.) John john.fo...@gmail.com said the following on 01/21/2010 03:43 PM: Just that John, I'm looking to setup a general purpose lab. I'd lean towards RF type stuff since I'm a HAM. Sent via BlackBerry by ATT -Original Message- From: J. Forster j...@quik.com Date: Thu, 21 Jan 2010 12:22:11 To: did...@cox.net; Discussion of precise time and frequency measurementtime-nuts@febo.com Subject: Re: [time-nuts] Test Equipment I made the mistake of setting up a purpose-built lab once, and soon discovered to do most things, you really need at least a basic general purpose lab. The only exception is, I suppose, if you are going to repair a specific instrument and do nothing else. Also, as others have pointed out, you really need to define what are your area(s) of interest. Specialized gear gets $$$ pretty quickly. -John John, That sounds like asking what is the best vehicle for you to buy. If you do not know what you want to do with it, I am not sure we can help you all that much. However, if you have a specific objective, I am sure you will get a lot of valuable information here. You have a (good) analog scope, you may want a power supply or two, a soldering iron and maybe a desoldering station if you do surface mount. You also want a good hand-held multimeter. Some sort of signal or function generator may be useful too. These vary
Re: [time-nuts] Test Equipment
Bird wattmeters, such as the model 43 thruline, are far from accurate devices. They are spec'd to be +/- 5% of the full scale reading of the installed slug. That means for a 100W slug, the error band is +/- 5W! If you happen to read 20W on the meter, the error band says your true power could be anywhere from 15 to 25 watts! As a comparison, an HP 432A wattmeter can achieve an ultimate accuracy of +/-0.2% +/- 10uW. -Chuck Harris Don Latham wrote: Amtronix did tell me that the power measurements were off, according to some folks who had compared them to Birds. Of course the E's could be calibrated. I don't have much below 30 MHz at present either. I sense that there may be enough around to warrant a Yahoo or Google group??? Don - Original Message - From: Bob Camp li...@cq.nu To: Discussion of precise time and frequency measurement time-nuts@febo.com Sent: Monday, February 01, 2010 6:42 PM Subject: Re: [time-nuts] Test Equipment Hi The big question in my mind about these is how well they do below 30 MHz. Until I know I can trust them it down there, I'm not selling very much stuff. I have yet to find a data sheet from before 2000 when they shipped with the sub-800 MHz stuff enabled. I get the impression that the E8285's never quite did as well below 30 MHz as the 8920's do. Bob ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Test Equipment -Memory cards
At 07:42 AM 2/2/2010 , Robert Atkinson wrote: than a 141T setup. I think you will find that the E8285A is the same as the 8924C and uses non-volatile RAM cards, not flash. These cards are rare now and have CMOS ram and a lithium coin cell. You MIGHT be able to read a flash card, but I'm pretty certain you can't write to them. If you're looking for PCMCIA SRAM cards, they're also used with vintage Fanuc CNC controls. You might end up paying a premium, but they are out there. (On the other side, I'm using an el-cheapo PCMCIA / PCI adapter in a desktop PC, with the stock XP drivers. No problems yet.) -- newell N5TNL ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
[time-nuts] Aside about Triangle Waveforms
A widely used WW-II aircraft radio altimeter used a triangular waveform to FM modulate a 400 MHz oscillator, employing a mechanical variable capacitor constructed similar to a permanent-magnet loudspeaker. To get the capacitor's diaphragm to reverse accurately, at the positive peak of the trianglular waveform, required a sharp, negative-going impulse to be added to the peak of the triangle, creating a sharp notch in the waveform about 30% deep. This makes me wonder about the limitations of speaker cones attempting to reproduce complex waveforms. If they had overall feedback for positional correction, the spectrum of the resulting driving waveform might contain some pretty complex components. Bruce Hunter ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Standard Resistor oil
You want the stuff sold for laxative use, USP Mineral Oil, sometimes trademarked Nujol. You do NOT want Baby Oil. -John Hi Some drug store mineral oil has extra stuff in it. This is one case where you want the cheap generic version rather than the improved name brand. Bob On Feb 1, 2010, at 5:56 PM, Chuck Harris wrote: Plain old mineral oil, also known as paraffin oil, as can be found in any drugstore. They should be filled above the element, but not quite full. Maybe 3/4 to 7/8 full. It isn't really important. -Chuck Harris Chris Erickson wrote: I bought some old Leeds Northrup standard resistors on ebay and and seller dumped the oil out before sending them. Does anyone know what the correct kind of oil to put back in these? How full should they be? ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
[time-nuts] OT: Practical PLL low noise?
Hello all, sorry for the OT, but I know there're many real electronic artists here. As an amateur radio operator I often use transverters, some home made. They usually can be made sigthly better (RF and noise-wise) than japanese transceivers. However often the LO xtal oscillator drifts too much for comfortable digital and weak signal work. Now the big question: is there any PLL design that can lock 22 MHz and 42 MHz xtal oscillator to a 10 MHz reference (typically from a GPSDO) without adding significant noise to the oscillators? The LOs usually go to a single or doube balanced diode mixer like the famous minicircuit ones, and at that point the RF signal has been already amplified by 10 or 20 dB stage(s). Other options would be ovenizing the LOs or making a DDS sinth. Now, what would be more practical approach from the home construction point? Thanks Frank IZ8DWF ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] OT: Practical PLL low noise?
Hello Francesco: you can use my xlock, http://www.qslnet.de/member/on4iy/xlock/xlock.html You will have to add a varicap to make the XO into a VCXO. There should not be any noticable noise increase due to the PLL - make sure you use a good reference 10MHz in yr GPSDO! 73 xtof on4iy On 02/02/10 16:50, francesco messineo wrote: Hello all, sorry for the OT, but I know there're many real electronic artists here. As an amateur radio operator I often use transverters, some home made. They usually can be made sigthly better (RF and noise-wise) than japanese transceivers. However often the LO xtal oscillator drifts too much for comfortable digital and weak signal work. Now the big question: is there any PLL design that can lock 22 MHz and 42 MHz xtal oscillator to a 10 MHz reference (typically from a GPSDO) without adding significant noise to the oscillators? The LOs usually go to a single or doube balanced diode mixer like the famous minicircuit ones, and at that point the RF signal has been already amplified by 10 or 20 dB stage(s). Other options would be ovenizing the LOs or making a DDS sinth. Now, what would be more practical approach from the home construction point? Thanks Frank IZ8DWF ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] OT: Practical PLL low noise?
Hi The first issue - your oscillator may be drifting quite a lot. If so, that's the first thing to check and possibly fix. A reasonable oscillator should be able to hold less than 100 Hz at 42 MHz under normal room conditions. Fixes range from circuit improvements, to a better crystal, to simply eliminating a draft that blows on the oscillator. If the oscillator is reasonably stable, it will need to be turned into a VCXO in order to lock it. If both oscillators use fundamental crystals, that should not be very hard. If they use higher overtone crystals it may be more of a challenge. Often you will find a tradeoff between good oscillator performance and wide tuning range. What ever chip you use to do the lock, keep the loop bandwidth small. The GPSDO will be noisy and it will not help you for phase noise. I would start the bandwidth at 100 Hz to be sure everything works ok and then start narrowing it to 10 Hz or less. At some point the loop will be to narrow to keep up with the changes and you will not be able to maintain phase lock. What ever loop bandwidth you use, keep the phase margin large. You do not need a fast locking loop. Instead you need one that has less tendency to peak. Phase margins should be above 70 degrees. The nice thing about doing this with a chip is that most of the manufacturers have cute little web applications / free downloads to design the loop filters for you. No digging out crazy formulas and wondering if you got it all right. Have fun ! Bob KB8TQ -Original Message- From: time-nuts-boun...@febo.com [mailto:time-nuts-boun...@febo.com] On Behalf Of francesco messineo Sent: Tuesday, February 02, 2010 10:50 AM To: Discussion of precise time and frequency measurement Subject: [time-nuts] OT: Practical PLL low noise? Hello all, sorry for the OT, but I know there're many real electronic artists here. As an amateur radio operator I often use transverters, some home made. They usually can be made sigthly better (RF and noise-wise) than japanese transceivers. However often the LO xtal oscillator drifts too much for comfortable digital and weak signal work. Now the big question: is there any PLL design that can lock 22 MHz and 42 MHz xtal oscillator to a 10 MHz reference (typically from a GPSDO) without adding significant noise to the oscillators? The LOs usually go to a single or doube balanced diode mixer like the famous minicircuit ones, and at that point the RF signal has been already amplified by 10 or 20 dB stage(s). Other options would be ovenizing the LOs or making a DDS sinth. Now, what would be more practical approach from the home construction point? Thanks Frank IZ8DWF ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Test Equipment -Memory cards
Hi The memory in the E8285A has a lithium cell associated with it. One of my big questions is weather the firmware goes away when the coin cell dies (battery backed SRAM) or if the firmware is in something a bit more robust. Hopefully it's sitting on the porch when I get home tonight -- 141T ... how many do you need in addition to an E8285 ... all sorts of questions to be answered. Bob -Original Message- From: time-nuts-boun...@febo.com [mailto:time-nuts-boun...@febo.com] On Behalf Of Robert Atkinson Sent: Tuesday, February 02, 2010 8:42 AM To: Discussion of precise time and frequency measurement Subject: Re: [time-nuts] Test Equipment -Memory cards Hi Don, Congratulations on the E8285A. I've an 8924C that does me nicely and came with a bunch of other stuff including two 10811A's and a crystal impedance meter (gotta keep on-topic) for £300 (~$500). Another useful instrument in the range that can sometimes be picked up cheaply is the 8922X if you get the 06 or 106 option you get a nice 1GHz digital SA with TG, a CW RF generator and low frequency scope. The GSM test stuff is an unwanted extra. A bit big, but better than a 141T setup. I think you will find that the E8285A is the same as the 8924C and uses non-volatile RAM cards, not flash. These cards are rare now and have CMOS ram and a lithium coin cell. You MIGHT be able to read a flash card, but I'm pretty certain you can't write to them. Robert G8RPI. --- On Mon, 1/2/10, Don Latham d...@montana.com wrote: From: Don Latham d...@montana.com Subject: Re: [time-nuts] Test Equipment To: Discussion of precise time and frequency measurement time-nuts@febo.com Date: Monday, 1 February, 2010, 22:34 Just bought one last week. As advertised, came with a cal and checkout sheet. BTW, cost another $150 to have manuals printed out. But, I'm old fashioned and have a hard time using manuals onscreen... I also got the feeling (phone order) that I can call Amtronix and at least reach a Real Person who will talk to me. I think the E8285A will replace at least three present instruments with better, once I master Instrument Basic :-). Does anyone know which low-cost PCMCIA memory card will work? They're on Epay for as little as $10 Don Bob Camp Hi I do believe the last (or maybe next to last) of the Amtronix E8285A's is now on it's way to a basement in Pennsylvania. If anybody else here is looking for one, I'd sure call Rick pretty quick. Bob -Original Message- From: time-nuts-boun...@febo.com [mailto:time-nuts-boun...@febo.com] On Behalf Of John Ackermann N8UR Sent: Thursday, January 21, 2010 4:14 PM To: john.fo...@gmail.com; Discussion of precise time and frequency measurement Subject: Re: [time-nuts] Test Equipment If RF measurement is your bag, and you're able to spend a couple of $K (actually, $2K if what I've seen recently holds), consider a communications service monitor like the HP 8920A/8920B/8921/8924/8935/E8285 (all pretty much the same thing). You get an RF generator, RF power meter, RX frequency meter and modulation analyzer, audio generator, audio analyzer, digital o'scope, and in most units a spectrum analyzer (many have a tracking generator, too) in one box. And I've probably forgotten a few things. If you get one with spec analyzer and tracking generator, there's software that lets you do swept insertion/return loss and cable fault finding. None of its capabilities are as good as those of a dedicated box performing a single function, but they're good enough for the vast majority of uses. An 8920 was the first significant piece of test gear I bought, and if I ever have to sell out, it'll be the last one to go. The prices came down a lot when Lucent surplused hundreds (thousands?) of them from their portable and cell phone production lines. I saw an 8935 with spec an, fully functional (as far as I could tell) for about $1500 this summer. A guy who sells and services a lot of these boxes is Rick at http://www.amtronix.com -- that web site will give you lots of info about the various versions and options. (I just noticed he has some Agilent 8285As as a hobbyist special with spec an and tracking generator for $650. That looks like a deal.) John john.fo...@gmail.com said the following on 01/21/2010 03:43 PM: Just that John, I'm looking to setup a general purpose lab. I'd lean towards RF type stuff since I'm a HAM. Sent via BlackBerry by ATT -Original Message- From: J. Forster j...@quik.com Date: Thu, 21 Jan 2010 12:22:11 To: did...@cox.net; Discussion of precise time and frequency measurementtime-nuts@febo.com Subject: Re: [time-nuts] Test Equipment I made the mistake of setting up a purpose-built lab once, and soon discovered to do most things, you really need at least a basic general purpose lab. The only exception is, I suppose, if you are going to repair a specific instrument and do nothing else. Also, as others have pointed out, you
Re: [time-nuts] Lady Heather Daemon for FreeBSD
Hi Sounds *very* useful. Which version(s) of FreeBSD have you tried it on? Thanks! Bob -Original Message- From: time-nuts-boun...@febo.com [mailto:time-nuts-boun...@febo.com] On Behalf Of Ralph Smith Sent: Tuesday, February 02, 2010 12:41 PM To: time-nuts@febo.com Subject: [time-nuts] Lady Heather Daemon for FreeBSD I have written a small daemon that attaches to a Thunderbolt and acts as a client for Lady Heather. In addition, if specified it will provide data to NTP through the shared memory driver. It currently compiles on FreeBSD, and OS X. Available at http://ralphsmith.org/~ralph/thunderbolt.tar.gz It is in a very early, rough stage at the moment, but I thought I would provide it for anyone that is interested. There are two programs: tboltd, and gpsdclient. First tboltd: Usage: tboltd [-t tty] [-v] [-p port] [-u unit] [-d] -t tty: Specify Thunderbolt serial port. Default 'cuau1'. -v:Increase verbosity level. -p port: UDP port to listen for client connections. Default 45000. -u unit: Unit number for NTP shared memory driver. Default to none -d:Do not detach and run in daemon mode. Some notes: - tboltd will currently accept as many clients as memory allows. - tboltd currently does not buffer requests from clients. Multiple clients risk having commands interleaved. - Logging needs to be implemented properly - tboltd blocks when there is no activity on the thunderbolt or client. On a Net4501 (133 MHz 486 equivalent), with 3 clients (1 Heather + 2 gpsdclient) it takes less than 1% CPU. - Likely to be bugs. - I invoke it on my Net4501 as tboltd -p 45000 -t cuau1 -u 0 - Corresponding ntp.conf entry # Use shared memory tos mindist 0.030 server 127.127.28.0 minpoll 4 maxpoll 4 prefer fudge 127.127.28.0 time1 0.0275 refid GPS Second: gpsdclient. gpsdclient will attach to tboltd or gpsd, receive Thunderbold TSIP packets, and drive an NTP shared memory segment. Usage: ./gpsdclient [-g] [-u unit] [-p port] host - The -g flag will sent the command to gpsd to put it in super-raw mode to deliver TSIP - Currently not configured to detach and run as a daemon, will fix that when I get around to it. This needs some cleanup and reorganization, but I thought some of you might be interested. Any comments or suggestions please let me know. -- Ralph Smith (AB4RS) ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Lady Heather Daemon for FreeBSD
On Tue, February 2, 2010 12:47 pm, Bob Camp wrote: Hi Sounds *very* useful. Which version(s) of FreeBSD have you tried it on? 8.0-STABLE, but it should be good for earlier versions. Ralph (AB4RS) ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] OT: Practical PLL low noise?
Hi Bob, On 2/2/10, Bob Camp li...@cq.nu wrote: Hi The first issue - your oscillator may be drifting quite a lot. If so, that's the first thing to check and possibly fix. A reasonable oscillator should be able to hold less than 100 Hz at 42 MHz under normal room conditions. Fixes range from circuit improvements, to a better crystal, to simply eliminating a draft that blows on the oscillator. it is drifting about 50 Hz during warm up, but the problem is thermal drift internally as season changes, as tx/rx periods change, and so on. If the oscillator is reasonably stable, it will need to be turned into a VCXO in order to lock it. If both oscillators use fundamental crystals, that should not be very hard. If they use higher overtone crystals it may be more of a challenge. Often you will find a tradeoff between good oscillator performance and wide tuning range. 22 MHz can be fundamental, 42 MHz is third overtone for sure. What ever chip you use to do the lock, keep the loop bandwidth small. The GPSDO will be noisy and it will not help you for phase noise. I would start the bandwidth at 100 Hz to be sure everything works ok and then start narrowing it to 10 Hz or less. At some point the loop will be to narrow to keep up with the changes and you will not be able to maintain phase lock. What ever loop bandwidth you use, keep the phase margin large. You do not need a fast locking loop. Instead you need one that has less tendency to peak. Phase margins should be above 70 degrees. The nice thing about doing this with a chip is that most of the manufacturers have cute little web applications / free downloads to design the loop filters for you. No digging out crazy formulas and wondering if you got it all right. thanks for the suggestions, any good candidate as a chip? 73 Frank IZ8DWF ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Lady Heather Daemon for FreeBSD
Hi I'm running 8 on most of my stuff now, so that should not be an issue. If you had come back with 6.1 that might have been reason to stop and think a bit. Does the program take care of all the serial line setup stuff, or are there links and stuff that need to be done to get it to talk right? Some of the NTP serial drivers seem to require a bit of a massage on the serial settings to make them happy. Bob -Original Message- From: time-nuts-boun...@febo.com [mailto:time-nuts-boun...@febo.com] On Behalf Of Ralph Smith Sent: Tuesday, February 02, 2010 12:54 PM To: Discussion of precise time and frequency measurement Subject: Re: [time-nuts] Lady Heather Daemon for FreeBSD On Tue, February 2, 2010 12:47 pm, Bob Camp wrote: Hi Sounds *very* useful. Which version(s) of FreeBSD have you tried it on? 8.0-STABLE, but it should be good for earlier versions. Ralph (AB4RS) ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Lady Heather Daemon for FreeBSD
On Tue, February 2, 2010 12:58 pm, Bob Camp wrote: Hi I'm running 8 on most of my stuff now, so that should not be an issue. If you had come back with 6.1 that might have been reason to stop and think a bit. Does the program take care of all the serial line setup stuff, or are there links and stuff that need to be done to get it to talk right? Some of the NTP serial drivers seem to require a bit of a massage on the serial settings to make them happy. Bob Nothing special required, it just blindly goes with the Thunderbolt's default configuration of 9600 8N1, and configures the serial line accordingly. Ralph ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
[time-nuts] Low noise PLL for transceiver locking
Frank, My suggestion would be to try injection locking, rather than a PLL. No change is made to the 22MHz and 42MHz oscillators, except to find a way to inject enough reference power to force them to lock to it. Injection locking works well with modest harmonic relationships, and gives good noise performance. The injection can be via a coupling link, or even at the cold end of an existing bypass capacitor. Narrow pulses often work best. I am confident that you could lock 22MHz to a 2MHz injection (divided from your 10MHz reference) if the 2MHz pulse was narrow enough and the 22MHz oscillator sufficiently stable. 42MHz is more of a challenge - you might need a double-step, such as first locking 7MHz to 1MHz from the reference, and using that to lock 42MHz. The literature on injection locked oscillators is quite limited. You get very interesting results as the lock drops out! 73, Murray ZL1BPU ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] OT: Practical PLL low noise?
Hi I happen to like the Analog Devices ADF4001 for this sort of thing. You would need two of them, one for each oscillator. The National chip mentioned earlier will also work. The 2306 it's self is obsolete, but I'm sure there are other National parts that will drop into the same socket. Either way you will need something like a PIC to shoot the settings into the chip. It looks like all that is taken care of on the board in the earlier post. The code on the PIC would need to be re-written to match up with what ever chip you decide to use. Bob -Original Message- From: time-nuts-boun...@febo.com [mailto:time-nuts-boun...@febo.com] On Behalf Of francesco messineo Sent: Tuesday, February 02, 2010 12:58 PM To: Discussion of precise time and frequency measurement Subject: Re: [time-nuts] OT: Practical PLL low noise? Hi Bob, On 2/2/10, Bob Camp li...@cq.nu wrote: Hi The first issue - your oscillator may be drifting quite a lot. If so, that's the first thing to check and possibly fix. A reasonable oscillator should be able to hold less than 100 Hz at 42 MHz under normal room conditions. Fixes range from circuit improvements, to a better crystal, to simply eliminating a draft that blows on the oscillator. it is drifting about 50 Hz during warm up, but the problem is thermal drift internally as season changes, as tx/rx periods change, and so on. If the oscillator is reasonably stable, it will need to be turned into a VCXO in order to lock it. If both oscillators use fundamental crystals, that should not be very hard. If they use higher overtone crystals it may be more of a challenge. Often you will find a tradeoff between good oscillator performance and wide tuning range. 22 MHz can be fundamental, 42 MHz is third overtone for sure. What ever chip you use to do the lock, keep the loop bandwidth small. The GPSDO will be noisy and it will not help you for phase noise. I would start the bandwidth at 100 Hz to be sure everything works ok and then start narrowing it to 10 Hz or less. At some point the loop will be to narrow to keep up with the changes and you will not be able to maintain phase lock. What ever loop bandwidth you use, keep the phase margin large. You do not need a fast locking loop. Instead you need one that has less tendency to peak. Phase margins should be above 70 degrees. The nice thing about doing this with a chip is that most of the manufacturers have cute little web applications / free downloads to design the loop filters for you. No digging out crazy formulas and wondering if you got it all right. thanks for the suggestions, any good candidate as a chip? 73 Frank IZ8DWF ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Lady Heather Daemon for FreeBSD
On Tue, February 2, 2010 12:53 pm, Ralph Smith wrote: On Tue, February 2, 2010 12:47 pm, Bob Camp wrote: Hi Sounds *very* useful. Which version(s) of FreeBSD have you tried it on? 8.0-STABLE, but it should be good for earlier versions. It should also port to Linux easily with minimal tweaking. I just don't have a Linux box available. Ralph ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Low noise PLL for transceiver locking
Hi Murray, On 2/2/10, Murray Greenman murray.green...@rakon.com wrote: Frank, My suggestion would be to try injection locking, rather than a PLL. No change is made to the 22MHz and 42MHz oscillators, except to find a way to inject enough reference power to force them to lock to it. Injection locking works well with modest harmonic relationships, and gives good noise performance. The injection can be via a coupling link, or even at the cold end of an existing bypass capacitor. Narrow pulses often work best. I am confident that you could lock 22MHz to a 2MHz injection (divided from your 10MHz reference) if the 2MHz pulse was narrow enough and the 22MHz oscillator sufficiently stable. 42MHz is more of a challenge - you might need a double-step, such as first locking 7MHz to 1MHz from the reference, and using that to lock 42MHz. I confess I've never heard about injection locking! Do you have any suggestion where to start from to learn about it? 73 Frank IZ8DWF ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Low noise PLL for transceiver locking
Hi Murray, On 2/2/10, Murray Greenman murray.green...@rakon.com wrote: Frank, My suggestion would be to try injection locking, rather than a PLL. No change is made to the 22MHz and 42MHz oscillators, except to find a way to inject enough reference power to force them to lock to it. Injection locking works well with modest harmonic relationships, and gives good noise performance. The injection can be via a coupling link, or even at the cold end of an existing bypass capacitor. Narrow pulses often work best. I am confident that you could lock 22MHz to a 2MHz injection (divided from your 10MHz reference) if the 2MHz pulse was narrow enough and the 22MHz oscillator sufficiently stable. 42MHz is more of a challenge - you might need a double-step, such as first locking 7MHz to 1MHz from the reference, and using that to lock 42MHz. I confess I've never heard about injection locking! Do you have any suggestion where to start from to learn about it? 73 Frank IZ8DWF ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Low noise PLL for transceiver locking
francesco messineo wrote: Hi Murray, On 2/2/10, Murray Greenmanmurray.green...@rakon.com wrote: Frank, My suggestion would be to try injection locking, rather than a PLL. No change is made to the 22MHz and 42MHz oscillators, except to find a way to inject enough reference power to force them to lock to it. Injection locking works well with modest harmonic relationships, and gives good noise performance. The injection can be via a coupling link, or even at the cold end of an existing bypass capacitor. Narrow pulses often work best. I am confident that you could lock 22MHz to a 2MHz injection (divided from your 10MHz reference) if the 2MHz pulse was narrow enough and the 22MHz oscillator sufficiently stable. 42MHz is more of a challenge - you might need a double-step, such as first locking 7MHz to 1MHz from the reference, and using that to lock 42MHz. I confess I've never heard about injection locking! Do you have any suggestion where to start from to learn about it? 73 Frank IZ8DWF You can start with the references listed at the bottom of the page: http://www.ko4bb.com/~bruce/InjectionLocking.html http://www.ko4bb.com/%7Ebruce/InjectionLocking.html Bruce ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Test Equipment
Hi Bob: I finally figured that out. I had the manuals printed, anyway. kinda added to the cost, but... Don Bob Camp Hi The 2.x firmware is the magic that lets it run down to HF type frequencies. You really do not want to upgrade the firmware. It would be very nice to find a back shelf somewhere with a set of original manuals for the 2.x version. Now if it just had a SR-620 counter built into it Bob On Feb 2, 2010, at 1:20 AM, Don Latham wrote: Oh, forgot. My firmware appears to be something like A.02.4 or something like that, and the manuals are A.05.0 or so. another possible problem. Don - Original Message - From: Bob Camp li...@cq.nu To: Discussion of precise time and frequency measurement time-nuts@febo.com Sent: Monday, February 01, 2010 6:42 PM Subject: Re: [time-nuts] Test Equipment Hi The big question in my mind about these is how well they do below 30 MHz. Until I know I can trust them it down there, I'm not selling very much stuff. I have yet to find a data sheet from before 2000 when they shipped with the sub-800 MHz stuff enabled. I get the impression that the E8285's never quite did as well below 30 MHz as the 8920's do. Bob On Feb 1, 2010, at 7:29 PM, Don Latham wrote: Hi Bob. Display shows use, definitely, but I can see info over the whole tube. I have a couple of things to do before I can start the learning curve, but am looking forward to using it. I'll have a bunch of test stuff for sale if this thing works out ;-) Don Bob Camp Hi I *suspect* that any size that was common in 2003 will be ok. I have no basis for that claim. That likely will limit you to 2 gig and down. Each time I called Amtronix, Rick answered on the first ring. It's definitely someplace I would recommend dealing with. How's the display on your unit? That sees to be the weakness of a lot of test gear these days. Can't wait to measure -100 dbc/Hz phase noise with mine :) Bob On Feb 1, 2010, at 5:34 PM, Don Latham wrote: Just bought one last week. As advertised, came with a cal and checkout sheet. BTW, cost another $150 to have manuals printed out. But, I'm old fashioned and have a hard time using manuals onscreen... I also got the feeling (phone order) that I can call Amtronix and at least reach a Real Person who will talk to me. I think the E8285A will replace at least three present instruments with better, once I master Instrument Basic :-). Does anyone know which low-cost PCMCIA memory card will work? They're on Epay for as little as $10 Don Bob Camp Hi I do believe the last (or maybe next to last) of the Amtronix E8285A's is now on it's way to a basement in Pennsylvania. If anybody else here is looking for one, I'd sure call Rick pretty quick. Bob -Original Message- From: time-nuts-boun...@febo.com [mailto:time-nuts-boun...@febo.com] On Behalf Of John Ackermann N8UR Sent: Thursday, January 21, 2010 4:14 PM To: john.fo...@gmail.com; Discussion of precise time and frequency measurement Subject: Re: [time-nuts] Test Equipment If RF measurement is your bag, and you're able to spend a couple of $K (actually, $2K if what I've seen recently holds), consider a communications service monitor like the HP 8920A/8920B/8921/8924/8935/E8285 (all pretty much the same thing). You get an RF generator, RF power meter, RX frequency meter and modulation analyzer, audio generator, audio analyzer, digital o'scope, and in most units a spectrum analyzer (many have a tracking generator, too) in one box. And I've probably forgotten a few things. If you get one with spec analyzer and tracking generator, there's software that lets you do swept insertion/return loss and cable fault finding. None of its capabilities are as good as those of a dedicated box performing a single function, but they're good enough for the vast majority of uses. An 8920 was the first significant piece of test gear I bought, and if I ever have to sell out, it'll be the last one to go. The prices came down a lot when Lucent surplused hundreds (thousands?) of them from their portable and cell phone production lines. I saw an 8935 with spec an, fully functional (as far as I could tell) for about $1500 this summer. A guy who sells and services a lot of these boxes is Rick at http://www.amtronix.com -- that web site will give you lots of info about the various versions and options. (I just noticed he has some Agilent 8285As as a hobbyist special with spec an and tracking generator for $650. That looks like a deal.) John john.fo...@gmail.com said the following on 01/21/2010 03:43 PM: Just that John, I'm looking to setup a general purpose lab. I'd lean towards RF type stuff since I'm a HAM. Sent via BlackBerry by ATT -Original Message- From: J. Forster j...@quik.com Date: Thu, 21 Jan 2010 12:22:11 To: did...@cox.net; Discussion of precise time and frequency
Re: [time-nuts] Triangle Waves
Bob Camp wrote: Hi The DAC will quite nicely produce a trapezoid (or clipped triangle wave). It's certainly on the list. --- I've had a lot of lunch time discussions with the NIST guys about their obsession with input levels. About all I can say is that I don't see the same sensitivities they do. I suspect a lot of the issue is that we're not using the exact same circuits / components. - Some of their papers are annoyingly incomplete, in that the measurement setups used were incompletely specified. Matching the isolation amplifier to the mixer for efficient power transfer is something I do plan to look at. I suspect it's only good up to a certain point and then you get into trouble. - Lots of things to check ... Bob Bruce ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Aside about Triangle Waveforms
Modern radar altimeters also use triangular wave FM modulation but at around 4.2GHz. Mix the return signal with a sample of the transmitter and you get an audio tone directly proportional to the round trip delay and thus height. works down to a few feet, pretty good for a real time time interval measurement. Some old techniques are hard to beat :-) Robert G8RPI. --- On Tue, 2/2/10, brucekar...@aol.com brucekar...@aol.com wrote: From: brucekar...@aol.com brucekar...@aol.com Subject: [time-nuts] Aside about Triangle Waveforms To: time-nuts@febo.com Date: Tuesday, 2 February, 2010, 15:14 A widely used WW-II aircraft radio altimeter used a triangular waveform to FM modulate a 400 MHz oscillator, employing a mechanical variable capacitor constructed similar to a permanent-magnet loudspeaker. To get the capacitor's diaphragm to reverse accurately, at the positive peak of the trianglular waveform, required a sharp, negative-going impulse to be added to the peak of the triangle, creating a sharp notch in the waveform about 30% deep. This makes me wonder about the limitations of speaker cones attempting to reproduce complex waveforms. If they had overall feedback for positional correction, the spectrum of the resulting driving waveform might contain some pretty complex components. Bruce Hunter ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
[time-nuts] FW: Injection locking
-Original Message- From: Murray Greenman Sent: Wednesday, 3 February 2010 9:00 a.m. To: 'time-nuts@febo.com' Subject: Injection locking Frank, Bruce's collection would be a good place to start. Thanks Bruce. Most of the examples relate to microwave applications, where often there is no alternative, but the approach works well on HF and VHF as well, and more importantly, can be achieved with existing oscillators with little modification. The IL technique works with ratios from 20:1 to 1:20 or more, and works well with the GPSDO as a reference. My experience is mostly with locking HF crystal oscillators. It works with overtone as well as fundamental oscillators. With an overtone oscillator you can couple into the mode suppression choke. With a tuned tank Pierce oscillator you can couple into the output tank. With a Colpitts, inject into the emitter, collector, or bottom of the crystal. I have made an excellent 10MHz CMOS gate oscillator with 2MHz injection into a varicap acting as one of the crystal load caps (output side). Kit VK2LL and others have used 10MHz injection to lock the 20MHz reference in common Icom HF transceivers. Arguably the father of the Injection Locking technique would be Vasil Uzunoglu, and I have some references for articles by him. The most readable article is Synchronous Oscillator outperforms the PLL (from EDN 1999) http://www.edn.com/contents/images/46326.pdf. It shows how to emitter-lock a conventional Colpitts oscillator. The secret here is to get the bias correct. The test and measurement techniques he uses are a good way to assess performance. Robert Adler (inventor of the TV remote!) also explored the IL technique. See: http://www.edn.com/contents/images/46326.pdf US Patent 4,355,404 Carrier Recovery Network for QPSK Modems employing Synchronized Oscillators, Uzunoglu 1982 US Patent 6,580,330 Injection Locked Oscillator Automatic Frequency Centering method and Apparatus, Katznelson Petrovic 2003 (has a good list of background papers to read) A study of locking oscillators... Proc IEEE R Adler 1973 http://ieeexplore.ieee.org/iel5/5/31361/31173/01451222.pdf http://en.wikipedia.org/wiki/Injection_locking http://www.amalgamate2000.com/radio-hobbies/radio/synchronous_oscillator .htm http://potol.eecs.berkeley.edu/~jr/research/PDFs/2009-01-ASPDAC-Bhansali -Roychowdhury-GenAdler.pdf http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.130.2535rep=re p1type=pdf That should keep you busy for a while! 73, Murray ZL1BPU ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] FW: Injection locking
Hi Murray and all, Yes, indeed injection locking looks very interesting, and I started reading around. Seems relatively easy for 22 MHz, but not as easy for 42 MHz (good values should be 6 or 7 MHz, right?). So far the practical circuit I've seen are few, and this would make me lean in favour of direct synthesys which indeed looks easy but forces me to abandon the old oscillator circuits. In the next days I'll try simulating a few ideas with spice and then decide. First wild idea: how about making two CMOS gate xtal oscillators with injection locking as you describe? I'd need 2 MHz (10 divided by five) for the 22 MHz, but how practical would be obtaining the 6 or 7 MHz from 10 MHz? It would need another oscillator locked Thanks Frank IZ8DWF On 2/2/10, Murray Greenman murray.green...@rakon.com wrote: -Original Message- From: Murray Greenman Sent: Wednesday, 3 February 2010 9:00 a.m. To: 'time-nuts@febo.com' Subject: Injection locking Frank, Bruce's collection would be a good place to start. Thanks Bruce. Most of the examples relate to microwave applications, where often there is no alternative, but the approach works well on HF and VHF as well, and more importantly, can be achieved with existing oscillators with little modification. The IL technique works with ratios from 20:1 to 1:20 or more, and works well with the GPSDO as a reference. My experience is mostly with locking HF crystal oscillators. It works with overtone as well as fundamental oscillators. With an overtone oscillator you can couple into the mode suppression choke. With a tuned tank Pierce oscillator you can couple into the output tank. With a Colpitts, inject into the emitter, collector, or bottom of the crystal. I have made an excellent 10MHz CMOS gate oscillator with 2MHz injection into a varicap acting as one of the crystal load caps (output side). Kit VK2LL and others have used 10MHz injection to lock the 20MHz reference in common Icom HF transceivers. Arguably the father of the Injection Locking technique would be Vasil Uzunoglu, and I have some references for articles by him. The most readable article is Synchronous Oscillator outperforms the PLL (from EDN 1999) http://www.edn.com/contents/images/46326.pdf. It shows how to emitter-lock a conventional Colpitts oscillator. The secret here is to get the bias correct. The test and measurement techniques he uses are a good way to assess performance. Robert Adler (inventor of the TV remote!) also explored the IL technique. See: http://www.edn.com/contents/images/46326.pdf US Patent 4,355,404 Carrier Recovery Network for QPSK Modems employing Synchronized Oscillators, Uzunoglu 1982 US Patent 6,580,330 Injection Locked Oscillator Automatic Frequency Centering method and Apparatus, Katznelson Petrovic 2003 (has a good list of background papers to read) A study of locking oscillators... Proc IEEE R Adler 1973 http://ieeexplore.ieee.org/iel5/5/31361/31173/01451222.pdf http://en.wikipedia.org/wiki/Injection_locking http://www.amalgamate2000.com/radio-hobbies/radio/synchronous_oscillator .htm http://potol.eecs.berkeley.edu/~jr/research/PDFs/2009-01-ASPDAC-Bhansali -Roychowdhury-GenAdler.pdf http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.130.2535rep=re p1type=pdf That should keep you busy for a while! 73, Murray ZL1BPU ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] FW: Injection locking
To generate either 6MHz or 7MHz from 10MHz one can always use something akin to a conjugate regenerative divider. For 7Mhz this requires a mixer a 7MHz bandpass filter, a 3MHz bandpass filter, a couple of power splitter/combiners, and a couple of amplifiers. For 6Mhz this requires a mixer a 6MHz bandpass filter, a 4MHz bandpass filter, a couple of power splitter/combiners, and a couple of amplifiers. Bruce francesco messineo wrote: Hi Murray and all, Yes, indeed injection locking looks very interesting, and I started reading around. Seems relatively easy for 22 MHz, but not as easy for 42 MHz (good values should be 6 or 7 MHz, right?). So far the practical circuit I've seen are few, and this would make me lean in favour of direct synthesys which indeed looks easy but forces me to abandon the old oscillator circuits. In the next days I'll try simulating a few ideas with spice and then decide. First wild idea: how about making two CMOS gate xtal oscillators with injection locking as you describe? I'd need 2 MHz (10 divided by five) for the 22 MHz, but how practical would be obtaining the 6 or 7 MHz from 10 MHz? It would need another oscillator locked Thanks Frank IZ8DWF On 2/2/10, Murray Greenmanmurray.green...@rakon.com wrote: -Original Message- From: Murray Greenman Sent: Wednesday, 3 February 2010 9:00 a.m. To: 'time-nuts@febo.com' Subject: Injection locking Frank, Bruce's collection would be a good place to start. Thanks Bruce. Most of the examples relate to microwave applications, where often there is no alternative, but the approach works well on HF and VHF as well, and more importantly, can be achieved with existing oscillators with little modification. The IL technique works with ratios from 20:1 to 1:20 or more, and works well with the GPSDO as a reference. My experience is mostly with locking HF crystal oscillators. It works with overtone as well as fundamental oscillators. With an overtone oscillator you can couple into the mode suppression choke. With a tuned tank Pierce oscillator you can couple into the output tank. With a Colpitts, inject into the emitter, collector, or bottom of the crystal. I have made an excellent 10MHz CMOS gate oscillator with 2MHz injection into a varicap acting as one of the crystal load caps (output side). Kit VK2LL and others have used 10MHz injection to lock the 20MHz reference in common Icom HF transceivers. Arguably the father of the Injection Locking technique would be Vasil Uzunoglu, and I have some references for articles by him. The most readable article is Synchronous Oscillator outperforms the PLL (from EDN 1999) http://www.edn.com/contents/images/46326.pdf. It shows how to emitter-lock a conventional Colpitts oscillator. The secret here is to get the bias correct. The test and measurement techniques he uses are a good way to assess performance. Robert Adler (inventor of the TV remote!) also explored the IL technique. See: http://www.edn.com/contents/images/46326.pdf US Patent 4,355,404 Carrier Recovery Network for QPSK Modems employing Synchronized Oscillators, Uzunoglu 1982 US Patent 6,580,330 Injection Locked Oscillator Automatic Frequency Centering method and Apparatus, Katznelson Petrovic 2003 (has a good list of background papers to read) A study of locking oscillators... Proc IEEE R Adler 1973 http://ieeexplore.ieee.org/iel5/5/31361/31173/01451222.pdf http://en.wikipedia.org/wiki/Injection_locking http://www.amalgamate2000.com/radio-hobbies/radio/synchronous_oscillator .htm http://potol.eecs.berkeley.edu/~jr/research/PDFs/2009-01-ASPDAC-Bhansali -Roychowdhury-GenAdler.pdf http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.130.2535rep=re p1type=pdf That should keep you busy for a while! 73, Murray ZL1BPU ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] FW: Injection locking
However injection locking also works when the frequencies ratios involved are rational numbers. For 22MHz and 10MHz, the corresponding ratio is 11/5 a rational number. For 42MHz and 10MHz, the frequency ratio is 21/5 a rational number Bruce Bruce Griffiths wrote: To generate either 6MHz or 7MHz from 10MHz one can always use something akin to a conjugate regenerative divider. For 7Mhz this requires a mixer a 7MHz bandpass filter, a 3MHz bandpass filter, a couple of power splitter/combiners, and a couple of amplifiers. For 6Mhz this requires a mixer a 6MHz bandpass filter, a 4MHz bandpass filter, a couple of power splitter/combiners, and a couple of amplifiers. Bruce francesco messineo wrote: Hi Murray and all, Yes, indeed injection locking looks very interesting, and I started reading around. Seems relatively easy for 22 MHz, but not as easy for 42 MHz (good values should be 6 or 7 MHz, right?). So far the practical circuit I've seen are few, and this would make me lean in favour of direct synthesys which indeed looks easy but forces me to abandon the old oscillator circuits. In the next days I'll try simulating a few ideas with spice and then decide. First wild idea: how about making two CMOS gate xtal oscillators with injection locking as you describe? I'd need 2 MHz (10 divided by five) for the 22 MHz, but how practical would be obtaining the 6 or 7 MHz from 10 MHz? It would need another oscillator locked Thanks Frank IZ8DWF On 2/2/10, Murray Greenmanmurray.green...@rakon.com wrote: -Original Message- From: Murray Greenman Sent: Wednesday, 3 February 2010 9:00 a.m. To: 'time-nuts@febo.com' Subject: Injection locking Frank, Bruce's collection would be a good place to start. Thanks Bruce. Most of the examples relate to microwave applications, where often there is no alternative, but the approach works well on HF and VHF as well, and more importantly, can be achieved with existing oscillators with little modification. The IL technique works with ratios from 20:1 to 1:20 or more, and works well with the GPSDO as a reference. My experience is mostly with locking HF crystal oscillators. It works with overtone as well as fundamental oscillators. With an overtone oscillator you can couple into the mode suppression choke. With a tuned tank Pierce oscillator you can couple into the output tank. With a Colpitts, inject into the emitter, collector, or bottom of the crystal. I have made an excellent 10MHz CMOS gate oscillator with 2MHz injection into a varicap acting as one of the crystal load caps (output side). Kit VK2LL and others have used 10MHz injection to lock the 20MHz reference in common Icom HF transceivers. Arguably the father of the Injection Locking technique would be Vasil Uzunoglu, and I have some references for articles by him. The most readable article is Synchronous Oscillator outperforms the PLL (from EDN 1999) http://www.edn.com/contents/images/46326.pdf. It shows how to emitter-lock a conventional Colpitts oscillator. The secret here is to get the bias correct. The test and measurement techniques he uses are a good way to assess performance. Robert Adler (inventor of the TV remote!) also explored the IL technique. See: http://www.edn.com/contents/images/46326.pdf US Patent 4,355,404 Carrier Recovery Network for QPSK Modems employing Synchronized Oscillators, Uzunoglu 1982 US Patent 6,580,330 Injection Locked Oscillator Automatic Frequency Centering method and Apparatus, Katznelson Petrovic 2003 (has a good list of background papers to read) A study of locking oscillators... Proc IEEE R Adler 1973 http://ieeexplore.ieee.org/iel5/5/31361/31173/01451222.pdf http://en.wikipedia.org/wiki/Injection_locking http://www.amalgamate2000.com/radio-hobbies/radio/synchronous_oscillator .htm http://potol.eecs.berkeley.edu/~jr/research/PDFs/2009-01-ASPDAC-Bhansali -Roychowdhury-GenAdler.pdf http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.130.2535rep=re p1type=pdf That should keep you busy for a while! 73, Murray ZL1BPU ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
[time-nuts] EFRATOM LPRO-101
Hello all. I received an EFRATOM LPRO-101 last week from eBay Fluke 1, and wired it up last night. The results are not good. Using a HP 5370B counter with calibration less than one year old, the frequency reads = 10,000,146,012.9 Power Applied = 24.9Vdc @ 0.30 amps after warm up Lamp Voltage = 5.9 Vdc Ext C Field = 2.437 XTL V Mon = 14.67 Bite = 4.57 Bite never goes low. I just sent an email to the outfit I purchased it from, but would like to get this working instead of waiting a month or so. Is it dead? Any thing I can do to revive it? Any suggestions? Thanks to all. Jim N0OBG -- The heart of the wise inclines to the right, but the heart of the fool to the left. Eccl 10:2 (NIV) ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] FW: Injection locking
On 2/2/10, Bruce Griffiths bruce.griffi...@xtra.co.nz wrote: However injection locking also works when the frequencies ratios involved are rational numbers. For 22MHz and 10MHz, the corresponding ratio is 11/5 a rational number. For 42MHz and 10MHz, the frequency ratio is 21/5 a rational number Then 2 MHz would work for both, obtaining 2 MHz from 10 MHz is quite easy, HP AN-301-1 plus a 74LS193 would do it. If I can make it work with the current oscillator design I'm using, I can easily adapt it also to already made transverters, very tempting. Frank ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] EFRATOM LPRO-101
Buy another one, they are so cheap :-) Out of the last lot of five that I bought from the same source, four work very well and one is marginal - sometimes it locks but it's noisy, sometime does not lock. Lamp and xtal voltage are ok, but it jumps up and down by 1E10. Very positive overall, IMHO. 73 - Marco IK1ODO At 22.08 02/02/2010, you wrote: Hello all. I received an EFRATOM LPRO-101 last week from eBay Fluke 1, and wired it up last night. The results are not good. Using a HP 5370B counter with calibration less than one year old, the frequency reads = 10,000,146,012.9 Power Applied = 24.9Vdc @ 0.30 amps after warm up Lamp Voltage = 5.9 Vdc Ext C Field = 2.437 XTL V Mon = 14.67 Bite = 4.57 Bite never goes low. I just sent an email to the outfit I purchased it from, but would like to get this working instead of waiting a month or so. Is it dead? Any thing I can do to revive it? Any suggestions? Thanks to all. Jim N0OBG ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] FW: Injection locking
francesco messineo wrote: On 2/2/10, Bruce Griffithsbruce.griffi...@xtra.co.nz wrote: However injection locking also works when the frequencies ratios involved are rational numbers. For 22MHz and 10MHz, the corresponding ratio is 11/5 a rational number. For 42MHz and 10MHz, the frequency ratio is 21/5 a rational number Then 2 MHz would work for both, obtaining 2 MHz from 10 MHz is quite easy, HP AN-301-1 plus a 74LS193 would do it. If I can make it work with the current oscillator design I'm using, I can easily adapt it also to already made transverters, very tempting. Frank Directly injecting the 10MHz signal should also work for either 22MHz or 42MHz. Bruce ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
[time-nuts] Injection Locking
Frank, As Bruce suggests, you can in theory lock any rational number ratio, including 11/5 and 21/5. However, the locking gain drops off as the ratio becomes more extreme, and thus the lock range and potential stability are degraded. Yes, you could certainly use a regenerative divider to generate 6MHz or 7MHz, but why not use another IL oscillator? Simply make a 6MHz gate oscillator (common micro crystal) and lock it to 2MHz, divided from 10MHz, then use that to lock 42MHz. If you use Johnson decade counter as divider (e.g. 74HC4017), you get a nice 100ns wide 2MHz pulse to lock to. In my experience locking higher frequencies to lower ones is easier if the reference consists of a pulse with width rather less than the period of the higher frequency, although this depends on how you inject. You'll need to buffer and shape the 6MHz to lock the 42MHz oscillator. Good idea, build a bunch of gate oscillators and experiment. I suggest you also try directly locking 42MHz to 2MHz, in order to determine the lock range. I know from experience at 2MHz to 10MHz (5/1) the lock range is immense, far more than the thermal drift of the 10MHz oscillator, and so I have a very phase stable result. I used mine to GPS lock a 2MHz OCXO, using a micro running at 10MHz, and I injection locked the 10MHz micro clock to the 2MHz, and so needed good phase stability. Not a micro or ASIC in sight! 73, Murray Zl1BPU ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
[time-nuts] R: Re: EFRATOM LPRO-101
I had exactly the same symptom on one of my units: a comparator circuit fails to detect the passing of the sweept voltage over the upper treshold of about 12.5V. The problem was R215 on the bottom face of the pcb, 100K. Sometimes it is not a matter of budget, at least in my case, and trying to fix a faulty device may have the taste of a challenge. Please check that resistor, and let me know. (a secondary check before checking R215: after warm-up, say 5 minutes, try interrupting the power supply for an instant: this forces the unit to restart sweeping the voltage, and, being warm, it could happen to lock before reaching the 12.5V treshold). Antonio I8IOV Buy another one, they are so cheap :-) Out of the last lot of five that I bought from the same source, four work very well and one is marginal - sometimes it locks but it's noisy, sometime does not lock. Lamp and xtal voltage are ok, but it jumps up and down by 1E10. Very positive overall, IMHO. 73 - Marco IK1ODO At 22.08 02/02/2010, you wrote: Hello all. I received an EFRATOM LPRO-101 last week from eBay Fluke 1, and wired it up last night. The results are not good. Using a HP 5370B counter with calibration less than one year old, the frequency reads = 10,000,146,012.9 Power Applied = 24.9Vdc @ 0.30 amps after warm up Lamp Voltage = 5.9 Vdc Ext C Field = 2.437 XTL V Mon = 14.67 Bite = 4.57 Bite never goes low. I just sent an email to the outfit I purchased it from, but would like to get this working instead of waiting a month or so. Is it dead? Any thing I can do to revive it? Any suggestions? Thanks to all. Jim N0OBG ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Aside about Triangle Waveforms
It's the AN/APN-1 in the USAAC version. There is also a USN version with slight differences. -John == A widely used WW-II aircraft radio altimeter used a triangular waveform to FM modulate a 400 MHz oscillator, employing a mechanical variable capacitor constructed similar to a permanent-magnet loudspeaker. To get the capacitor's diaphragm to reverse accurately, at the positive peak of the trianglular waveform, required a sharp, negative-going impulse to be added to the peak of the triangle, creating a sharp notch in the waveform about 30% deep. This makes me wonder about the limitations of speaker cones attempting to reproduce complex waveforms. If they had overall feedback for positional correction, the spectrum of the resulting driving waveform might contain some pretty complex components. Bruce Hunter ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] FW: Injection locking
Hi If you do go the injection locking route check a couple of things: 1) Be sure to do the math and keep the 3db bandwidth down to the ~20 Hz range. Otherwise you will be getting more phase noise than you probably should. Generally this means having some kind of control on how much power you are injecting. 2) Consider what impact (if any) the extra signals running around in your radio will have. The harmonics of what ever you inject are one issue. The intermodulation products between the injection and the oscillator in some cases can be another. None of this is to say it does not work. Only that there are a few things that may (or may not) be issues. Bob -Original Message- From: time-nuts-boun...@febo.com [mailto:time-nuts-boun...@febo.com] On Behalf Of francesco messineo Sent: Tuesday, February 02, 2010 4:14 PM To: Discussion of precise time and frequency measurement Subject: Re: [time-nuts] FW: Injection locking On 2/2/10, Bruce Griffiths bruce.griffi...@xtra.co.nz wrote: However injection locking also works when the frequencies ratios involved are rational numbers. For 22MHz and 10MHz, the corresponding ratio is 11/5 a rational number. For 42MHz and 10MHz, the frequency ratio is 21/5 a rational number Then 2 MHz would work for both, obtaining 2 MHz from 10 MHz is quite easy, HP AN-301-1 plus a 74LS193 would do it. If I can make it work with the current oscillator design I'm using, I can easily adapt it also to already made transverters, very tempting. Frank ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
[time-nuts] LORAN C update
American LORAN C will shut down at 2000z Feb 8 Dual rated chains that serve Canada will operate till 1 Oct. As an example North East 9960 will be off but Canadian 5930 will operate till Oct. Have used 5930 for years from Boston as a reference. North American LORAN will be completely quite by Oct 1. Time to see if the Austrons will sync to Europe then. Have locked 1 time already at night. Regards ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] LORAN C update
Hi At least when they were brand new, the Austron's had no problem at all locking to chains in Iceland and Europe from the central US. That was with the US chains going full bore Bob On Feb 2, 2010, at 5:37 PM, paul swed wrote: American LORAN C will shut down at 2000z Feb 8 Dual rated chains that serve Canada will operate till 1 Oct. As an example North East 9960 will be off but Canadian 5930 will operate till Oct. Have used 5930 for years from Boston as a reference. North American LORAN will be completely quite by Oct 1. Time to see if the Austrons will sync to Europe then. Have locked 1 time already at night. Regards ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Triangle Waves
Hi The NIST guys at least at lunch would come back with We only put enough in to get the paper published. The rest of he details will appear in another paper down the road. Some of the stuff has been coming out a piece at a time for 30 years now . Bob On Feb 2, 2010, at 2:32 PM, Bruce Griffiths wrote: Bob Camp wrote: Hi The DAC will quite nicely produce a trapezoid (or clipped triangle wave). It's certainly on the list. --- I've had a lot of lunch time discussions with the NIST guys about their obsession with input levels. About all I can say is that I don't see the same sensitivities they do. I suspect a lot of the issue is that we're not using the exact same circuits / components. - Some of their papers are annoyingly incomplete, in that the measurement setups used were incompletely specified. Matching the isolation amplifier to the mixer for efficient power transfer is something I do plan to look at. I suspect it's only good up to a certain point and then you get into trouble. - Lots of things to check ... Bob Bruce ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Triangle Waves
Bob Camp wrote: Hi At least from the last time I tried it: If you use a sine wave input source, it's got to be an amazingly good 10 Hz sine wave. A normal audio generator will not produce a 10 Hz output with good enough short term stability / noise to give you useful data. Audio generators may be out there that will do the job, but I certainly don't have one, and have never come across one. Since the output of the mixer is basically a triangle wave, it makes sense to use that as your test source. A triangle wave also has the nice property that it's easy on the math. You don't have any approximation issues with the integers going into the DAC. That shoves the inevitable digital crud higher in frequency. Another nice thing about a pure digital approach is that it provides a clean trigger for the start channel of the counter you are testing things with. You can even set up the DAC to put out square waves to see just how good various bits of the chain are. Tough to do that with anything other than another arbitrary function generator. I agree that the reference is going to be an issue and that a LED stack may be the way to go. No matter how you generate the test tone, power supply noise will be an issue. The output amplifier on the DAC is my biggest worry. I could go with a current out DAC and something like an OP-27. That won't give me 1nV/Hz either, but it will at least be within shouting distance of it. Sigma deltas might be a third option. I have no idea what their low frequency flicker noise looks like. So, other than the noise issue (which obviously needs to be analyzed / tested / pounded on) any other issues with the approach? At least from what I have seen in the past, level sensitivity on the inputs shows up pretty fast in the output beat note as you vary the input signals that are supposed to be saturating the mixer. If they are doing their job, a 2 db level change produces a very small change in the output. If you have something amiss in that department, you will see it pretty fast. On that I'm pretty much in agreement with Rubiola's stuff. Since I intend to mate the isolation amps up directly on the same board as the mixer, there is no real need for a 50 ohm interface between them. If the mixer looks like 18.26 ohms, the amp output can be transformed to that level rather than 50 ohms. Everything is matched (over a 1/8 trace) and you don't burn up power in a bunch of resistors. How well that idea works - time will tell. It's easy to put the resistors in if it flunks out. So many things to try Just a reality check question here... a simple triangle oscillator is very easily created by two op-amps, one for an integrator and one for Schmitt trigger operation. If you want better long-term stability open the loop and insert a 10 Hz from your favourite divider chain of a trusted 10 MHz or so. Would such a design be limiting your measurement goals considerable, and would any flaws be reasonably to overcome by better design? Cheers, Magnus ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Triangle Waves
If you use the integrator-hysteresis approach, make VERY sure the FB capacitor has excellent dC/dV, otherwise the ramp will NOT be linear. I built one ages ago, using a ceramic capacitor and it produced near a sine. -John == Bob Camp wrote: Hi At least from the last time I tried it: If you use a sine wave input source, it's got to be an amazingly good 10 Hz sine wave. A normal audio generator will not produce a 10 Hz output with good enough short term stability / noise to give you useful data. Audio generators may be out there that will do the job, but I certainly don't have one, and have never come across one. Since the output of the mixer is basically a triangle wave, it makes sense to use that as your test source. A triangle wave also has the nice property that it's easy on the math. You don't have any approximation issues with the integers going into the DAC. That shoves the inevitable digital crud higher in frequency. Another nice thing about a pure digital approach is that it provides a clean trigger for the start channel of the counter you are testing things with. You can even set up the DAC to put out square waves to see just how good various bits of the chain are. Tough to do that with anything other than another arbitrary function generator. I agree that the reference is going to be an issue and that a LED stack may be the way to go. No matter how you generate the test tone, power supply noise will be an issue. The output amplifier on the DAC is my biggest worry. I could go with a current out DAC and something like an OP-27. That won't give me 1nV/Hz either, but it will at least be within shouting distance of it. Sigma deltas might be a third option. I have no idea what their low frequency flicker noise looks like. So, other than the noise issue (which obviously needs to be analyzed / tested / pounded on) any other issues with the approach? At least from what I have seen in the past, level sensitivity on the inputs shows up pretty fast in the output beat note as you vary the input signals that are supposed to be saturating the mixer. If they are doing their job, a 2 db level change produces a very small change in the output. If you have something amiss in that department, you will see it pretty fast. On that I'm pretty much in agreement with Rubiola's stuff. Since I intend to mate the isolation amps up directly on the same board as the mixer, there is no real need for a 50 ohm interface between them. If the mixer looks like 18.26 ohms, the amp output can be transformed to that level rather than 50 ohms. Everything is matched (over a 1/8 trace) and you don't burn up power in a bunch of resistors. How well that idea works - time will tell. It's easy to put the resistors in if it flunks out. So many things to try Just a reality check question here... a simple triangle oscillator is very easily created by two op-amps, one for an integrator and one for Schmitt trigger operation. If you want better long-term stability open the loop and insert a 10 Hz from your favourite divider chain of a trusted 10 MHz or so. Would such a design be limiting your measurement goals considerable, and would any flaws be reasonably to overcome by better design? Cheers, Magnus ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Triangle Waves
Magnus Danielson wrote: Bob Camp wrote: Hi At least from the last time I tried it: If you use a sine wave input source, it's got to be an amazingly good 10 Hz sine wave. A normal audio generator will not produce a 10 Hz output with good enough short term stability / noise to give you useful data. Audio generators may be out there that will do the job, but I certainly don't have one, and have never come across one. Since the output of the mixer is basically a triangle wave, it makes sense to use that as your test source. A triangle wave also has the nice property that it's easy on the math. You don't have any approximation issues with the integers going into the DAC. That shoves the inevitable digital crud higher in frequency. Another nice thing about a pure digital approach is that it provides a clean trigger for the start channel of the counter you are testing things with. You can even set up the DAC to put out square waves to see just how good various bits of the chain are. Tough to do that with anything other than another arbitrary function generator. I agree that the reference is going to be an issue and that a LED stack may be the way to go. No matter how you generate the test tone, power supply noise will be an issue. The output amplifier on the DAC is my biggest worry. I could go with a current out DAC and something like an OP-27. That won't give me 1nV/Hz either, but it will at least be within shouting distance of it. Sigma deltas might be a third option. I have no idea what their low frequency flicker noise looks like. So, other than the noise issue (which obviously needs to be analyzed / tested / pounded on) any other issues with the approach? At least from what I have seen in the past, level sensitivity on the inputs shows up pretty fast in the output beat note as you vary the input signals that are supposed to be saturating the mixer. If they are doing their job, a 2 db level change produces a very small change in the output. If you have something amiss in that department, you will see it pretty fast. On that I'm pretty much in agreement with Rubiola's stuff. Since I intend to mate the isolation amps up directly on the same board as the mixer, there is no real need for a 50 ohm interface between them. If the mixer looks like 18.26 ohms, the amp output can be transformed to that level rather than 50 ohms. Everything is matched (over a 1/8 trace) and you don't burn up power in a bunch of resistors. How well that idea works - time will tell. It's easy to put the resistors in if it flunks out. So many things to try Just a reality check question here... a simple triangle oscillator is very easily created by two op-amps, one for an integrator and one for Schmitt trigger operation. If you want better long-term stability open the loop and insert a 10 Hz from your favourite divider chain of a trusted 10 MHz or so. Would such a design be limiting your measurement goals considerable, and would any flaws be reasonably to overcome by better design? Cheers, Magnus For beat frequencies in the 1-100Hz range one only need verify the ZCD jitter and delay variations etc., to within a few nanosec. In the short term such jitter tantalisingly close to what a well designed audio oscillator is capable of. Unfortunately the trigger jitter in most counters is very large for frequencies in this range so verifying the low jitter of an audio oscillator requires using a ZCD or equivalent. To avoid this somewhat circular verification one would prefer a source for which the zero crossing slew rate and jitter are very low by design. Low pass filtering of the divider output may suffice for initial verification but one would need to use a fully synchronous divider chain or the equivalent thereof as far as jitter and wander etc is concerned. Bruce ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Triangle Waves
Bruce Griffiths wrote: Bob Camp wrote: Hi At least from the last time I tried it: If you use a sine wave input source, it's got to be an amazingly good 10 Hz sine wave. A normal audio generator will not produce a 10 Hz output with good enough short term stability / noise to give you useful data. Audio generators may be out there that will do the job, but I certainly don't have one, and have never come across one. Since the output of the mixer is basically a triangle wave, it makes sense to use that as your test source. A triangle wave also has the nice property that it's easy on the math. You don't have any approximation issues with the integers going into the DAC. That shoves the inevitable digital crud higher in frequency. When both the RF and LO ports are saturated, the mixer output waveform depends on how the IF port is terminated. The output is indeed approximately triangular with your IF port termination method when both the RF and LO ports are saturated. My experience says that it also depends on the relative phase... so it shifts between approx falling saw to approx tri to approx rising saw to approx falling saw over a 360 degree beat period. Kind of comforting to see this shift occurring slowly on the scope. With the IF port terminated in a capacitor when both RF and LO ports are saturated the output waveform is quasi trapezoidal. When only the LO port is saturated the IF output is sinusoidal. Another nice thing about a pure digital approach is that it provides a clean trigger for the start channel of the counter you are testing things with. You can even set up the DAC to put out square waves to see just how good various bits of the chain are. Tough to do that with anything other than another arbitrary function generator. I agree that the reference is going to be an issue and that a LED stack may be the way to go. No matter how you generate the test tone, power supply noise will be an issue. The output amplifier on the DAC is my biggest worry. I could go with a current out DAC and something like an OP-27. That won't give me 1nV/Hz either, but it will at least be within shouting distance of it. Sigma deltas might be a third option. I have no idea what their low frequency flicker noise looks like. Producing a high amplitude (eg 20V pp) output and attenuating it down to say 2V pp or so typical of a mixer will significantly reduce the noise due to the output amplifier. So, other than the noise issue (which obviously needs to be analyzed / tested / pounded on) any other issues with the approach? At least from what I have seen in the past, level sensitivity on the inputs shows up pretty fast in the output beat note as you vary the input signals that are supposed to be saturating the mixer. If they are doing their job, a 2 db level change produces a very small change in the output. If you have something amiss in that department, you will see it pretty fast. On that I'm pretty much in agreement with Rubiola's stuff. Yes but NIST used a saturated mixer and still found that the mixer phase shift depended on how hard you drive the diodes. This should not come as a big surprise, as diodes change their capacitance with applied voltage. Long term variations in isolation amplifier output due to temperature variations may be significant. I haven't heard of AGC being applied to this particular system, but their use on stabilizing mixer phase detector gain should be known. Allowing loop gain to change with input signal strength is not always a good idea. Since I intend to mate the isolation amps up directly on the same board as the mixer, there is no real need for a 50 ohm interface between them. If the mixer looks like 18.26 ohms, the amp output can be transformed to that level rather than 50 ohms. Everything is matched (over a 1/8 trace) and you don't burn up power in a bunch of resistors. How well that idea works - time will tell. It's easy to put the resistors in if it flunks out. I've considered that type of arrangement... it could provide some opportunities not allowed by cabled interconnect. For reasonable frequencies much simpler transmission models can be used. Cheers, Magnus ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Triangle Waves
Bob Camp wrote: Hi The DAC will quite nicely produce a trapezoid (or clipped triangle wave). It's certainly on the list. --- I've had a lot of lunch time discussions with the NIST guys about their obsession with input levels. About all I can say is that I don't see the same sensitivities they do. I suspect a lot of the issue is that we're not using the exact same circuits / components. - Matching the isolation amplifier to the mixer for efficient power transfer is something I do plan to look at. I suspect it's only good up to a certain point and then you get into trouble. If you stay at say 10 MHz or so, the isolational amplifier is close to the mixer and a more lumped model can be used. 1/12 of the rise time is used as rule of thumb, but I am sure you know this well. Best performance rather than best power transfer may be a better goal, but what manifest best performance for such a system is not easier said than analysed. Cheers, Magnus ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Triangle Waves
Magnus Danielson wrote: When both the RF and LO ports are saturated, the mixer output waveform depends on how the IF port is terminated. The output is indeed approximately triangular with your IF port termination method when both the RF and LO ports are saturated. My experience says that it also depends on the relative phase... so it shifts between approx falling saw to approx tri to approx rising saw to approx falling saw over a 360 degree beat period. Kind of comforting to see this shift occurring slowly on the scope. Your description is for the beat frequency output or for the input signal waveforms? With a finite frequency beat signal one gets a periodic modulation of the phase difference between the 2 inputs. Bruce Cheers, Magnus ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Triangle Waves
time-nuts-boun...@febo.com wrote on 02/02/2010 07:20:24 PM: From: Bruce Griffiths bruce.griffi...@xtra.co.nz To: Discussion of precise time and frequency measurement time-nuts@febo.com Date: 02/02/2010 07:27 PM Subject: Re: [time-nuts] Triangle Waves Sent by: time-nuts-boun...@febo.com Magnus Danielson wrote: [snip] Just a reality check question here... a simple triangle oscillator is very easily created by two op-amps, one for an integrator and one for Schmitt trigger operation. If you want better long-term stability open the loop and insert a 10 Hz from your favourite divider chain of a trusted 10 MHz or so. Would such a design be limiting your measurement goals considerable, and would any flaws be reasonably to overcome by better design? Cheers, Magnus For beat frequencies in the 1-100Hz range one only need verify the ZCD jitter and delay variations etc., to within a few nanosec. In the short term such jitter tantalisingly close to what a well designed audio oscillator is capable of. Unfortunately the trigger jitter in most counters is very large for frequencies in this range so verifying the low jitter of an audio oscillator requires using a ZCD or equivalent. Would integration of a 50% duty cycle square wave generate an adequate triangle wave? Modern opamps make pretty good low-noise integrators, although one would need to use a good integration capacitor to ensure linear ramps. The square wave would come from a simple binary divider chain, which will clean many things up and ensure a stable duty cycle, whatever the nature of the original signal source. Joe Gwinn ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Triangle Waves
Joseph M Gwinn wrote: time-nuts-boun...@febo.com wrote on 02/02/2010 07:20:24 PM: From: Bruce Griffithsbruce.griffi...@xtra.co.nz To: Discussion of precise time and frequency measurement time-nuts@febo.com Date: 02/02/2010 07:27 PM Subject: Re: [time-nuts] Triangle Waves Sent by: time-nuts-boun...@febo.com Magnus Danielson wrote: [snip] Just a reality check question here... a simple triangle oscillator is very easily created by two op-amps, one for an integrator and one for Schmitt trigger operation. If you want better long-term stability open the loop and insert a 10 Hz from your favourite divider chain of a trusted 10 MHz or so. Would such a design be limiting your measurement goals considerable, and would any flaws be reasonably to overcome by better design? Cheers, Magnus For beat frequencies in the 1-100Hz range one only need verify the ZCD jitter and delay variations etc., to within a few nanosec. In the short term such jitter tantalisingly close to what a well designed audio oscillator is capable of. Unfortunately the trigger jitter in most counters is very large for frequencies in this range so verifying the low jitter of an audio oscillator requires using a ZCD or equivalent. Would integration of a 50% duty cycle square wave generate an adequate triangle wave? Modern opamps make pretty good low-noise integrators, although one would need to use a good integration capacitor to ensure linear ramps. The square wave would come from a simple binary divider chain, which will clean many things up and ensure a stable duty cycle, whatever the nature of the original signal source. Joe Gwinn The integration function requires a low frequency cutoff (either a servoloop or a resistor shunting the integration capacitor) to avoid integrator saturation. This inevitably distorts the triangle wave, however it should be possible to reduce the triangular wave distortion by predistorting the integrator input current. Bruce ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Triangle Waves
No. Unless the up and down ramps were PERFECTLY matched and the opamp had ZERO offset, the imperfections would integrate up or down, eventually saturating the opamp. You could make the integrator less perfect by adding a R in parallel w/ the FB cap, but it'll not be an integrator any more. -John === Would integration of a 50% duty cycle square wave generate an adequate triangle wave? Modern opamps make pretty good low-noise integrators, although one would need to use a good integration capacitor to ensure linear ramps. The square wave would come from a simple binary divider chain, which will clean many things up and ensure a stable duty cycle, whatever the nature of the original signal source. Joe Gwinn ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Triangle Waves
Hi At least the last time I tried it, the filter a square wave / integrate based on a square wave approach both appeared to give performance that was inadequate. Simply put, the triangle wave should give *better* performance than a similar wave generated off of a pair of good oscillators. That was not the case. Could there have been errors made - sure. Exactly what did I do - long time ago, details are in a log book that probably doesn't even exist any more. Ideally the signal source would be much better than the limiter I'm trying to test. If I want to verify a 10 ns limiter, a triangle wave good to a ns or so would be a nice thing to have. It would also be nice to easily verify that device. The objective is a quick test of a limiter rather than the world's best low frequency R-C oscillator. That of course assumes I can womp up the DAC gizmo easily. Bob On Feb 2, 2010, at 7:58 PM, Joseph M Gwinn wrote: time-nuts-boun...@febo.com wrote on 02/02/2010 07:20:24 PM: From: Bruce Griffiths bruce.griffi...@xtra.co.nz To: Discussion of precise time and frequency measurement time-nuts@febo.com Date: 02/02/2010 07:27 PM Subject: Re: [time-nuts] Triangle Waves Sent by: time-nuts-boun...@febo.com Magnus Danielson wrote: [snip] Just a reality check question here... a simple triangle oscillator is very easily created by two op-amps, one for an integrator and one for Schmitt trigger operation. If you want better long-term stability open the loop and insert a 10 Hz from your favourite divider chain of a trusted 10 MHz or so. Would such a design be limiting your measurement goals considerable, and would any flaws be reasonably to overcome by better design? Cheers, Magnus For beat frequencies in the 1-100Hz range one only need verify the ZCD jitter and delay variations etc., to within a few nanosec. In the short term such jitter tantalisingly close to what a well designed audio oscillator is capable of. Unfortunately the trigger jitter in most counters is very large for frequencies in this range so verifying the low jitter of an audio oscillator requires using a ZCD or equivalent. Would integration of a 50% duty cycle square wave generate an adequate triangle wave? Modern opamps make pretty good low-noise integrators, although one would need to use a good integration capacitor to ensure linear ramps. The square wave would come from a simple binary divider chain, which will clean many things up and ensure a stable duty cycle, whatever the nature of the original signal source. Joe Gwinn ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Triangle Waves
time-nuts-boun...@febo.com wrote on 02/02/2010 08:19:26 PM: From: Bruce Griffiths bruce.griffi...@xtra.co.nz To: Discussion of precise time and frequency measurement time-nuts@febo.com Date: 02/02/2010 08:20 PM Subject: Re: [time-nuts] Triangle Waves Sent by: time-nuts-boun...@febo.com Joseph M Gwinn wrote: time-nuts-boun...@febo.com wrote on 02/02/2010 07:20:24 PM: From: Bruce Griffithsbruce.griffi...@xtra.co.nz To: Discussion of precise time and frequency measurement time-nuts@febo.com Date: 02/02/2010 07:27 PM Subject: Re: [time-nuts] Triangle Waves Sent by: time-nuts-boun...@febo.com Magnus Danielson wrote: [snip] Just a reality check question here... a simple triangle oscillator is very easily created by two op-amps, one for an integrator and one for Schmitt trigger operation. If you want better long-term stability open the loop and insert a 10 Hz from your favourite divider chain of a trusted 10 MHz or so. Would such a design be limiting your measurement goals considerable, and would any flaws be reasonably to overcome by better design? Cheers, Magnus For beat frequencies in the 1-100Hz range one only need verify the ZCD jitter and delay variations etc., to within a few nanosec. In the short term such jitter tantalisingly close to what a well designed audio oscillator is capable of. Unfortunately the trigger jitter in most counters is very large for frequencies in this range so verifying the low jitter of an audio oscillator requires using a ZCD or equivalent. Would integration of a 50% duty cycle square wave generate an adequate triangle wave? Modern opamps make pretty good low-noise integrators, although one would need to use a good integration capacitor to ensure linear ramps. The square wave would come from a simple binary divider chain, which will clean many things up and ensure a stable duty cycle, whateverthe nature of the original signal source. Joe Gwinn The integration function requires a low frequency cutoff (either a servoloop or a resistor shunting the integration capacitor) to avoid integrator saturation. This inevitably distorts the triangle wave, however it should be possible to reduce the triangular wave distortion by predistorting the integrator input current. Yes, there would need to be some kind of drift compensation (I favor a opamp servoloop), but given that we are trying to measure ZCD jitter (versus long-term wander), isn't this good enough? The distortion will be small and stable, and so will not cause jitter. Joe Gwinn ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] OT: Practical PLL low noise?
Frank, You might want to take a look in here also. Hardly gets any simpler than this ;-) http://w3ref.cfn.ist.utl.pt/cupido/reflock.html Luis Cupido ct1dmk. On 02/02/10 16:50, francesco messineo wrote: Hello all, sorry for the OT, but I know there're many real electronic artists here. As an amateur radio operator I often use transverters, some home made. They usually can be made sigthly better (RF and noise-wise) than japanese transceivers. However often the LO xtal oscillator drifts too much for comfortable digital and weak signal work. Now the big question: is there any PLL design that can lock 22 MHz and 42 MHz xtal oscillator to a 10 MHz reference (typically from a GPSDO) without adding significant noise to the oscillators? The LOs usually go to a single or doube balanced diode mixer like the famous minicircuit ones, and at that point the RF signal has been already amplified by 10 or 20 dB stage(s). Other options would be ovenizing the LOs or making a DDS sinth. Now, what would be more practical approach from the home construction point? Thanks Frank IZ8DWF ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Triangle Waves
Joseph M Gwinn wrote: time-nuts-boun...@febo.com wrote on 02/02/2010 08:19:26 PM: From: Bruce Griffithsbruce.griffi...@xtra.co.nz To: Discussion of precise time and frequency measurement time-nuts@febo.com Date: 02/02/2010 08:20 PM Subject: Re: [time-nuts] Triangle Waves Sent by: time-nuts-boun...@febo.com Joseph M Gwinn wrote: time-nuts-boun...@febo.com wrote on 02/02/2010 07:20:24 PM: From: Bruce Griffithsbruce.griffi...@xtra.co.nz To: Discussion of precise time and frequency measurement time-nuts@febo.com Date: 02/02/2010 07:27 PM Subject: Re: [time-nuts] Triangle Waves Sent by: time-nuts-boun...@febo.com Magnus Danielson wrote: [snip] Just a reality check question here... a simple triangle oscillator is very easily created by two op-amps, one for an integrator and one for Schmitt trigger operation. If you want better long-term stability open the loop and insert a 10 Hz from your favourite divider chain of a trusted 10 MHz or so. Would such a design be limiting your measurement goals considerable, and would any flaws be reasonably to overcome by better design? Cheers, Magnus For beat frequencies in the 1-100Hz range one only need verify the ZCD jitter and delay variations etc., to within a few nanosec. In the short term such jitter tantalisingly close to what a well designed audio oscillator is capable of. Unfortunately the trigger jitter in most counters is very large for frequencies in this range so verifying the low jitter of an audio oscillator requires using a ZCD or equivalent. Would integration of a 50% duty cycle square wave generate an adequate triangle wave? Modern opamps make pretty good low-noise integrators, although one would need to use a good integration capacitor to ensure linear ramps. The square wave would come from a simple binary divider chain, which will clean many things up and ensure a stable duty cycle, whateverthe nature of the original signal source. Joe Gwinn The integration function requires a low frequency cutoff (either a servoloop or a resistor shunting the integration capacitor) to avoid integrator saturation. This inevitably distorts the triangle wave, however it should be possible to reduce the triangular wave distortion by predistorting the integrator input current. Yes, there would need to be some kind of drift compensation (I favor a opamp servoloop), but given that we are trying to measure ZCD jitter (versus long-term wander), isn't this good enough? The distortion will be small and stable, and so will not cause jitter. Joe Gwinn Yes one shouldn't lose sight of the goal which isnt a perfect triangular wave, but merely a low jitter one. The major problem is the Johnson noise of the resistors used in the integrator. If for example one uses a simple RC filter using 25k plus 10uF and drives it with a 10Hz square wave the output noise at dc is about 20nv/rtHz. The output slew rate with say a 5V amplitude square wave is about 1V pp and the zero crossing jitter due to Johnson noise is on the order of 3ns. Bruce ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Triangle Waves
Bruce Griffiths wrote: Joseph M Gwinn wrote: time-nuts-boun...@febo.com wrote on 02/02/2010 08:19:26 PM: From: Bruce Griffithsbruce.griffi...@xtra.co.nz To: Discussion of precise time and frequency measurement time-nuts@febo.com Date: 02/02/2010 08:20 PM Subject: Re: [time-nuts] Triangle Waves Sent by: time-nuts-boun...@febo.com Joseph M Gwinn wrote: time-nuts-boun...@febo.com wrote on 02/02/2010 07:20:24 PM: From: Bruce Griffithsbruce.griffi...@xtra.co.nz To: Discussion of precise time and frequency measurement time-nuts@febo.com Date: 02/02/2010 07:27 PM Subject: Re: [time-nuts] Triangle Waves Sent by: time-nuts-boun...@febo.com Magnus Danielson wrote: [snip] Just a reality check question here... a simple triangle oscillator is very easily created by two op-amps, one for an integrator and one for Schmitt trigger operation. If you want better long-term stability open the loop and insert a 10 Hz from your favourite divider chain of a trusted 10 MHz or so. Would such a design be limiting your measurement goals considerable, and would any flaws be reasonably to overcome by better design? Cheers, Magnus For beat frequencies in the 1-100Hz range one only need verify the ZCD jitter and delay variations etc., to within a few nanosec. In the short term such jitter tantalisingly close to what a well designed audio oscillator is capable of. Unfortunately the trigger jitter in most counters is very large for frequencies in this range so verifying the low jitter of an audio oscillator requires using a ZCD or equivalent. Would integration of a 50% duty cycle square wave generate an adequate triangle wave? Modern opamps make pretty good low-noise integrators, although one would need to use a good integration capacitor to ensure linear ramps. The square wave would come from a simple binary divider chain, which will clean many things up and ensure a stable duty cycle, whateverthe nature of the original signal source. Joe Gwinn The integration function requires a low frequency cutoff (either a servoloop or a resistor shunting the integration capacitor) to avoid integrator saturation. This inevitably distorts the triangle wave, however it should be possible to reduce the triangular wave distortion by predistorting the integrator input current. Yes, there would need to be some kind of drift compensation (I favor a opamp servoloop), but given that we are trying to measure ZCD jitter (versus long-term wander), isn't this good enough? The distortion will be small and stable, and so will not cause jitter. Joe Gwinn Yes one shouldn't lose sight of the goal which isnt a perfect triangular wave, but merely a low jitter one. The major problem is the Johnson noise of the resistors used in the integrator. If for example one uses a simple RC filter using 25k plus 10uF and drives it with a 10Hz square wave the output noise at dc is about 20nv/rtHz. The output slew rate with say a 5V amplitude square wave is about 1V pp and the zero crossing jitter due to Johnson noise is on the order of 3ns. Bruce Make that 25K plus 2uF producing ~2V pp quasi triangular wave with a slew rate of around 40V/s and a noise bandwidth of about 5Hz producing a zero crossing jitter of around 1.1ns due to Johnson noise from the 25K resistor. To ensure this isnt degraded by the logic supply noise an extremely low noise logic supply (at least for the output stage) is necessary. Bruce ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Triangle Waves
Joseph M Gwinn wrote: time-nuts-boun...@febo.com wrote on 02/02/2010 09:13:26 PM: From: Bruce Griffithsbruce.griffi...@xtra.co.nz To: Discussion of precise time and frequency measurement time-nuts@febo.com Date: 02/02/2010 09:16 PM Subject: Re: [time-nuts] Triangle Waves Sent by: time-nuts-boun...@febo.com Joseph M Gwinn wrote: time-nuts-boun...@febo.com wrote on 02/02/2010 08:19:26 PM: From: Bruce Griffithsbruce.griffi...@xtra.co.nz To: Discussion of precise time and frequency measurement time-nuts@febo.com Date: 02/02/2010 08:20 PM Subject: Re: [time-nuts] Triangle Waves Sent by: time-nuts-boun...@febo.com Joseph M Gwinn wrote: time-nuts-boun...@febo.com wrote on 02/02/2010 07:20:24 PM: From: Bruce Griffithsbruce.griffi...@xtra.co.nz To: Discussion of precise time and frequency measurement time-nuts@febo.com Date: 02/02/2010 07:27 PM Subject: Re: [time-nuts] Triangle Waves Sent by: time-nuts-boun...@febo.com Magnus Danielson wrote: [snip] Just a reality check question here... a simple triangle oscillator is very easily created by two op-amps, one for an integrator and one for Schmitt trigger operation. If you want better long-term stability open the loop and insert a 10 Hz from your favourite divider chain of a trusted 10 MHz or so. Would such a design be limiting your measurement goals considerable, and would any flaws be reasonably to overcome by better design? Cheers, Magnus For beat frequencies in the 1-100Hz range one only need verify the ZCD jitter and delay variations etc., to within a few nanosec. In the short term such jitter tantalisingly close to what a well designed audio oscillator is capable of. Unfortunately the trigger jitter in most counters is very large for frequencies in this range so verifying the low jitter of an audio oscillator requires using a ZCD or equivalent. Would integration of a 50% duty cycle square wave generate an adequate triangle wave? Modern opamps make pretty good low-noise integrators, although one would need to use a good integration capacitorto ensure linear ramps. The square wave would come from a simple binary divider chain, which will clean many things up and ensure a stable duty cycle, whateverthe nature of the original signal source. Joe Gwinn The integration function requires a low frequency cutoff (either a servoloop or a resistor shunting the integration capacitor) to avoid integrator saturation. This inevitably distorts the triangle wave, however it should be possible to reduce the triangular wave distortion by predistorting the integrator input current. Yes, there would need to be some kind of drift compensation (I favor a opamp servoloop), but given that we are trying to measure ZCD jitter (versus long-term wander), isn't this good enough? The distortion will be small and stable, and so will not cause jitter. Joe Gwinn Yes one shouldn't lose sight of the goal which isnt a perfect triangular wave, but merely a low jitter one. The major problem is the Johnson noise of the resistors used in the integrator. If for example one uses a simple RC filter using 25k plus 10uF and drives it with a 10Hz square wave the output noise at dc is about 20nv/rtHz. The output slew rate with say a 5V amplitude square wave is about 1V pp and the zero crossing jitter due to Johnson noise is on the order of 3ns. I've lost track of our jitter objective, and why we need to achieve it. Also, if the intent is to measure the inherent jitter of a ZCD circuit, we may be better off using a really clean sinewave, as it will be easier to generate a clean enough sinewave than trianglewave. The fact that we will use a triangle or trapezoid in practice will change the numbers somewhat, but the ranking of proposed circuits by their sinewave jitter should carry over correctly, so long as the same fundamental frequency is used. Joe Gwinn Figure 8 (attached) from Collin's paper indicates that the jitter of a 100Hz wien bridge oscillator is of the order of a few hundred nanosec or so. This was taken using a 3 stage limiter and a 1 sec counter gate time. It may be feasible to do better. It would also appear to be feasible to produce a 10Hz sinewave with ns jitter by low pass filtering a square wave using a combination of active and passive RC filtering. Bruce attachment: Wien_Osc_Stability.png___ time-nuts mailing list --
Re: [time-nuts] Triangle Waves
On Feb 2, 2010, at 9:29 PM, Joseph M Gwinn wrote: time-nuts-boun...@febo.com wrote on 02/02/2010 09:13:26 PM: From: Bruce Griffiths bruce.griffi...@xtra.co.nz To: Discussion of precise time and frequency measurement time-nuts@febo.com Date: 02/02/2010 09:16 PM Subject: Re: [time-nuts] Triangle Waves Sent by: time-nuts-boun...@febo.com Joseph M Gwinn wrote: time-nuts-boun...@febo.com wrote on 02/02/2010 08:19:26 PM: From: Bruce Griffithsbruce.griffi...@xtra.co.nz To: Discussion of precise time and frequency measurement time-nuts@febo.com Date: 02/02/2010 08:20 PM Subject: Re: [time-nuts] Triangle Waves Sent by: time-nuts-boun...@febo.com Joseph M Gwinn wrote: time-nuts-boun...@febo.com wrote on 02/02/2010 07:20:24 PM: From: Bruce Griffithsbruce.griffi...@xtra.co.nz To: Discussion of precise time and frequency measurement time-nuts@febo.com Date: 02/02/2010 07:27 PM Subject: Re: [time-nuts] Triangle Waves Sent by: time-nuts-boun...@febo.com Magnus Danielson wrote: [snip] Just a reality check question here... a simple triangle oscillator is very easily created by two op-amps, one for an integrator and one for Schmitt trigger operation. If you want better long-term stability open the loop and insert a 10 Hz from your favourite divider chain of a trusted 10 MHz or so. Would such a design be limiting your measurement goals considerable, and would any flaws be reasonably to overcome by better design? Cheers, Magnus For beat frequencies in the 1-100Hz range one only need verify the ZCD jitter and delay variations etc., to within a few nanosec. In the short term such jitter tantalisingly close to what a well designed audio oscillator is capable of. Unfortunately the trigger jitter in most counters is very large for frequencies in this range so verifying the low jitter of an audio oscillator requires using a ZCD or equivalent. Would integration of a 50% duty cycle square wave generate an adequate triangle wave? Modern opamps make pretty good low-noise integrators, although one would need to use a good integration capacitorto ensure linear ramps. The square wave would come from a simple binary divider chain, which will clean many things up and ensure a stable duty cycle, whateverthe nature of the original signal source. Joe Gwinn The integration function requires a low frequency cutoff (either a servoloop or a resistor shunting the integration capacitor) to avoid integrator saturation. This inevitably distorts the triangle wave, however it should be possible to reduce the triangular wave distortion by predistorting the integrator input current. Yes, there would need to be some kind of drift compensation (I favor a opamp servoloop), but given that we are trying to measure ZCD jitter (versus long-term wander), isn't this good enough? The distortion will be small and stable, and so will not cause jitter. Joe Gwinn Yes one shouldn't lose sight of the goal which isnt a perfect triangular wave, but merely a low jitter one. The major problem is the Johnson noise of the resistors used in the integrator. If for example one uses a simple RC filter using 25k plus 10uF and drives it with a 10Hz square wave the output noise at dc is about 20nv/rtHz. The output slew rate with say a 5V amplitude square wave is about 1V pp and the zero crossing jitter due to Johnson noise is on the order of 3ns. I've lost track of our jitter objective, and why we need to achieve it. Also, if the intent is to measure the inherent jitter of a ZCD circuit, we may be better off using a really clean sinewave, as it will be easier to generate a clean enough sinewave than trianglewave. The fact that we will use a triangle or trapezoid in practice will change the numbers somewhat, but the ranking of proposed circuits by their sinewave jitter should carry over correctly, so long as the same fundamental frequency is used. Joe Gwinn The objective is to check out a limiter that will have performance at better than 10 ns level when driven with a beat note in the 5 to 10 Hz range. A signal 5 to 10X better than the limiter target performance would be adequate. That puts the desired signal in the vicinity of 1 ppb avar at 1 second. Ideally the signal source would also provide a stable square wave trigger to drive the start channel of a counter or the sweep of a scope. I 'm sure some level of troubleshooting on the limiter will be needed eventually. Generating a good trigger off of a slow waveform isn't all that easy. The trigger also may be helpful in verifying the performance of the signal source it's self. Bob ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to
Re: [time-nuts] Triangle Waves
On Feb 2, 2010, at 9:57 PM, Bruce Griffiths wrote: Joseph M Gwinn wrote: Figure 8 (attached) from Collin's paper indicates that the jitter of a 100Hz wien bridge oscillator is of the order of a few hundred nanosec or so. This was taken using a 3 stage limiter and a 1 sec counter gate time. It may be feasible to do better. It would also appear to be feasible to produce a 10Hz sinewave with ns jitter by low pass filtering a square wave using a combination of active and passive RC filtering. That figure pretty clearly shows the issue of getting a good trigger on a low frequency signal. It's a pretty big guess to figure out what the 100 Hz oscillator is doing. At least to my eye it's pretty far from the 1 ppb / second level. As long as we're on the Collins paper - has anybody dug up the final paper referenced? I believe it's titled something like Hard Limiter Experimental Results? Bob Bruce Wien_Osc_Stability.png___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
[time-nuts] USB Optical Spectrometer
Hi all, This is for those asking about the Ocean Optics Optical Spectrometers. The guy who sells them is: Roland Guilmet rolandguil...@yahoo.com I've sent him the emails of those who expressed an interest to me off-list. Please deal directly with him. Best, -John = ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] Triangle Waves
Bob Camp wrote: On Feb 2, 2010, at 9:57 PM, Bruce Griffiths wrote: Joseph M Gwinn wrote: Figure 8 (attached) from Collin's paper indicates that the jitter of a 100Hz wien bridge oscillator is of the order of a few hundred nanosec or so. This was taken using a 3 stage limiter and a 1 sec counter gate time. It may be feasible to do better. It would also appear to be feasible to produce a 10Hz sinewave with ns jitter by low pass filtering a square wave using a combination of active and passive RC filtering. That figure pretty clearly shows the issue of getting a good trigger on a low frequency signal. It's a pretty big guess to figure out what the 100 Hz oscillator is doing. At least to my eye it's pretty far from the 1 ppb / second level. As long as we're on the Collins paper - has anybody dug up the final paper referenced? I believe it's titled something like Hard Limiter Experimental Results? Bob Bruce The final paper doesn't appear to have been published as far as I can tell. Oliver Collins appeared to move on to other things (ie moved from John Hopkins to Notre Dame) at that time. He appears to still be at the University of Notre Dame. You could always try contacting him. http://www.nd.edu/~wand/ http://www.nd.edu/%7Ewand/ Bruce ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
Re: [time-nuts] EFRATOM LPRO-101
Mine worked great for about two weeks, but now BITE goes high again after about 45 minutes. It's quite stable when it's locked. Haven't had a chance to look closer yet. Dave - Original Message - From: Jim j...@commo.com To: time-nuts@febo.com Sent: Tuesday, February 2, 2010 2:08:06 PM GMT -07:00 US/Canada Mountain Subject: [time-nuts] EFRATOM LPRO-101 Hello all. I received an EFRATOM LPRO-101 last week from eBay Fluke 1, and wired it up last night. The results are not good. Using a HP 5370B counter with calibration less than one year old, the frequency reads = 10,000,146,012.9 Power Applied = 24.9Vdc @ 0.30 amps after warm up Lamp Voltage = 5.9 Vdc Ext C Field = 2.437 XTL V Mon = 14.67 Bite = 4.57 Bite never goes low. I just sent an email to the outfit I purchased it from, but would like to get this working instead of waiting a month or so. Is it dead? Any thing I can do to revive it? Any suggestions? Thanks to all. Jim N0OBG -- The heart of the wise inclines to the right, but the heart of the fool to the left. Eccl 10:2 (NIV) ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there. ___ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
