[time-nuts] Low cost SDR suitable for phase noise measurement?
Bob Camp
kb8tq at n1k.org
Fri Aug 12 13:09:09 UTC 2016
Hi
There have been a lot of threads on DMTD’s here. There have not been quite as many
on single mixer setups. The same basic principles apply. I would be careful about duplicating
the original NIST DMTD from the first paper. I believe that “That’s what you get when you have
a statistics guy do an RF project” is about as close to a direct quote as I can come up with after
nearly 40 years …. One key point - you need about 120 to 140 db of isolation between the mixers in
the DMTD. At that level, you have to measure it. A calculated number is likely to be wrong.
There are a few other issues.
Without a proper limiter, the DMTD or single mixer will actually be worse than your 5370 by a
significant margin. With a limiter, you will still need a computing counter to get it to work well.
The deal is fairly simple:
You are after 1x10^-13 accuracy
You start with 10 MHz
You go down to 10 Hz
Thus you get 1x10^6 from the heterodyne process. You still need 13-6 = 7 orders of magnitude.
Thus you need to measure the 10 Hz to about 7 places accurately. Resolution does not matter
much in this case, it’s accuracy you are after. Without a proper limiter, your 10 Hz will come out as
10.1 with the rest of the digits purely random.
Fixing that with an ADC only works if you have an ADC that is quiet below the audio range and
that is not blocked (AC coupled) at to high a frequency. You also need to write a bunch of custom
DSP software to get it running.
Lots of fun !!
Bob
> On Aug 12, 2016, at 1:29 AM, Bob Stewart <bob at evoria.net> wrote:
>
> Hi Bob,
> Thanks. Well, I guess the next thing to do is to get a USB sound card, and lay down some copper. I did manage to get some DBMs and power splitters from Mini-circuits, and I'll use the LPF from the NIST DMTD and see what I can come up with. Hmm, I'll have to look at the LPFs, won't I? No amps or ZCDs, of course. I think I'll start with a single mixer to see what that gets me. But I'll go ahead and layout the board for a DMTD with power splitter and two LPFs for the most versatility. If I can get something better out of it than the 5370, it does solve a lot of problems for me. The 8640B should be fine as a heterodyne oscillator. Also, I can experiment with a pair of FCC-1/FCC-2 DDS systems I have that Bob Okas designed and sold before he passed away, as well to see if they have any value for this application. I have choices for master reference oscillator, as well.
>
> Bob
> -----------------------------------------------------------------
> AE6RV.com
>
> GFS GPSDO list:
> groups.yahoo.com/neo/groups/GFS-GPSDOs/info
>
> From: Bob Camp <kb8tq at n1k.org>
> To: Discussion of precise time and frequency measurement <time-nuts at febo.com>
> Sent: Thursday, August 11, 2016 9:41 PM
> Subject: Re: [time-nuts] Low cost SDR suitable for phase noise measurement?
>
> Hi
>
> In a DMTD (or in the equally useful single mixer setup) the advantage for frequency measurement
> comes from the downconversion process. The further you downconvert (10 Hz vs 100 Hz) the
> greater your potential resolution. The design approach is to push the beat note as low as you
> can without running into other limits. With conventional setups that is in the 1 to 10 Hz range.
>
> If you want to fiddle a bit, there is no particular reason why you must use a limiter after the
> mixer. An ADC after the mixer could be used to process the audio and estimate it’s frequency.
> The same sort of ADC setup might give you some phase noise data.
>
> The gotcha with all this is that a single mixer / quadrature approach will give you *much* better
> phase noise data than a single or dual mixer setup. If phase noise is the objective, there is no
> reason to run the mixer setup as a DMTD. With a DMTD you digitize the carrier (audio beat note)
> and then try to dig down 195 db to get to the phase noise. With a quadrature approach you null out
> the carrier. Then all you have to deal with is the noise.
>
> Bob
>
>
>> On Aug 11, 2016, at 5:15 PM, Bob Stewart <bob at evoria.net> wrote:
>>
>> Hi Bob,
>> If I understand the question correctly, yes. The DMTDs that I've encountered so far use a zero-crossing detector and a TIC. I wanted to build sort of a hybrid system that has a DMTD front end and an SDR back end. It seemed to me that that would be sort of half way between a DMTD and a Timepod. But whereas the Timepod digitizes 4 RF signals, this would digitize 2 audio signals. It may be that it suffers all the disadvantages of both and has the advantages of neither, but I thought I'd build it and see where it leads.
>> As to the 1 to 10 Hz range, doesn't that depend on the heterodyne frequency you choose? I mean, you don't have to use a 10 Hz offset, do you?
>>
>> Bob
>> -----------------------------------------------------------------
>> AE6RV.com
>>
>> GFS GPSDO list:
>> groups.yahoo.com/neo/groups/GFS-GPSDOs/info
>>
>> From: Bob Camp <kb8tq at n1k.org>
>> To: Discussion of precise time and frequency measurement <time-nuts at febo.com>
>> Sent: Thursday, August 11, 2016 3:13 PM
>> Subject: Re: [time-nuts] Low cost SDR suitable for phase noise measurement?
>>
>> Hi
>>
>> Just to clarify:
>>
>> You are looking for a way to digitize the output(s) of the mixer(s) in a DMTD. The
>> “target signal” is in the 1 to 10 Hz range.
>>
>> Is that correct or are you looking for something to use instead of a DMTD?
>>
>> Bob
>>
>>
>>> On Aug 11, 2016, at 3:29 PM, Bob Stewart <bob at evoria.net> wrote:
>>>
>>> Hi Brooke,
>>> It looks like I've asked a question poorly, yet again. So, let me try again. I have this idea of combining a DMTD with an SDR. So, said that way, it looks like what I really need is a 2-channel audio SDR. And having said that, maybe what I'm looking for is an external USB sound card. Those I can find. I can probably even figure out how to convert them to DC and supply them with a disciplined oscillator, if needed.
>>>
>>> Bob
>>> -----------------------------------------------------------------
>>> AE6RV.com
>>>
>>> GFS GPSDO list:
>>> groups.yahoo.com/neo/groups/GFS-GPSDOs/info
>>>
>>> From: Brooke Clarke <brooke at pacific.net>
>>> To: Discussion of precise time and frequency measurement <time-nuts at febo.com>
>>> Sent: Thursday, August 11, 2016 1:36 PM
>>> Subject: Re: [time-nuts] Low cost SDR suitable for phase noise measurement?
>>>
>>> Hi Bob:
>>>
>>> The SDR-IQ has a lower frequency limit of 500 Hz.
>>> http://www.prc68.com/I/Bats.shtml#SDRIQ
>>>
>>> There is a mod that replaces the on board 66 MHz oscillator with one locked to 10 MHz.
>>>
>>> --
>>> Have Fun,
>>>
>>> Brooke Clarke, N6GCE
>>> http://www.PRC68.com
>>> http://www.end2partygovernment.com/2012Issues.html
>>> The lesser of evils is still evil.
>>>
>>> -------- Original Message --------
>>>> A number of posts have appeared about Zynq SDRs on the "Shera revisited" thread. So, I'd like to ask the more general question: Are there any low cost SDRs that are suitable for measuring phase noise when coupled with a DMTD? I believe what I'm looking for is something with enough bits to work in the audio range down to DC. Yes, I could use a modified sound card, but I'm really looking for something not sound card related. My thought was to use a DMTD with a heterodyne frequency of maybe 5KHz.
>>>>
>>>> Bob -----------------------------------------------------------------
>>>> AE6RV.com
>>>>
>>>> GFS GPSDO list:
>>>> groups.yahoo.com/neo/groups/GFS-GPSDOs/info
>>>> _______________________________________________
>>>> time-nuts mailing list -- time-nuts at 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 at 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 at 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 at 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 at 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 at 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 at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
More information about the Time-nuts_lists.febo.com
mailing list