Re: 122GHz transverter with external 10Mhz reference


 

I can see the thought behind it. The frequency can be several kHz off and the signal can drift several kHz (maybe more) and you can't tell the difference on FM. A 100 Hz drift on CW or SSB is annoying and a kHz drift is starting to get difficult to deal with.

But my experience with my Isotemp OCXO's free-running at 10 MHz has been fine up through 47 GHz. Frequency has been very accurate and stable.

I haven't used a GPS disciplined OCXO yet. I think I mentioned that, in a paper by OK1DXD published in the 2013 MUD Proceedings, he stated that 1 pps takes too long to lock in a GPS disciplined system, which is why he used 10 kHz as the lock frequency.

What is the difficulty with a GPS disciplined OCXO on CW supposed to be?

73, Zack W9SZ


On Thu, Apr 16, 2020 at 5:35 AM Ev Tupis via groups.io <w2ev=yahoo.com@groups.io> wrote:
Hi Mike,

Thank you for taking the time to lay this out as you have.

I'm always cautious when I hear opinion statements such as "this will/won't work" when supporting data is missing.  In this case, "GPSDO's are good enough for FM but not CW".  I'm trying to understand what that means quantitatively vs. what Tim has demonstrated in his videos when using a direct, external 10 MHz reference.

I'll take a look at the videos that you shared.  I'll follow-up with additional questions if I have them after that.

Again, I (and likely others) sincerely appreciate the time that you took to lay this out.

Kindly,
Ev, W2EV







On Wednesday, April 15, 2020, 4:18:51 PM EDT, Mike Lavelle K6ML <miclevel@...> wrote:





Ev ...

Not sure if when you say 'mentioned earlier in this thread' you were referring to the various GPS units or the various OCXOs that folks have mentioned (lots of posts on both; you included the OCXO list in your post).

As other posters mentioned, we are really talking about two different alternatives:

(1) the original design (kit as shipped) has an on board 10 MHz TCXO and firmware which accepts a 1 pps signal from an external GPS unit.  The firmware 'disciplines' the TCXO to follow the 1 pps signal.  There is a file (from VK3CV) in this group's wiki (https://groups.io/g/The122GProject/files/Hardware%20and%20Parts.pdf) that recommends a NEO-6 GPS (and also a tripod and a riflescope) that you may want to use.  Any GPS with a 1pps output should work (higher frequencies than 1 pps are not supported by the original firmware). Note that, for (1) you're not looking for a GPSDO, just a GPS with 1 pps output, as the board already provides the "DO" part.

(2) elsewhere in the wiki, you will find a mod that you can do to disable the on board TCXO that disregards the DO firmware and instead uses an external 10 MHz reference that you supply (https://groups.io/g/The122GProject/files/Hardware%20and%20Parts.pdf) in its place.  The simplest way to do that would be to use a 10 MHz OCXO like the ones in my list.

Tim has posted several youTube videos that are examples of (1) using various 1 pps GPS units (NEO6, NEO7, his lab test gear standard, etc).  They all show that the TCXO frequency drifts with temp and that method (1) cannot fully correct the drift... the drift is not a problem for FM, but is for CW.  The results don't vary much with the GPS unit.  I suspect the drift will be a bit less once the PCB is enclosed in a box that provides some insulation from thermal air currents, but it may not completely go away.  If you haven't you should listen to the videos.

FM / NEO 6 : https://www.youtube.com/watch?v=5ngaM_gfkYA
CW / NEO 6 : https://www.youtube.com/watch?v=GKfi10Mdc1g&t=6s
CW / other GPS's:
https://www.youtube.com/watch?v=9Fj641S70BM
https://www.youtube.com/watch?v=1BHaocTx6iI
https://www.youtube.com/watch?v=X4HHPSn1oyk


Tim doesn't have a video of a simple OCXO (method #2), but he does have two videos that connect an external GPSDO (instead of the simpler external OCXO).  The first (using a very good bench GPSDO which includes a good OCXO applied to external oscillator mod) is a good approximation of (method #2), just ignore the (very) occasional 'blip' sound:

https://www.youtube.com/watch?v=TiMGFrocmr8

On the other hand, this video shows what a poor external GPSDO can sound like (chirps every 1 sec):

https://www.youtube.com/watch?v=xLHYBbV-qCI&t=20s

But, again, method #2 in its simplest for is just a good external OCXO, not an external GPSDO, and it will not have the blips.

73 & HTH,

Mike


-

> On April 15, 2020 at 4:46 AM "Ev Tupis via groups.io" <w2ev=yahoo.com@groups.io> wrote:
>
>
> Guys, seriously.  What is the operational/functional difference between using a NEO-8M vs the units mentioned earlier in this thread, please?
>
>
>
>
>
>
> On Tuesday, April 14, 2020, 8:26:42 AM EDT, Ev Tupis via groups.io <w2ev=yahoo.com@groups.io> wrote:
>
>
>
>
>
> I need to ask a Greenhorn question, please.  Could someone help me (and probably others who haven't been brave enough to ask) about a comparison of the devices shown earlier in this tread to the one below?  It was mentioned earlier as a fine unit and is still clearly available (with enough time to receive it before final units are shipped to project participants).
>
> Ev, W2EV
>
>
> NEO-8M GPS Satellite Positioning Module for Arduino STM32 C51 replace NEO-7M | eBay
>
>
>
> $7.99
>  
> NEO-8M GPS Satellite Positioning Module for Arduino STM32 C51 replace NE...
> With a USB interface, you can watch the computer positioning effect directly phone line. with SMA interface can ... 
>
>
>
>
> I
>
>
>
>
>
>
> On Tuesday, April 14, 2020, 12:05:52 AM EDT, Zack Widup <w9sz.zack@...> wrote:
>
>
>
>
>
> Thanks Mike. The OCXO118-12 is also 5 ppB -30 to +60 C. Somewhere I also have a phase noise sheet for it, too. I don't remember exactly what it was, but I remember when I saw it I thought it had a really good phase noise.
>
> 73, Zack W9SZ
>
> On Mon, Apr 13, 2020 at 10:38 PM Mike Lavelle K6ML <miclevel@...> wrote:
> >  
> >  
> >  Hi, Zack...
> >
> >
> >  
> >
> >
> >  isotemp 134-10   5 ppB -30 to +60C
> >
> >
> >  MTI 250             10ppB   -30 to +70
> >
> >
> >  MTI260               0.2 ppB  -30 to +70
> >
> >
> >  Oscilloquartz  8863    1 ppB stb (0.2, 0.6 options)  -10 to +70
> >
> >
> >  CTS 196              5 ppB   -10 to +70
> >
> >
> >  
> >
> >
> >  I'm using CTS196's (there were bunches of them on ePay a while back) with good results.
> >
> >
> >  After a few minutes warmup, we always find each other within about 1 kHz at 122 GHz.
> >
> >
> >  
> >
> >
> >  mike
> >
> >
> >  
> >
> >
> >  
> >
> >
> >>  On April 13, 2020 at 1:47 PM Zack Widup <w9sz.zack@...> wrote:
> >>
> >>
> >>  
> >>  What is an example of a good double oven SC cut OCXO with several ppB temp stability?
> >>
> >>
> >>  
> >>
> >>
> >>  I have been using an ISOTEMP OCXO118-12 without GPS disciplining on lower bands. I actually have a few of them and calibrate them to a rubidium standard every month or so. So far their stability and accuracy has been adequate through 47 GHz. There is some noticeable (but acceptable) jitter on 47 GHz. I haven't been able to determine if the jitter is from the OCXO or from the phase-locked crystal oscillator I'm using to generate the LO. I'm about to try one of these out with my 76 GHz system I am finishing up on.
> >>
> >>
> >>  
> >>
> >>
> >>  73, Zack W9SZ
> >>
> >>
> >>  
> >>
> >>
> >>  
> >>
> >>
> >>  
> >>  On Mon, Apr 13, 2020 at 1:52 PM Mike Lavelle K6ML < miclevel@...> wrote:
> >>
> >>
> >>>  
> >>> [Edited Message Follows]
> >>> Interesting link, Barry.   Looks well done.
> >>>
> >>> One thing to remember: a GPSDO is not a magic cure for a cheap oscillator.   If the oscillator being disciplined has several ppm of temp drift (that's 100's of kHz at 122 GHz over full temp range) and pops and jumps (typical of an AT cut xtal), the GPS 1pps can only accomplish so much by disciplining once per sec.   That's what you heard on the original video of the on board tcxo that comes with this kit.   In addition, if the disciplining firmware does not filter the corrections, there can be sudden step changes with a 1pps tick (even if no leakthru).
> >>>
> >>> But, if you're gonna take the trouble to find a good double oven SC cut OCXO with several ppB temp stability instead of several ppm, the value add of the GPS disciplining it is less... I just net my surplus ppB OCXO to be on freq before leaving home and rely on its stability while out hill topping.   Signals are easy to find in a 1 kHz pass band.  (and if you have even a simple, narrow panafall display, you can easily find a weak signal a few kHz off, if need be).  I think GPS is only necessary when trying to work subaudible, barely visible on waterfall, digital mode QSOs.    
> >>>
> >>> Also, if you change your QTH (head to a hilltop), the GPS will need some time to adjust to the new position fix, and may even introduce an error for a while before settling down.
> >>>  
> >>>
> >>>  
> >>>
> >>
> >>
> >>
> >>
> >>   Virus-free.  www.avast.com  
> >>
> >
> >  
> >  
> >
> >
> >
> >
>
>
> Virus-free.  www.avast.com  
>
>
>
>
>
>
>
>
>






Virus-free. www.avast.com

Join The122GProject@groups.io to automatically receive all group messages.