Re: [time-nuts] GPSDO with all-digital phase/time measurement?
hau...@keteu.org said: That being said, I find myself wondering as follows: Suppose that we count OCXO cycles (at, say, 10 MHz) using one of the MCU's timer/counter peripherals, and periodically sample the counter value with an interrupt triggered on the rising edge of the GPS 1pps. Assume that this interrupt is the highest priority in the system, so that our measurement is fully deterministic, having only the +/- one cycle ambiguity inherent in the counting. Also assume that we keep the counter running continuously. In general, things like highest priority interrupt don't guarantee what you want. The fine print depends upon your MCU. Some instructions take more than 1 cycle. Consider what happens if your interrupt arrives just after another interrupt got started. For this problem, I'd check the fine print on the counter/timers. They often have a mode where it will copy the counter to a holding register (and generate an interrupt) on the rising edge of an input pin. If you also get an interrupt on the counter overflowing, you can work out the total number of cycles between rising edges. -- The other problem you need to be aware of if you want to build a GPSDO is hanging bridges. tvb: Motorola GPS M12+ Sawtooth http://www.leapsecond.com/pages/m12/sawtooth.htm Tom Clark and Rick Hambly: Timing for VLBI http://gpstime.com/files/tow-time2009.pdf -- These are my opinions. I hate spam. ___ 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] 5370 processor boards available
I think a better solution would be to find a very large super cap and power the BBB from that while giving it a power fail interrupt to quickly sync the file system. The advantage of something like the BBB is that it runs Linux so you have a nice environment in which to run your code. The disadvantage of Linux is that it's complicated and you have things like file systems that can get trashed. Now, in addition of being the sysadmin for your PC(s), you have to be an admin for your lab gear. That may be more interesting that you were expecting. A friend reports that his scope caught a virus... The Linux ext4 file system is pretty robust. There are lots of PCs out there that mostly survive power failures. If I was worried about the file system getting trashed on power off, I'd work on the software long before I added a super-cap. I think my first try would be to run the file system read-only until I figured out that I needed to write a file. Then I would know something about how much data I wanted to write and the usage patterns. You can help a lot with flush() in the right places in your code. That may cost performance if you are writing a lot of data. Another approach is to make sure you can put the disk back together easily and quickly. Then it's not such a big deal if/when the disk gets trashed. That's probably a good idea anyway. Power fail isn't the only thing that can trash a disk. It shouldn't take much more than a simple script to format the disk and copy over all the bits from a backup place on a PC. Maybe it's install the standard distro package and then add your bits. (That's assuming you can take the disk out of the BBB and plug it into a PC.) -- These are my opinions. I hate spam. ___ 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] 5370 processor boards available
On 27 February 2014 08:26, Hal Murray hmur...@megapathdsl.net wrote: The advantage of something like the BBB is that it runs Linux so you have a nice environment in which to run your code. The disadvantage of Linux is that it's complicated and you have things like file systems that can get trashed. Now, in addition of being the sysadmin for your PC(s), you have to be an admin for your lab gear. That may be more interesting that you were expecting. A friend reports that his scope caught a virus... Consumer devices like routers and modems run Linux. Does anyone know how they get around this issue? I wonder if the OS is stored on a read-only device, and that is loaded into a read/write device when the device is powered on. It is great to hear about this project - even though I no longer have a 5370B. I can see CPU upgrades could bring new life to a lot of old instruments, but it clearly takes a lot of dedication to sort out the code. Something very complex like a vector network analyzer would probably be particularly challenging, Dave ___ 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] 5370 processor boards available
In message canx10ha2qmiwc2js8xfn3jat1vxntnnbzxa8-fhq7cybf52...@mail.gmail.com , Dr. David Kirkby writes: On 27 February 2014 08:26, Hal Murray hmur...@megapathdsl.net wrote: Consumer devices like routers and modems run Linux. Does anyone know how they get around this issue? They generally don't mount their filesystems read/write, but only read/only. I've done similar things with FreeBSD in many systems (see: nanobsd) but I don't have time or clue to figure out how to do that with Linux. I belive that busybox is somewhat akin to nanobsd, but don't know how to get that onto the BBB hardware. -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@freebsd.org | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence. ___ 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] 5370 processor boards available
On 2/27/2014 6:38 AM, Poul-Henning Kamp wrote: They generally don't mount their filesystems read/write, but only read/only. I've done similar things with FreeBSD in many systems (see: nanobsd) but I don't have time or clue to figure out how to do that with Linux. The way TiVo does this is to have a RO / partition, and a RW /var partition. Normally, any writing is done in /var. If it gets corrupted, it gets rebuilt at boot time. But, they don't have to deal with user account and changing config files (those are stored outside the Linux partitions). But, the concept could be easily extended so perhaps / is only RW during short times when changes which need to be non-volatile are made. (soft link /tmp to somewhere in /var, too) I belive that busybox is somewhat akin to nanobsd, but don't know how to get that onto the BBB hardware. bb is just an all-in-one binary, which provides many standard *nix commands, often in a form just different enough from the gnu coreutils and POSIX specs to screw things up when you're not watching. ___ 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] 5370 processor boards available
Not that I can code anything. I do like several comments I see considering real life always messes everything up. The ability to download new code easily. To complement that the ability to upload the current system with IPs and such. Most routers support this approach. The other comment make the file system RO if possible. Batteries and such are always a mess to get right. Regards Paul. WB8TSL On Thu, Feb 27, 2014 at 6:38 AM, Poul-Henning Kamp p...@phk.freebsd.dkwrote: In message canx10ha2qmiwc2js8xfn3jat1vxntnnbzxa8-fhq7cybf52...@mail.gmail.com , Dr. David Kirkby writes: On 27 February 2014 08:26, Hal Murray hmur...@megapathdsl.net wrote: Consumer devices like routers and modems run Linux. Does anyone know how they get around this issue? They generally don't mount their filesystems read/write, but only read/only. I've done similar things with FreeBSD in many systems (see: nanobsd) but I don't have time or clue to figure out how to do that with Linux. I belive that busybox is somewhat akin to nanobsd, but don't know how to get that onto the BBB hardware. -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@freebsd.org | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence. ___ 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] 5370 processor boards available
The BBB has 2GB of flash on board (non removable) and has a micro SD socket. Would not be too hard to keep a backup copy of the OS and apps on the SD card so that it would be easy to boot from SD and reload the built-in flash if the BBB fails to boot from the built-in image. That would not be a replacement for a workable, safe boot process that can be interrupted without trashing the file system. Just a few days ago, an overnight storm caused power to flicker maybe 5 times in less than a minute. My Raspberry Pi managed to survive it, but I am not sure that will always be the case. I think a super cap with proper shutdown routine would probably the easiest to implement. Alternately, running the flash read only, copy OS, apps and everything else to a RAM disk and run from the RAM disk would probably be the safest. Not sure you can do that with 512MB of RAM without some serious pruning of the Linux kernel. Didier KO4BB On February 27, 2014 2:26:13 AM CST, Hal Murray hmur...@megapathdsl.net wrote: I think a better solution would be to find a very large super cap and power the BBB from that while giving it a power fail interrupt to quickly sync the file system. The advantage of something like the BBB is that it runs Linux so you have a nice environment in which to run your code. The disadvantage of Linux is that it's complicated and you have things like file systems that can get trashed. Now, in addition of being the sysadmin for your PC(s), you have to be an admin for your lab gear. That may be more interesting that you were expecting. A friend reports that his scope caught a virus... The Linux ext4 file system is pretty robust. There are lots of PCs out there that mostly survive power failures. If I was worried about the file system getting trashed on power off, I'd work on the software long before I added a super-cap. I think my first try would be to run the file system read-only until I figured out that I needed to write a file. Then I would know something about how much data I wanted to write and the usage patterns. You can help a lot with flush() in the right places in your code. That may cost performance if you are writing a lot of data. Another approach is to make sure you can put the disk back together easily and quickly. Then it's not such a big deal if/when the disk gets trashed. That's probably a good idea anyway. Power fail isn't the only thing that can trash a disk. It shouldn't take much more than a simple script to format the disk and copy over all the bits from a backup place on a PC. Maybe it's install the standard distro package and then add your bits. (That's assuming you can take the disk out of the BBB and plug it into a PC.) -- These are my opinions. I hate spam. ___ 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. -- Sent from my Motorola Droid Razr 4G LTE wireless tracker while I do other things. ___ 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] 5370 processor boards available
The only time there is any exposure is during a write operation. When the processor board is used to run the 5370, how often is data written and what is the exposure interval? I would be willing to bet that Linux already has a power-fail NMI input. I would bet that you can find out what the worst-case PF NMI latency is and then ensure that the PS output stays up at least that long without having to worry about a battery or super cap. (I bet PF latency is under 1ms.) In that case, all you need to do is assert the PF interrupt line. Hmm, check to see if it is already there and based on nothing more than sagging Vcc/Vdd. I bet it is and therefore this whole discussion may be moot. But the idea of making the boot partition RO and the /var partition R/W makes a whole lot of sense to me. -- Brian Lloyd, WB6RQN/J79BPL 706 Flightline Drive Spring Branch, TX 78070 br...@lloyd.com +1.916.877.5067 ___ 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] 5370 processor boards available
On 2/27/14 6:40 AM, Didier Juges wrote: The BBB has 2GB of flash on board (non removable) and has a micro SD socket. Would not be too hard to keep a backup copy of the OS and apps on the SD card so that it would be easy to boot from SD and reload the built-in flash if the BBB fails to boot from the built-in image. Alternately, running the flash read only, copy OS, apps and everything else to a RAM disk and run from the RAM disk would probably be the safest. Not sure you can do that with 512MB of RAM without some serious pruning of the Linux kernel. Back in 2004, I was running a stripped down linux off a small compact flash (256MB or 512MB, I think) on a VIA Eden 700 MHz motherboard with very limited RAM. It included busybox to provide the usual command line utilities. As I dimly recall, it booted from a compressed filesystem (on CF) to ramdisk only. I also had a version of Debian that net-booted, and also ran entirely in ram. I could probably dig it up if someone's interested. We run RTEMS (www.rtems.org) at work, which provides a Posix compliant interface, but is designed for embedded systems, and has a bunch of targets (ours is SPARC, but there's x86 versions, for sure). You can definitely fit that in well under 128 MB. We've found that most everything that compiles and runs under Linux will compile/run under RTEMS, unless you're using something peculiar to Linux, or you need dynamic loading/linking (RTEMS is statically linked). That is, we have very few #ifdef LINUX kind of things in our code. ___ 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] Serial port splitter s/w
A number of people have reported issues with the Raspberry Pi Ethernet hardware when used for NTP, as it is actually a USB-Ethernet bridge and the drivers may not be all they could be. I have not had problems myself but I do not run NTP on it. The Beaglebone Black (BBB) is supposed to be better in that regard since the Ethernet MAC is directly on the CPU. The BBB is also entirely open source, unlike the proprietary Broadcom CPU on the RPi. Didier KO4BB On February 26, 2014 11:32:36 AM CST, Chris Albertson albertson.ch...@gmail.com wrote: It's not going to work. If the purpose of running the Thunderbolt is only to drive NTP then you don't need LH. NTP's only tags the pulses to the nearest microsecond, nano sec on accuracy is lost on NTP. I'd even say the TB is the wrong GPS for NTP. It costs to much and uses to much power. But if you are also, or mainly, using the Thunderbolt for it's 10Mhz and NTP is a secondary function then a TB makes sense. Yes you NEED the PPS on the serial cable. Thunderbolts do not send NMEA. Thunderbolts send their own data format that is unique to Trimble. Don't modify the GPS receiver. Make a special cable adapter. When you do this pay attention to polarity of the PPS signal. It is easy to get it backwards. You want the raising edge of the TB pulse to interrupt the computer. It you invert the signal the wrong number of times the time will be off by the ouse length and I don't know if the pulse length is controlled to the level the leading edge is.Remember RS232 uses negative and positive voltage, data lines use negative logic, control lines positive. The TB's PPS is TTL level. Many rs232 ports do accept t/l level if you get the polariy correct. Again don't even bother to run an NTP server without PPS. You may as well just get time from some internet time servers. You can NOT control a GPS from two ports. Both NTP and LH will try to send commands to the GPS. Likely, almost certainly you need to build a small circuit board the has two connectors that face the TB (PPS and serial) and one that faces the computer. The little perfboard makes a neat way to or connect cables but you could solder up a y--cable The best thing to do is get a cheaper GPS, and one that uses less power to drive the NTP server. The old Motorola Oncore series are cheap and the new breed of very small GPSes are good too. DOn't spend more than $40 or $50 on a GPS to drive NTP as ,again NTP record microseconds. You could free up that Windows PC too. It is not the best platform for NTP. Asmall ARM based system (even the Rasbery Pi) will outperform a Windows based NTP server. and use a LOT less power (Power cost for a NTP server is more than you think, it came to about $300 a year for me if I used a standard PC and a thunderbolt. Switching to a very tiny ARM based system and a smaller GPS gave as good performance and power savings paid for the hardware in 1/2 a year. $.21/KWH about 170W and 8760 hours per year comes to a $300 power bill. My current system is powered by a 1000 mw plug-in power cube and does not need a cooling fan. On Wed, Feb 26, 2014 at 7:53 AM, David C. Partridge david.partri...@perdrix.co.uk wrote: I'm running Meinberg NTP on the Windows 7 x64 machine to which my Thunderbolt is attached. I'd like to be able to share the serial port between LH and NTP so that I can run the machine as an NTP Stratum 1 server locked to the TB, and also be able to use LH to check things. I looked around the with Google, and saw *numerous serial port splitters. Which is recommended? Also what's the best way how to configure NTP to lock the the TB on a serial port? Do I need to modify the TB to deliver the PPS down one of the serial data lines or will NTP work well by parsing the NMEA time messages? Many thanks David Partridge ___ 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. -- Chris Albertson Redondo Beach, California ___ 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. -- Sent from my Motorola Droid Razr 4G LTE wireless tracker while I do other things. ___ 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] GPSDO with all-digital phase/time measurement?
Hi Hal, I've reviewed this a bit and I used the wrong terminology. Either Timer1 or Timer3 can use the T1CKI input as a clock. Along with that, the CCP1 or CCP2 pin can be used to trigger a capture of the timer in use into the CCPR1 or CCPR2 register pair. I had considered this as an interrupt function, but after looking at the manual, it's a matter of switchable dedicated hardware. Sorry for the bad info. Bob From: Hal Murray hmur...@megapathdsl.net To: Bob Stewart b...@evoria.net; Discussion of precise time and frequency measurement time-nuts@febo.com Cc: hmur...@megapathdsl.net Sent: Thursday, February 27, 2014 1:13 AM Subject: Re: [time-nuts] GPSDO with all-digital phase/time measurement? b...@evoria.net said: At least on the PIC I'm using, the CCP and timer interrupts don't seem to be synchronous with the PIC clock. I could be mistaken. Unless you have a very strange architecture, it doesn't make sense for an interrupt to not be synchronous with the CPU clock. You are in the middle of an add instruction, and now you want to start an interrupt. What does that mean? I expect there is the standard 2 FF synchronizer on all the input pins. Things like the counter/timers run on the CPU clock, taking their input after the synchronizer. I don't remember seeing a data sheet that comes out and says that, but sometimes you can get some (strong?) hints with things like minimum pulse widths or max clock rate. -- These are my opinions. I hate spam. ___ 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] Serial port splitter s/w
Yes the BBB has about twice or more the performance over the Pi except in graphics. The Pi has a faster GPU. But in the case of an NTP server you likely would never connect a monitor or keyboard to the computer so the GPU will go unused. However the NTP is very un-demanding and will only see on average of maybe about one network packet per second or less. In fact using an entire BBB or Pi just for NTP is kind of a waste as it could be doing other things at the same time. Put some disks on it as use it as a file server for backups for the other computers. Lots of other uses you could put it to. On another list server I'm reading about how to use discarded cell phones. These older Andriod cell phones that people don't want are many times available for fee and have mamory and processors like those in the BBB or Pi. They use very little power, have built-in barrty backup and have nice built in displays. They can be re-purposed as tiny computers. (Think of a TIC or GPSDO with a color display and touch screen user interface.) The problem is making the physical connections, that takes some work and reading. On Thu, Feb 27, 2014 at 6:56 AM, Didier Juges shali...@gmail.com wrote: A number of people have reported issues with the Raspberry Pi Ethernet hardware when used for NTP, as it is actually a USB-Ethernet bridge and the drivers may not be all they could be. I have not had problems myself but I do not run NTP on it. The Beaglebone Black (BBB) is supposed to be better in that regard since the Ethernet MAC is directly on the CPU. The BBB is also entirely open source, unlike the proprietary Broadcom CPU on the RPi. Didier KO4BB On February 26, 2014 11:32:36 AM CST, Chris Albertson albertson.ch...@gmail.com wrote: It's not going to work. If the purpose of running the Thunderbolt is only to drive NTP then you don't need LH. NTP's only tags the pulses to the nearest microsecond, nano sec on accuracy is lost on NTP. I'd even say the TB is the wrong GPS for NTP. It costs to much and uses to much power. But if you are also, or mainly, using the Thunderbolt for it's 10Mhz and NTP is a secondary function then a TB makes sense. Yes you NEED the PPS on the serial cable. Thunderbolts do not send NMEA. Thunderbolts send their own data format that is unique to Trimble. Don't modify the GPS receiver. Make a special cable adapter. When you do this pay attention to polarity of the PPS signal. It is easy to get it backwards. You want the raising edge of the TB pulse to interrupt the computer. It you invert the signal the wrong number of times the time will be off by the ouse length and I don't know if the pulse length is controlled to the level the leading edge is.Remember RS232 uses negative and positive voltage, data lines use negative logic, control lines positive. The TB's PPS is TTL level. Many rs232 ports do accept t/l level if you get the polariy correct. Again don't even bother to run an NTP server without PPS. You may as well just get time from some internet time servers. You can NOT control a GPS from two ports. Both NTP and LH will try to send commands to the GPS. Likely, almost certainly you need to build a small circuit board the has two connectors that face the TB (PPS and serial) and one that faces the computer. The little perfboard makes a neat way to or connect cables but you could solder up a y--cable The best thing to do is get a cheaper GPS, and one that uses less power to drive the NTP server. The old Motorola Oncore series are cheap and the new breed of very small GPSes are good too. DOn't spend more than $40 or $50 on a GPS to drive NTP as ,again NTP record microseconds. You could free up that Windows PC too. It is not the best platform for NTP. Asmall ARM based system (even the Rasbery Pi) will outperform a Windows based NTP server. and use a LOT less power (Power cost for a NTP server is more than you think, it came to about $300 a year for me if I used a standard PC and a thunderbolt. Switching to a very tiny ARM based system and a smaller GPS gave as good performance and power savings paid for the hardware in 1/2 a year. $.21/KWH about 170W and 8760 hours per year comes to a $300 power bill. My current system is powered by a 1000 mw plug-in power cube and does not need a cooling fan. On Wed, Feb 26, 2014 at 7:53 AM, David C. Partridge david.partri...@perdrix.co.uk wrote: I'm running Meinberg NTP on the Windows 7 x64 machine to which my Thunderbolt is attached. I'd like to be able to share the serial port between LH and NTP so that I can run the machine as an NTP Stratum 1 server locked to the TB, and also be able to use LH to check things. I looked around the with Google, and saw *numerous serial port splitters. Which is recommended? Also what's the best way how to configure NTP to lock the the TB on a serial port? Do I need to modify the TB to deliver the
Re: [time-nuts] Serial port splitter s/w
A number of people have reported issues with the Raspberry Pi Ethernet hardware when used for NTP, as it is actually a USB-Ethernet bridge and the drivers may not be all they could be. I have not had problems myself but I do not run NTP on it. The Beaglebone Black (BBB) is supposed to be better in that regard since the Ethernet MAC is directly on the CPU. The BBB is also entirely open source, unlike the proprietary Broadcom CPU on the RPi. Didier KO4BB Didier, Here are the results I saw when using Ethernet sync for NTP on a Raspberry Pi: http://www.satsignal.eu/ntp/Raspberry-Pi-NTP.html#results Syncing just from my Cable Modem WAN connection it's quite bad (offsets +/- 3 milliseconds), but when syncing to local stratum-1 servers the offsets were reduced to something slightly in excess of +/- 20 microseconds. Adding PPS reduces that to a couple of microseconds (with Linux 3.6.11 compiled as a non-tickless system) as seen further down under the Current performance heading. Perhaps there are equivalent graphs for the BBB system somewhere? The Raspberry Pi is sufficiently open that you can recompile programs, and even recompile the kernel for it. 73, David GM8ARV -- SatSignal Software - Quality software written to your requirements Web: http://www.satsignal.eu Email: david-tay...@blueyonder.co.uk ___ 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] GPSDO with all-digital phase/time measurement?
Hi Alex, As these guys have told me, and as I have learned, Rb standards are very regular, but that doesn't mean they are very accurate. So, unless your Rb standard was being disciplined to the right frequency you'd probably have to do some sort of analysis to see the quantizing errors on your 1PPS pulse. Bob From: Alexander Pummer alex...@ieee.org To: Bob Stewart b...@evoria.net; Discussion of precise time and frequency measurement time-nuts@febo.com Sent: Thursday, February 27, 2014 7:42 AM Subject: Re: [time-nuts] GPSDO with all-digital phase/time measurement? it would be interesting to see the accuracy of the 1pps pulses by comparing them with a second 1pps pulse, which is derived from a rubidium standard, which on his own does not have quantizing errors, 73 KJ6UHN Alex ___ 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] GPSDO with all-digital phase/time measurement?
it would be interesting to see the accuracy of the 1pps pulses by comparing them with a second 1pps pulse, which is derived from a rubidium standard, which on his own does not have quantizing errors, 73 KJ6UHN Alex On 2/26/2014 6:03 PM, Bob Stewart wrote: Hi Mark, I'm neither an engineer, nor an expert, but here are my comments. I think that the idea of 100ns/T is wrong. There are several variables that control accuracy, but the time between pulses from your OCXO (assuming no phase or frequency drift) isn't one of them. So, that gives 1/T. Here the problem is that T must get large before your accuracy can be good. You can achieve very good accuracy, but at the cost of waiting thousands of seconds between phase points; i.e. where your 1PPS coincides with the 10 millionth OCXO pulse. The theoretical maximum would be infinity, of course, but your oscillator won't be that stable. Another big problem is the accuracy of the 1PPS pulse. I'm using an Adafruit GPS receiver, and it's listed as accurate to within 10ns. And it is, but you have to be wary of exactly what that means. It doesn't mean +/- 5ns. So, as your 1PPS pulse bobs back and forth, you will often encounter an OCXO pulse up to 10ns early, or up to 10ns late. So, might you count 9,999,999 pulses from the OCXO immediately followed by 10,000,001 pulses. Neither of those, by itself is a signal to change the EFC voltage to your OCXO. In fact, it is normal for your count to alternate between the two for long periods, if you are very very close to exactly 10MHz, just from the quantization error on the 1PPS. It is also normal for 1/T to control the time between phase crossings. So you have to wait for two miscounts in a row in the same direction to make a change. And even then, you can't be 100% sure that it's not due to the quantization errors in your 1PPS signal. The better GPS receivers will output a quantization error value every second. But if you're using the 1/T method, there's nothing you can do with it, so you have to live with whatever quantization errors you get. Anyway, those are my experiences. Bob - AE6RV From: Mark Haun hau...@keteu.org To: time-nuts@febo.com Sent: Wednesday, February 26, 2014 3:51 PM Subject: [time-nuts] GPSDO with all-digital phase/time measurement? Hi everyone, I'm new to the list, and have been reading the recent threads on Arduino-based GPSDOs and the pros/cons of 10-kHz vs 1-Hz time pulses with interest. As I understand it, there are a couple of reasons why one needs a time-interval / phase measurement implemented outside the MCU: 1) Time resolution inside the MCU is limited by its clock period, which is much too coarse. The GPSDO would ping-pong within a huge dead zone. 2) Software tends to inject non-determinism into the timing. snip Thanks, Mark KJ6PC ___ 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] 5370 processor boards available
If having a battery in the device makes some of you queasy, you could always put it outside, on the back. But I seriously believe there are ways to make battery backup which will not leak. D. On 27 Feb 2014 06:52, Chuck Harris cfhar...@erols.com wrote: That fix presumes that the power line never quits at inappropriate times. This winter has provided me with ample reminders that power can go out anytime. I think a better solution would be to find a very large super cap and power the BBB from that while giving it a power fail interrupt to quickly sync the file system. -Chuck Harris John Seamons wrote: On Feb 25, 2014, at 12:59 PM, John Seamons j...@jks.com wrote: I may have a solution for the power-off problem that doesn't involve batteries or supercaps. It has the added advantage of providing instant-on. From the latest documentation: One solution to the annoyance of having to halt the BBB before the instrument is powered off is simply to keep the BBB running by providing it a secondary source of power via the USB-mini port. The BBB already understands how to select between two sources of power: the USB-mini and +5V barrel connectors (the latter of which is actually being delivered from the 5370 via the board expansion connectors). The app detects when the instrument has powered down and resumes running when power is restored. USB power could come from an external USB hub or charger. You may have to obtain a longer USB cable than the one supplied. Be certain to only use the USB-mini connector that is adjacent to the Ethernet RJ45 connector. Do not use the USB-A connector on the other end of the board as that port will not accept input power. Some of you might even figure out how to derive +5V for the USB-mini cable from the 5370 oven power supply card that is always powered if the instrument is off but the line cord is plugged in. But beware of the potential problem of conducting noise from the BBB to the reference oscillator over such a connection. Some experimentation and measurement is necessary. The planned USB/Ethernet connector card that replaces the current HPIB one at the back-panel could host a voltage regulator. Aligator clip connections to the '+25V UNREG' and GND test points on the oven power supply would alleviate the need for any soldering. ___ 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] 5370 processor boards available
You can fit linux with a comfortable amount of stuff in 20 MB, but you want more than that for more advanced programs. You should not use the built in ssd for writing. It has limited writes and cannot be replaced. Use a card. You can tailor linux so that it has minimal services but the question of how to make it shut down in a predictable amount of time is a complex issue and you are best off finding an existing project that focuses on this. So rather than limit yourself to an esoteric set of requirements I recommend using a battery that'll run the BBB for 15 minutes and will charge in an hour while it's on. My day job is among others as a linux admin. D. On 27 Feb 2014 15:47, Didier Juges shali...@gmail.com wrote: The BBB has 2GB of flash on board (non removable) and has a micro SD socket. Would not be too hard to keep a backup copy of the OS and apps on the SD card so that it would be easy to boot from SD and reload the built-in flash if the BBB fails to boot from the built-in image. That would not be a replacement for a workable, safe boot process that can be interrupted without trashing the file system. Just a few days ago, an overnight storm caused power to flicker maybe 5 times in less than a minute. My Raspberry Pi managed to survive it, but I am not sure that will always be the case. I think a super cap with proper shutdown routine would probably the easiest to implement. Alternately, running the flash read only, copy OS, apps and everything else to a RAM disk and run from the RAM disk would probably be the safest. Not sure you can do that with 512MB of RAM without some serious pruning of the Linux kernel. Didier KO4BB On February 27, 2014 2:26:13 AM CST, Hal Murray hmur...@megapathdsl.net wrote: I think a better solution would be to find a very large super cap and power the BBB from that while giving it a power fail interrupt to quickly sync the file system. The advantage of something like the BBB is that it runs Linux so you have a nice environment in which to run your code. The disadvantage of Linux is that it's complicated and you have things like file systems that can get trashed. Now, in addition of being the sysadmin for your PC(s), you have to be an admin for your lab gear. That may be more interesting that you were expecting. A friend reports that his scope caught a virus... The Linux ext4 file system is pretty robust. There are lots of PCs out there that mostly survive power failures. If I was worried about the file system getting trashed on power off, I'd work on the software long before I added a super-cap. I think my first try would be to run the file system read-only until I figured out that I needed to write a file. Then I would know something about how much data I wanted to write and the usage patterns. You can help a lot with flush() in the right places in your code. That may cost performance if you are writing a lot of data. Another approach is to make sure you can put the disk back together easily and quickly. Then it's not such a big deal if/when the disk gets trashed. That's probably a good idea anyway. Power fail isn't the only thing that can trash a disk. It shouldn't take much more than a simple script to format the disk and copy over all the bits from a backup place on a PC. Maybe it's install the standard distro package and then add your bits. (That's assuming you can take the disk out of the BBB and plug it into a PC.) -- These are my opinions. I hate spam. ___ 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. -- Sent from my Motorola Droid Razr 4G LTE wireless tracker while I do other things. ___ 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] 5370 processor boards available
In message CA+9GZUhJLqQ7K-cNXsSzPMgSOH6Oyd774SM_e0=ykljgyd4...@mail.gmail.com , cheater00 . writes: If having a battery in the device makes some of you queasy, you could always put it outside, on the back. But I seriously believe there are ways to make battery backup which will not leak. Seriously: Keeping the computer alive and eating power is silly. We just need to find out how to get the Linux configured correctly. (Or port it all to FreeBSD, which I'm considering, because there I know how to do it :-) -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@freebsd.org | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence. ___ 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] 5370 processor boards available
On Thu, Feb 27, 2014 at 12:03 PM, Poul-Henning Kamp p...@phk.freebsd.dkwrote: Seriously: Keeping the computer alive and eating power is silly. We just need to find out how to get the Linux configured correctly. This is a solved problem. I don't understand why this discussion is dragging on. E.g liveCD, ramdisk, Linux routers, thin (diskless) clients. The relevant variables are the distribution and if you want to use a network disk. I will say the underlying concerns are valid. I have two Raspberry Pis and three Beable Bone Blacks. I've had four SD cards fail among the group and they're not even on most of the time. CompactFlash is much better. ___ 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] 5370 processor boards available
There is a spin of FreeBSD called NanoBSD, little light on the documentation but the name pretty much says what it is about. Either way it looks like the BBB is opening up quite a few minds. Finally a good combination of power, i/o etc. and not to mention the two PRU's .. just read up on them last night, very interesting. -pete On Thu, Feb 27, 2014 at 9:21 AM, Chuck Harris cfhar...@erols.com wrote: Do you have experience with FreeBSD on ARM processors? The simple solution, which may not be possible, is to make all software on the flash disk RO, and put any part of the file system that needs to be written onto a RAM disk... probably should be the /var directory. The only time the flash disk must be written is when some configuration parameter must be changed, or the linux/program image must be changed. The routines that make those changes could be fashioned so that they flush the caches immediately after they are finished. From observation, the BBB keeps operating for several seconds after power is removed. If a power failure interrupt is created that will notify linux instantly on power failure, it would be possible to forbid configuration changes that take longer than a couple of seconds. -Chuck Harris Poul-Henning Kamp wrote: In message CA+9GZUhJLqQ7K-cNXsSzPMgSOH6Oyd774SM_e0=ykLJg yd4...@mail.gmail.com , cheater00 . writes: If having a battery in the device makes some of you queasy, you could always put it outside, on the back. But I seriously believe there are ways to make battery backup which will not leak. Seriously: Keeping the computer alive and eating power is silly. We just need to find out how to get the Linux configured correctly. (Or port it all to FreeBSD, which I'm considering, because there I know how to do it :-) ___ 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] 5370 processor boards available
Backup batteries don't make me queasy. The easiest, cheapest backup battery would be a small 6V 3AH sealed lead acid battery. It is trivial to make a good float charger using a 3 terminal regulator and a diode, and they do not leak, vent, or anything problematic. The little 6V 3AH packs are used in emergency lighting, so they are readily available. Float them at 6.9V, and they will last at least 5 years. I suggested a super cap because they are cheap, and tend to last a very long time without leaking. 100F will keep a BBB up for quite a while if it isn't driving the network, or usb devices. Regardless of the power safe mechanism, it would be best that the BBB get an interrupt at the moment power is deemed bad, and starts an orderly shutdown. To guard against power bouncing, it should be prevented from restarting until several seconds after shutdown is completed... some sort of timer should be part of the power safe mechanism. -Chuck Harris cheater00 . wrote: If having a battery in the device makes some of you queasy, you could always put it outside, on the back. But I seriously believe there are ways to make battery backup which will not leak. D. On 27 Feb 2014 06:52, Chuck Harris cfhar...@erols.com wrote: That fix presumes that the power line never quits at inappropriate times. This winter has provided me with ample reminders that power can go out anytime. I think a better solution would be to find a very large super cap and power the BBB from that while giving it a power fail interrupt to quickly sync the file system. -Chuck Harris ___ 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] 5370 processor boards available
At 11:03 AM 2/27/2014, Poul-Henning Kamp wrote: Seriously: Keeping the computer alive and eating power is silly. We just need to find out how to get the Linux configured correctly. Exactly. We use the PC-Engines ALIX x86 single board computer stuff at work, and even I didn't have too much trouble getting a pretty much stock Debian installation 'adjusted' to run with the CF card mounted as read-only. I've yet to see a problem with power cycles. (But I've only done the power cycle bench test a few hundred loops.) -- 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.
Re: [time-nuts] 5370 processor boards available
Do you have experience with FreeBSD on ARM processors? The simple solution, which may not be possible, is to make all software on the flash disk RO, and put any part of the file system that needs to be written onto a RAM disk... probably should be the /var directory. The only time the flash disk must be written is when some configuration parameter must be changed, or the linux/program image must be changed. The routines that make those changes could be fashioned so that they flush the caches immediately after they are finished. From observation, the BBB keeps operating for several seconds after power is removed. If a power failure interrupt is created that will notify linux instantly on power failure, it would be possible to forbid configuration changes that take longer than a couple of seconds. -Chuck Harris Poul-Henning Kamp wrote: In message CA+9GZUhJLqQ7K-cNXsSzPMgSOH6Oyd774SM_e0=ykljgyd4...@mail.gmail.com , cheater00 . writes: If having a battery in the device makes some of you queasy, you could always put it outside, on the back. But I seriously believe there are ways to make battery backup which will not leak. Seriously: Keeping the computer alive and eating power is silly. We just need to find out how to get the Linux configured correctly. (Or port it all to FreeBSD, which I'm considering, because there I know how to do it :-) ___ 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] frequency comparator reading question
The question can't be answered without knowing what the range switch does. In my experience, the cycle that is divided into 360 degrees is the period of the input signal regardless of the range switch. You don't say, but the usual GPSDO produces a 10 MHz signal, unless it's for a telco application. If the phase meter goes through one cycle in a second there is a one cps difference between the signals. One cycle at 10 MHz is one part in 10E7. One cycle in a kilohertz is 1 part in 1000. 1 cycle in 100 seconds for 10 MHz is one part in 10E9 Does the time for a phase rotation vary with the range switch? Bill Hawkins -Original Message- From: Bob Camp Sent: Wednesday, February 26, 2014 9:16 PM Not having one here, about all I can guess is that there are 360 degrees in a cycle. If it's going through 360 degrees in 10 seconds it's 0.1 Hz off at what ever point it's comparing. If it takes 100 seconds that's 0.01 Hz. Yes I get this pesky decimal point stuff wrong from time to time ... Bob On Feb 26, 2014, at 9:19 PM, Paul A. Cianciolo pa...@snet.net wrote: I have a Fluke montronics frequency comparator. It has 2 inputs, one from my GPS and one from my DUT. After a given oscillator is warmed up, I can read the meter in parts 10 -X There are 2 meters, One for phase 0 to 360 degrees, and one for part to the -nth In the 10-9th position on the selector switch the a given oscillator will show a 0 to 360 degrees travel lets say in 1 second. ___ 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] 5370 processor boards available
On Thu, Feb 27, 2014 at 6:03 PM, Poul-Henning Kamp p...@phk.freebsd.dk wrote: In message CA+9GZUhJLqQ7K-cNXsSzPMgSOH6Oyd774SM_e0=ykljgyd4...@mail.gmail.com , cheater00 . writes: If having a battery in the device makes some of you queasy, you could always put it outside, on the back. But I seriously believe there are ways to make battery backup which will not leak. Seriously: Keeping the computer alive and eating power is silly. We just need to find out how to get the Linux configured correctly. (Or port it all to FreeBSD, which I'm considering, because there I know how to do it :-) Oh, of course it's not about keeping the computer on at all times, it's about having a contingency for the 0.1% case when your computer does not shut down in the assumed time. Unless you develop your system and everything running in it as hard real time, you cannot escape the fact that sometimes the thing will take longer than expected to shut down. And at some point that will cause data loss. A suitable battery costs $5. Is your time worth so much less than $5 that you actually consider the task of building your own Linux distribution for this case, or even just finding one that will work relatively problem-free? D. ___ 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] 5370 processor boards available
On Thu, Feb 27, 2014 at 6:22 PM, Scott Newell newell+timen...@n5tnl.com wrote: At 11:03 AM 2/27/2014, Poul-Henning Kamp wrote: Seriously: Keeping the computer alive and eating power is silly. We just need to find out how to get the Linux configured correctly. Exactly. We use the PC-Engines ALIX x86 single board computer stuff at work, and even I didn't have too much trouble getting a pretty much stock Debian installation 'adjusted' to run with the CF card mounted as read-only. I've yet to see a problem with power cycles. (But I've only done the power cycle bench test a few hundred loops.) -- newell N5TNL My assumption is that, unlike an embedded computer which is plug-and-forget, this thing will get modified, developed on, played around with, etc, all of which includes userland software (= possibly long shutdown times) and disk access (= possibly long shutdown times). It's trivial to run a read-only system but that's limiting and takes out the fun of tinkering with the BBB to make your own applications and other stuff. You might say that we could run the stock system RO and add any user-supplied stuff to RW media and just say that it isn't going to survive a power-down. But that's not really a comfortable situation. D. ___ 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] 5370 processor boards available
On 27 Feb 2014 16:55, cheater00 . cheate...@gmail.com wrote: If having a battery in the device makes some of you queasy, you could always put it outside, on the back. But I seriously believe there are ways to make battery backup which will not leak. Two batteries, each with a diode and bought to a common point seems logical to me. It only needs one present so either one can be changed with no loss of power. Guarding against battery faulty batteries that leak prematurely probably needs a physical barrier, but I can't believe it is rocket science ___ 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] frequency comparator reading question
These units multiply the frequency delta and the phase change quoted is probably 10^n times the actual difference. The output is f + n*delta(f) I have but do not use a Montronix 100-7 (before the Fluke purchase) Alan G3NYK - Original Message - From: Bill Hawkins b...@iaxs.net To: 'Discussion of precise time and frequency measurement' time-nuts@febo.com Sent: Thursday, February 27, 2014 5:55 PM Subject: Re: [time-nuts] frequency comparator reading question The question can't be answered without knowing what the range switch does. In my experience, the cycle that is divided into 360 degrees is the period of the input signal regardless of the range switch. You don't say, but the usual GPSDO produces a 10 MHz signal, unless it's for a telco application. If the phase meter goes through one cycle in a second there is a one cps difference between the signals. One cycle at 10 MHz is one part in 10E7. One cycle in a kilohertz is 1 part in 1000. 1 cycle in 100 seconds for 10 MHz is one part in 10E9 Does the time for a phase rotation vary with the range switch? Bill Hawkins -Original Message- From: Bob Camp Sent: Wednesday, February 26, 2014 9:16 PM Not having one here, about all I can guess is that there are 360 degrees in a cycle. If it's going through 360 degrees in 10 seconds it's 0.1 Hz off at what ever point it's comparing. If it takes 100 seconds that's 0.01 Hz. Yes I get this pesky decimal point stuff wrong from time to time ... Bob On Feb 26, 2014, at 9:19 PM, Paul A. Cianciolo pa...@snet.net wrote: I have a Fluke montronics frequency comparator. It has 2 inputs, one from my GPS and one from my DUT. After a given oscillator is warmed up, I can read the meter in parts 10 -X There are 2 meters, One for phase 0 to 360 degrees, and one for part to the -nth In the 10-9th position on the selector switch the a given oscillator will show a 0 to 360 degrees travel lets say in 1 second. ___ 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] 5370 processor boards available
It is interesting to watch this thread. The most interesting thing is the fact that no one has asked or answered the basic question: is there a problem? There are a lot of solutions for what may be a complete non-issue. Given that Linux and BSD are reliable and stable on standard PC platforms connected to mains power without a UPS, that suggests to me that there is likely no issue. Sure, back in the bad ol' days you had to sync;sync;sync the filesystem prior to shutdown and then, if there had been a power fail, fsck it on power up. But those days have been gone for a LONG time. The modern filesystems reduce the exposure to the possibility of file system corruption to a tiny probability, and then the system further reduces that by providing a power-fail detection system that allows the critical pending writes to be flushed prior to final power-fail, thus leaving the FS in a completely deterministic state. I suppose that on the night of a full moon when the lightning flashes and the dog barks, conditions might be right to corrupt the file system, but then, that could happen when the battery goes dead or catches fire too. ;-) It seems to me that the bard put it pretty well, ... much ado about nothing. -- Brian Lloyd, WB6RQN/J79BPL 706 Flightline Drive Spring Branch, TX 78070 br...@lloyd.com +1.916.877.5067 ___ 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] 5370 processor boards available
In message CAE3hgTcOU5cT3y-T9gYGJz4xeoob9dU8UEM4YgPuwvDWX6Q=6...@mail.gmail.com , Brian Lloyd writes: The only time there is any exposure is during a write operation. When the processor board is used to run the 5370, how often is data written and what is the exposure interval? It runs a full Linux, so there are logfiles, daemons and all sorts of stuff which writes. It needs to be pared down to a sensible embedded configuration. -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@freebsd.org | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence. ___ 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] 5370 processor boards available
In message 530f740a.7030...@erols.com, Chuck Harris writes: Do you have experience with FreeBSD on ARM processors? A little bit, I just started playing with it again recently. The simple solution, which may not be possible, is to make all software on the flash disk RO, and put any part of the file system that needs to be written onto a RAM disk... probably should be the /var directory. That's exactly what NanoBSD does, with a few more wrinkles to boot: https://www.freebsd.org/doc/en/articles/nanobsd/ -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@freebsd.org | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence. ___ 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] 5370 processor boards available
In message ca+9gzujtzon9+uxg1r1brxx8zyv4wnjq22qpsbityako20w...@mail.gmail.com , cheater00 . writes: Oh, of course it's not about keeping the computer on at all times, it's about having a contingency for the 0.1% case when your computer does not shut down in the assumed time. You're missing the point: If all the permanent filesystems are mounted read-only, and only ram-disks are mounted read-write, you don't even need to shut down, you can just yank the power. -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@freebsd.org | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence. ___ 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] 5370 processor boards available
There you have the problem in the above sentence from Poul-Henning. It needs to be yank the power proof. Because that is the way the 5370 shuts down. That said the board and software thats been created by one person is simply amazing and well done. We are discussing a fine adjustment. Batteries are an incredible mess. I suspect we need a fast detect interrupt from the incoming power supply. My experience with other Linuxes is that they waste way to much time flushing stuff. So the comment about making the file system RO after you have added the IP makes great sense. The ability to upload and then re-download also is good I can do that on my routers and believe actually this board can easily do the same. Its just at the end of the day when I am tired I will hit the off switch. Ooops lucky me I get to reload everything. Regards Paul WB8TSL On Thu, Feb 27, 2014 at 2:21 PM, Poul-Henning Kamp p...@phk.freebsd.dkwrote: In message ca+9gzujtzon9+uxg1r1brxx8zyv4wnjq22qpsbityako20w...@mail.gmail.com , cheater00 . writes: Oh, of course it's not about keeping the computer on at all times, it's about having a contingency for the 0.1% case when your computer does not shut down in the assumed time. You're missing the point: If all the permanent filesystems are mounted read-only, and only ram-disks are mounted read-write, you don't even need to shut down, you can just yank the power. -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@freebsd.org | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence. ___ 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] 5370 processor boards available
Hi Brian, it's less about caching pending writes to file handles that are waiting inside the system. That's most likely to work well within a short period of time. The issue is if you're using the system to its full extent. Among others you'd like to write applications for the BBB that enhance the 5370's capabilities. You don't want to bother with hard real time, compilation, memory layout, etc, all of them problems each of which will make your development time 10x longer and your fun time 20x shorter. So instead you use something like Python or Java. Then, you also don't really bother to ensure that your program can be interrupted at any time, since that's another level of complexity. Coroutines? Continuation passing style? Reactive programming? Why bother? Then you make a few other decisions that make your program easier to make and more difficult to shut down immediately in a clean manner. So let's say it's running and it is collecting data which is crucial to you. Maybe you again didn't bother to make sure that if the program crashes it will not corrupt its data. So now your BBB gets the power loss interrupt, sends kill -15 to all processes and most of them exit. Your program doesn't so likely Linux tries to send signals 2, then 1, then 9 at which point your program will certainly crash (sig 9 means crash now). So your program will likely eat all the data. Note that all of the shortcuts I mentioned above are fairly reasonable defaults that save days of work per project for no improvement in reliability, since, after all, a suitable battery only costs $5. Cheers, D. On Thu, Feb 27, 2014 at 7:47 PM, Brian Lloyd br...@lloyd.com wrote: It is interesting to watch this thread. The most interesting thing is the fact that no one has asked or answered the basic question: is there a problem? There are a lot of solutions for what may be a complete non-issue. Given that Linux and BSD are reliable and stable on standard PC platforms connected to mains power without a UPS, that suggests to me that there is likely no issue. Sure, back in the bad ol' days you had to sync;sync;sync the filesystem prior to shutdown and then, if there had been a power fail, fsck it on power up. But those days have been gone for a LONG time. The modern filesystems reduce the exposure to the possibility of file system corruption to a tiny probability, and then the system further reduces that by providing a power-fail detection system that allows the critical pending writes to be flushed prior to final power-fail, thus leaving the FS in a completely deterministic state. I suppose that on the night of a full moon when the lightning flashes and the dog barks, conditions might be right to corrupt the file system, but then, that could happen when the battery goes dead or catches fire too. ;-) It seems to me that the bard put it pretty well, ... much ado about nothing. -- Brian Lloyd, WB6RQN/J79BPL 706 Flightline Drive Spring Branch, TX 78070 br...@lloyd.com +1.916.877.5067 ___ 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] 5370 processor boards available
Most routers use a similar model and can save important settings but still recover from a crash with no problem. There are several router distros that are good examples on how to do it. I would suggest looking at Voyage Linux http://linux.voyage.hk/ for an example. They have a specific versing for the BeagleBone Black. You can run user space apps on it just fine and it can save changes as necessary. I'm sure there are other examples as well. I'm just looking to making the box more reliable and stable. Running apps on it seems more complex than I'm up for. However a web page to control and read back seems like a nice trick. Demian On Thu, Feb 27, 2014 at 2:21 PM, Poul-Henning Kamp p...@phk.freebsd.dkwrote: In message ca+9gzujtzon9+uxg1r1brxx8zyv4wnjq22qpsbityako20w...@mail.gmail.com , cheater00 . writes: Oh, of course it's not about keeping the computer on at all times, it's about having a contingency for the 0.1% case when your computer does not shut down in the assumed time. You're missing the point: If all the permanent filesystems are mounted read-only, and only ram-disks are mounted read-write, you don't even need to shut down, you can just yank the power. -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@freebsd.org | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence. ___ 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] 5370 processor boards available
In message CA+9GZUgrLw05f=pvh5tji+vlkfstwatibcpe1t3xfgwq2+s...@mail.gmail.com , cheater00 . writes: Note that all of the shortcuts I mentioned above are fairly reasonable defaults that save days of work per project for no improvement in reliability, since, after all, a suitable battery only costs $5. Lets just say that I seem to use fairly, resonable, save, improvement and suitable in a different way than you do. The easiest way to do the kind of stuff you're talking about, is to NFS mount a filesystem on the BBB. -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@freebsd.org | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence. ___ 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] 5370 processor boards available
On 27 February 2014 17:31, Chuck Harris cfhar...@erols.com wrote: Backup batteries don't make me queasy. The easiest, cheapest backup battery would be a small 6V 3AH sealed lead acid battery. It is trivial to make a good float charger using a 3 terminal regulator and a diode, and they do not leak, vent, or anything problematic. The little 6V 3AH packs are used in emergency lighting, so they are readily available. Float them at 6.9V, and they will last at least 5 years. I thought there was an issue with even sealed lead acids of them emitting small amounts of sulfuric acid, which is not great in a bit of test equipment. Dave ___ 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] 5370 processor boards available
I have worked on lots of equipment that was sealed with a SLA inside, and I have never seen any sign of corrosion. Only times I have seen a problem is in the case of UPS's that were left discharged in unheated warehouses... the electrolyte froze and the battery cases bulged greatly, but still no leakage. -Chuck Harris Dr. David Kirkby wrote: On 27 February 2014 17:31, Chuck Harris cfhar...@erols.com wrote: Backup batteries don't make me queasy. The easiest, cheapest backup battery would be a small 6V 3AH sealed lead acid battery. It is trivial to make a good float charger using a 3 terminal regulator and a diode, and they do not leak, vent, or anything problematic. The little 6V 3AH packs are used in emergency lighting, so they are readily available. Float them at 6.9V, and they will last at least 5 years. I thought there was an issue with even sealed lead acids of them emitting small amounts of sulfuric acid, which is not great in a bit of test equipment. Dave ___ 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] 5370 processor boards available
Totally agree with Chucks comment I have opened many system that have been in service at least 5 years virtually no evidence of corrosion. I have however opened system that do not manage the batteries correctly and they are a mess. But we really are drifting away from the main thread. Just want to turn the 5370 off at the end of the day and back on the next. No think-um. Regards Paul On Thu, Feb 27, 2014 at 4:26 PM, Chuck Harris cfhar...@erols.com wrote: I have worked on lots of equipment that was sealed with a SLA inside, and I have never seen any sign of corrosion. Only times I have seen a problem is in the case of UPS's that were left discharged in unheated warehouses... the electrolyte froze and the battery cases bulged greatly, but still no leakage. -Chuck Harris Dr. David Kirkby wrote: On 27 February 2014 17:31, Chuck Harris cfhar...@erols.com wrote: Backup batteries don't make me queasy. The easiest, cheapest backup battery would be a small 6V 3AH sealed lead acid battery. It is trivial to make a good float charger using a 3 terminal regulator and a diode, and they do not leak, vent, or anything problematic. The little 6V 3AH packs are used in emergency lighting, so they are readily available. Float them at 6.9V, and they will last at least 5 years. I thought there was an issue with even sealed lead acids of them emitting small amounts of sulfuric acid, which is not great in a bit of test equipment. Dave ___ 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] 5370 processor boards available
But we really are drifting away from the main thread. Yes, indeed. It's time to wrap up the operating system and file system tangents and get back to time frequency. Those of you with positive suggestions for improvements of the 5370 processor mod kit can email them to John Seamons directly. He has done an incredible job and the direction this thread has taken is unfortunate. /tvb ___ 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] HP53131A Fan/Blower
Has anyone found a suitable replacement or a source for the fan/blower in the HP53131A ? I have talked to Agilent and the fan/blower is not a seperate item, its part of the power supply board. TRY FREE IM TOOLPACK at http://www.imtoolpack.com/default.aspx?rc=if5 Capture screenshots, upload images, edit and send them to your friends through IMs, post on Twitter®, Facebook®, MySpace™, LinkedIn® – FAST! ___ 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] 5370 processor boards available
Now that you mention it, I have another small embedded system with 64MB of RAM that runs Linux 3.4 just fine off a flash drive, so running a RAM disk in 512MB should be feasible. Didier KO4BB On February 27, 2014 9:12:45 AM CST, Jim Lux jim...@earthlink.net wrote: On 2/27/14 6:40 AM, Didier Juges wrote: The BBB has 2GB of flash on board (non removable) and has a micro SD socket. Would not be too hard to keep a backup copy of the OS and apps on the SD card so that it would be easy to boot from SD and reload the built-in flash if the BBB fails to boot from the built-in image. Alternately, running the flash read only, copy OS, apps and everything else to a RAM disk and run from the RAM disk would probably be the safest. Not sure you can do that with 512MB of RAM without some serious pruning of the Linux kernel. Back in 2004, I was running a stripped down linux off a small compact flash (256MB or 512MB, I think) on a VIA Eden 700 MHz motherboard with very limited RAM. It included busybox to provide the usual command line utilities. As I dimly recall, it booted from a compressed filesystem (on CF) to ramdisk only. I also had a version of Debian that net-booted, and also ran entirely in ram. I could probably dig it up if someone's interested. We run RTEMS (www.rtems.org) at work, which provides a Posix compliant interface, but is designed for embedded systems, and has a bunch of targets (ours is SPARC, but there's x86 versions, for sure). You can definitely fit that in well under 128 MB. We've found that most everything that compiles and runs under Linux will compile/run under RTEMS, unless you're using something peculiar to Linux, or you need dynamic loading/linking (RTEMS is statically linked). That is, we have very few #ifdef LINUX kind of things in our code. ___ 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. -- Sent from my Motorola Droid Razr 4G LTE wireless tracker while I do other things. ___ 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] 5370 processor boards available
The only time there is any exposure is during a write operation That is not true. By default, Linux updates the last access time for each file it reads. The last access time is stored with the file, so each file read actually causes the file to be written to as well, opening the door to all sort of mayhem is power is lost during boot. Aside the performance penalty, you can trash your drive while you were just reading it. One easy optimization for embedded Linux is to turn that off. Google something like turn off last access time Didier KO4BB On February 27, 2014 9:06:31 AM CST, Brian Lloyd br...@lloyd.com wrote: The only time there is any exposure is during a write operation. When the processor board is used to run the 5370, how often is data written and what is the exposure interval? I would be willing to bet that Linux already has a power-fail NMI input. I would bet that you can find out what the worst-case PF NMI latency is and then ensure that the PS output stays up at least that long without having to worry about a battery or super cap. (I bet PF latency is under 1ms.) In that case, all you need to do is assert the PF interrupt line. Hmm, check to see if it is already there and based on nothing more than sagging Vcc/Vdd. I bet it is and therefore this whole discussion may be moot. But the idea of making the boot partition RO and the /var partition R/W makes a whole lot of sense to me. -- Brian Lloyd, WB6RQN/J79BPL 706 Flightline Drive Spring Branch, TX 78070 br...@lloyd.com +1.916.877.5067 ___ 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. -- Sent from my Motorola Droid Razr 4G LTE wireless tracker while I do other things. ___ 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] 5370 processor boards available
And after all that, *STILL* no one has been able to answer the question, Is there a problem that must be solved? Oh, lots of supposition, rules of thumb, boatloads of experience, etc., but still no determination that something really needs to be solved. One of the interesting things that came from building the internet protocol suite was a whole bunch of, well, let's just try it and see, that resulted in a whole LOT of, Oh, wow, it turns out it's just not a problem because it just doesn't happen. I would be willing to lay a fiver that this is one of those situations. -- Brian Lloyd, WB6RQN/J79BPL 706 Flightline Drive Spring Branch, TX 78070 br...@lloyd.com +1.916.877.5067 ___ 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] 5370 processor boards available
Looks like I win the fiver. Johns created a great board for the 5370. However you can't just turn the 5370 off as this lazy person is used to. Plus I really have to say after a full day of time-nuttery I won't remember to shut the linux down. So thats the need good old shutdown controlled by the power off button. But as Tom says please send the thoughts to John. Regards Paul WB8TSL On Thu, Feb 27, 2014 at 8:55 PM, Brian Lloyd br...@lloyd.com wrote: And after all that, *STILL* no one has been able to answer the question, Is there a problem that must be solved? Oh, lots of supposition, rules of thumb, boatloads of experience, etc., but still no determination that something really needs to be solved. One of the interesting things that came from building the internet protocol suite was a whole bunch of, well, let's just try it and see, that resulted in a whole LOT of, Oh, wow, it turns out it's just not a problem because it just doesn't happen. I would be willing to lay a fiver that this is one of those situations. -- Brian Lloyd, WB6RQN/J79BPL 706 Flightline Drive Spring Branch, TX 78070 br...@lloyd.com +1.916.877.5067 ___ 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] 5370 processor boards available
Hi I realize this is a bit off the main topic of … but here goes: Is there any performance data on how the card does with a 5370A and / or a 5370B compared to the original CPU on the exact same box? Put another way - does the counter get better or worse with the new card? I realize that an A will do some things with B firmware, that’s not the question I’m asking. I’m looking for A to A or B to B timing data. Bob On Feb 27, 2014, at 10:03 PM, paul swed paulsw...@gmail.com wrote: Looks like I win the fiver. Johns created a great board for the 5370. However you can't just turn the 5370 off as this lazy person is used to. Plus I really have to say after a full day of time-nuttery I won't remember to shut the linux down. So thats the need good old shutdown controlled by the power off button. But as Tom says please send the thoughts to John. Regards Paul WB8TSL On Thu, Feb 27, 2014 at 8:55 PM, Brian Lloyd br...@lloyd.com wrote: And after all that, *STILL* no one has been able to answer the question, Is there a problem that must be solved? Oh, lots of supposition, rules of thumb, boatloads of experience, etc., but still no determination that something really needs to be solved. One of the interesting things that came from building the internet protocol suite was a whole bunch of, well, let's just try it and see, that resulted in a whole LOT of, Oh, wow, it turns out it's just not a problem because it just doesn't happen. I would be willing to lay a fiver that this is one of those situations. -- Brian Lloyd, WB6RQN/J79BPL 706 Flightline Drive Spring Branch, TX 78070 br...@lloyd.com +1.916.877.5067 ___ 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] 5370 processor boards available
On Feb 28, 2014, at 4:34 PM, Bob Camp li...@rtty.us wrote: Is there any performance data on how the card does with a 5370A and / or a 5370B compared to the original CPU on the exact same box? Put another way - does the counter get better or worse with the new card? I realize that an A will do some things with B firmware, that’s not the question I’m asking. I’m looking for A to A or B to B timing data. Yes. I made some measurements with the previous microcontroller-based board (48 MHz sam7x) versus the stock CPU card. I have not yet had a chance to repeat that exercise with the current board. Look at the 23-Jul-2011 entry on www.jks.com/5370/5370.html Very much dependent on what the measurement is, but a range of 5-72% faster. ___ 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] 5370 processor boards available
On Thu, Feb 27, 2014 at 9:03 PM, paul swed paulsw...@gmail.com wrote: Looks like I win the fiver. Really? You power-failed it and corrupted the file system? Johns created a great board for the 5370. However you can't just turn the 5370 off as this lazy person is used to. Really? You tried it and screwed up the file system? Plus I really have to say after a full day of time-nuttery I won't remember to shut the linux down. So thats the need good old shutdown controlled by the power off button. But as Tom says please send the thoughts to John. Regards Paul WB8TSL -- Brian Lloyd, WB6RQN/J79BPL 706 Flightline Drive Spring Branch, TX 78070 br...@lloyd.com +1.916.877.5067 ___ 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] 5370 processor boards available
Le 28 févr. 2014 à 02:55, Brian Lloyd a écrit : And after all that, *STILL* no one has been able to answer the question, Is there a problem that must be solved? Oh, lots of supposition, rules of thumb, boatloads of experience, etc., but still no determination that something really needs to be solved. It depends on what level of security is required. I don't have one of these boxes, but it is legitimate to identify possible weak points. Volatile file systems are one of those. The original box had no issues on that front. Embedded systems such as linux with flash memory do. I have just had a Soekris FreeBSD NTP server go feet up due to file system corruption caused by a power failure. It is no great deal to create a new system, but I hadn't made a drop duplicate flash card so it will take some time. So one NTP server is down for a day. If a proud owner of one of these great boards doesn't care about that sort of issue, then OK, else one of the suggestions imposes. ___ 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] 5370 processor boards available
In message 7984e000-057c-4790-9d20-e4dac1f60...@rtty.us, Bob Camp writes: Is there any performance data on how the card does with a 5370A and / or a 5370B compared to the original CPU on the exact same box? Put another way - does the counter get better or worse with the new card? I realize that an A will do some things with B firmware, that=92s not the question I'm asking. I'm looking for A to A or B to B timing data. I have spent most of my time trying to answer exactly that question and I have not been able to devise any experiment that shows a difference in noiselevels with a credible statistical uncertainty. Interestingly, it is pretty evident from my experiments that the phase-noise of whatever EXT CLK source I use is the main cause of one-shot noise, so if anybody happens to have a *really* clean 10MHz and a 5370, it would be interesting to hear how low it can go. -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@freebsd.org | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence. ___ 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.