A good approach to carrier recovery would be the following chain - 1 AGC or limiter amplifier - I would use AGC followed by a limiter to keep the input to the doubler constant. 2 Analog Frequency Doubler with a bit of Q in the tank tuned to 2X carrier frequency Digital when it screws up does so with bad manners. The tank provides some frequency memory - not obtainable with a digital doubler. 3. A slow PLL that locks onto and tracks the 2X carrier frequency. I would use a counted down VCXO for this to ensure a long coast time. Loop bandwidth should be of the order of seconds or more and have a damping factor of 1 or more. the classic 0.707 damping tracks fast but overshoots and hunts before settling down. Fast tracking is not the requirement here. Stable control of the PLL is. 4. A 2x divider stage to output the original carrier frequency. This could be a flip flop as the amplitude output of the PLL would be constant.
I have made a few PSK and BPSk carrier recover circuits of this type for both micrwave spread spectrum links (20 GHz and UP and also for use with FSK at VHF). If the carrier is stable and there is no Doppler a very slow Pll does a world of good. See Gardner "Phase Lock Loop Techniques". the whole thing could be built on little more than 8 square inches of PWB and I am being generous. MSK is alive and well in a couple of FAA procurements, both past and present. The appeal is the constant envelope and reduction of adjacent sidebands. John Roos K6IQL -----Original Message----- From: time-nuts-request <[email protected]> To: time-nuts <[email protected]> Sent: Sun, Aug 17, 2014 8:21 am Subject: time-nuts Digest, Vol 121, Issue 45 Send time-nuts mailing list submissions to [email protected] To subscribe or unsubscribe via the World Wide Web, visit https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts or, via email, send a message with subject or body 'help' to [email protected] You can reach the person managing the list at [email protected] When replying, please edit your Subject line so it is more specific than "Re: Contents of time-nuts digest..." Today's Topics: 1. Re: Cutler NAA on 24.0kHz.... (Kenneth G. Gordon) 2. Re: Cutler NAA on 24.0kHz.... (Kenneth G. Gordon) 3. Re: Cutler NAA on 24.0kHz.... (Kenneth G. Gordon) 4. Re: Cutler NAA on 24.0kHz.... (Kenneth G. Gordon) 5. Re: Cutler NAA on 24.0kHz.... (paul swed) 6. Re: Cutler NAA on 24.0kHz.... (Kenneth G. Gordon) 7. MIT Flea (bownes) 8. Proper way to manually connect Vfc (Ole Petter Ronningen) ---------------------------------------------------------------------- Message: 1 Date: Sat, 16 Aug 2014 09:20:34 -0700 From: "Kenneth G. Gordon" <[email protected]> To: Discussion of precise time and frequency measurement <[email protected]> Subject: Re: [time-nuts] Cutler NAA on 24.0kHz.... Message-ID: <[email protected]> Content-Type: text/plain; charset=ISO-8859-1 On 16 Aug 2014 at 11:33, Bob Camp wrote: > Hi > > I would be *very* surprised if the NAA antenna was 50% efficient (transmitter RF > to radiated power)..... According to this: http://www.scribd.com/doc/145116437/THE-BIGGEST-LITTLE-ANTENNA-IN -THE-WORLD-US-Navy-s-VLF-antenna-at-Cutler-Maine The company which designed and built the dual trideco antenna system at Cutler had to guarantee >50% radiation efficiency, and they achieved an antenna radiation efficiency of 74.9% when using the 6 panel trideco. When I read this, I was truly amazed. Although, this site: http://www.navy-radio.com/commsta/cutler.htm does say that with 2 MW input, the ERP is 1 MW, which would indicate at least 50% radiation efficiency. I am still amazed. > Given that it?s already up and running with good signal levels, that?s not a big > deal. Very true indeed. Ken W7EKB ------------------------------ Message: 2 Date: Sat, 16 Aug 2014 09:26:40 -0700 From: "Kenneth G. Gordon" <[email protected]> To: paul swed <[email protected]> Cc: [email protected] Subject: Re: [time-nuts] Cutler NAA on 24.0kHz.... Message-ID: <[email protected]> Content-Type: text/plain; charset=US-ASCII On 16 Aug 2014 at 10:24, paul swed wrote: > Ken > At least last night NAA was running just fine using a fluke 207 and 4 ft of > wire. The antenna is behind a metal rack that shields it in NAAs direction. Ha! At VLF you could probably bury your antenna in a grounded, steel pipe 4 feet into the ground and still hear NAA. > I > did that test out of curiosity. I LOVE curious... :-) > Granted its 2 MW but then again the antenna is at best 50% efficient. See previous post. > Who knows maybe they have sections of the antenna down for maintenance. The reason NAA has a double trideco is so they can continue to transmit with one section down for maintenance. It turns out that Cutler has a much-bigger-than-usual problem with lightning... Ken W7EKB ------------------------------ Message: 3 Date: Sat, 16 Aug 2014 09:44:49 -0700 From: "Kenneth G. Gordon" <[email protected]> To: paul swed <[email protected]> Cc: [email protected] Subject: Re: [time-nuts] Cutler NAA on 24.0kHz.... Message-ID: <[email protected]> Content-Type: text/plain; charset=US-ASCII On 16 Aug 2014 at 12:31, paul swed wrote: > I wonder why? And I complain about my antennas. > It turns out that Cutler has a much-bigger-than-usual problem with > lightning... Well, apparently, first of all, Cutler is situated in what turns out to be an especially lightning-prone area of the country. Secondly, apparently the very high RF field "encourages" ligntning strikes to the antenna system. I read about this in a paper concerning the insulators used on that, and other, Navy antennas. The MTBF of failure for those insulators is phenomenal: something like 1700 years at worst-case. I'll try to find that paper again and post the link for those interested. Kenneth G. Gordon W7EKB "Courage is being scared to death but saddling up anyway."--- John Wayne ------------------------------ Message: 4 Date: Sat, 16 Aug 2014 09:58:34 -0700 From: "Kenneth G. Gordon" <[email protected]> To: paul swed <[email protected]> Cc: [email protected] Subject: Re: [time-nuts] Cutler NAA on 24.0kHz.... Message-ID: <[email protected]> Content-Type: text/plain; charset=US-ASCII On 16 Aug 2014 at 12:48, paul swed wrote: > > As a test this morning I connected the fluke 207 vlf radio that has a 1.1KHz IF > out to a XY scope using a very stable 1.1 Khz synth function gen. Indeed the msk > is a classic 4 corner eye pattern. Kewel. > Also I looked at the tracor 900 de-MSK-r > option and they split the path and create a new signal that according to > calculations would be 180 degrees and then sum it with the original. But from my VERY preliminary glance at that schematic, it looks as though the signal passes through two op amps. Doesn't the phase through any amp change by 180 degrees? Therefore, through two op amps, it would swap a full 360, which would equal zero at the output?. > That is > perplexing as that would be 0. They do pass it through a 100 Hz BPF. I wonder if > in reality the phase shift is 90 degrees and then summed. Its not really making > sense as to how that gets rid of the MSK. No description either. But it does > some how because that signal is simply re-inserted into the 100Hz IF that feeds > the phase detectors and correction/locking circuits. Lets see if the diagram of > the tracor d-msk-r will come through It did. Thanks. Ken W7EKB ------------------------------ Message: 5 Date: Sat, 16 Aug 2014 13:35:52 -0400 From: paul swed <[email protected]> To: Kenneth G Gordon <[email protected]>, Discussion of precise time and frequency measurement <[email protected]> Subject: Re: [time-nuts] Cutler NAA on 24.0kHz.... Message-ID: <cad2jfaj8y4jxrzedpi5cfn0cj3--n8ohy-ombxbrxuffjom...@mail.gmail.com> Content-Type: text/plain; charset=UTF-8 Kenneth on the opamps that is correct. But I put little U's to indicate phase. They actually represent the top half of the input cycle. In the top path it inverts once The bottom path twice. So that makes the top 180 out and the bottom in phase with the original. However the 2 X RC sets the bottom path at I believe 180 degrees from the input. The final RC in the top and bottom path account for opamp filter delay and note they are equal. So thats has me scratching my head as to how this removes the MSK and leaves a carrier that can lock. One of the classic approaches to recover carrier or get rid of BPSK modulation is to simply double the incoming carrier. Works great if you don't loose the signal. But I do not see this circuit doing that. Regards Paul On Sat, Aug 16, 2014 at 12:20 PM, Kenneth G. Gordon < [email protected]> wrote: > On 16 Aug 2014 at 11:33, Bob Camp wrote: > > > Hi > > > > I would be *very* surprised if the NAA antenna was 50% efficient > (transmitter RF > > to radiated power)..... > > According to this: > > http://www.scribd.com/doc/145116437/THE-BIGGEST-LITTLE-ANTENNA-IN > -THE-WORLD-US-Navy-s-VLF-antenna-at-Cutler-Maine > > The company which designed and built the dual trideco antenna system at > Cutler had to guarantee >50% radiation efficiency, and they achieved an > antenna radiation efficiency of 74.9% when using the 6 panel trideco. > > When I read this, I was truly amazed. > > Although, this site: > > http://www.navy-radio.com/commsta/cutler.htm > > does say that with 2 MW input, the ERP is 1 MW, which would indicate at > least 50% radiation efficiency. > > I am still amazed. > > > Given that it?s already up and running with good signal levels, that?s > not a big > > deal. > > Very true indeed. > > Ken W7EKB > _______________________________________________ > time-nuts mailing list -- [email protected] > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > ------------------------------ Message: 6 Date: Sat, 16 Aug 2014 11:56:35 -0700 From: "Kenneth G. Gordon" <[email protected]> To: paul swed <[email protected]> Cc: [email protected] Subject: Re: [time-nuts] Cutler NAA on 24.0kHz.... Message-ID: <[email protected]> Content-Type: text/plain; charset=US-ASCII On 16 Aug 2014 at 13:35, paul swed wrote: > > Kenneth on the opamps that is correct. > But I put little U's to indicate phase. They actually represent the top half of > the input cycle. Yes, I saw those, but unless I am mistaken, you didn't add a "U" after the second opamp, which would have returned the phase to the input's. > In the top path it inverts once. I see twice: once through the first op amp and again through the second one. The second one then outputs to the IF. Anyway, to me, it is a very interesting and simple circuit. I LIKE "simple". I am a great believer in the KISS principle. Ken W7EKB ------------------------------ Message: 7 Date: Sun, 17 Aug 2014 01:56:47 -0400 From: bownes <[email protected]> To: Discussion of precise time and frequency measurement <[email protected]> Subject: [time-nuts] MIT Flea Message-ID: <[email protected]> Content-Type: text/plain; charset=us-ascii Apologies for to those not in New England or not going to the Flea. I'm heading out to the MIT Flea at oh-freaking-dark in the morning and I know there are often several time nuts who go. If you happen to see a guy wander by in a Horton Emergency Vehicles hat looking very tired, say Hi! ------------------------------ Message: 8 Date: Sun, 17 Aug 2014 14:15:19 +0200 From: Ole Petter Ronningen <[email protected]> To: [email protected] Subject: [time-nuts] Proper way to manually connect Vfc Message-ID: <cac+ophh1j8yhy4n9zn4so+x_kepwqghdrzfuzwpg31ptikv...@mail.gmail.com> Content-Type: text/plain; charset=UTF-8 Hello, all I've started to build up a little collection of various OCXO's, and measuring various deviations and variances and whatnot. I'm putting them into proper enclosures, but up untill now I've adjusted them simply using the Vref through a reasonably low tempco (20-100ppm/C) multiturn pot, with just a .1uF cap on the Vfc pin to ground. Is this in fact the best way to do it? Or is there another (reasonably simple) way to improve on that setup? I presume the reference-voltage present on most OCXO's are "clean enough" to meet spec, but are there improvements to be had by using a separate low noise regulator, for instance? How about filtering, is there any reason to spend much effort on that, as the OCXO's are in separate enclosures, with coax soldered directly onto the 10Mhz output? Not sure where the noise would come from, but I stand to be corrected.. Sorry if this has been repeatedly answered (as I have a feeling it must have been), but I failed to find it in the archives (perhaps poorly chosen search-terms on my part) Thank you Ole P ------------------------------ _______________________________________________ time-nuts mailing list [email protected] https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts End of time-nuts Digest, Vol 121, Issue 45 ****************************************** _______________________________________________ time-nuts mailing list -- [email protected] To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
