Cheap GPSDO Buying Help 2024?

[NT0Z]

Hi, gang,

My "unusual and obscure" BG7TBL GPSDO variant (5-8 years old) has developed a heat-related intermittency issue, and if I can't fix it, I'm in the market for a cheap GPSDO. I am already in a bit over my head, so there's that...

Compared to many Super Metrologists, my needs are ridiculously minimal. I use the GPSDO to discipline a frequency counter that, to date, has never needed to count above 100 MHz. In fact, without a prescaler it CAN'T count above 100 MHz.   I simply need to corral that device to an accuracy level of about "1 Hz in 50 MHz," which isn't overly special. I also have a spectrum analyzer that has a 10-MHz ref input, so when I build a distribution amp, that thing can play along, too.

My now-ailing unit has two 10-MHz outputs, one sine, one square. When it was working I successfully connected it to my venerable 5328A counter using both outputs, but I'm not sure whether a sine wave reference is "better" than a square.

Does it matter to the general variety of home lab devices?

I ask because I have been seeing some inexpensive GPSDOs (and OCXOs) on eBay for $40-$80. They seem to put out square-wave signals at 10 MHz, so if I need sine -- or want sine -- that might be a problem.

I learned to use Lady Heather to check in on my now-ailing GPSDO, so I don't know whether these inexpensive units can "speak to Lady Heather." I don't use the GPS time-keeping function from the software, etc, I just use it to confirm that My unit is seeing a bunch of satellites and that the ref output is locked." For now, perhaps forever, the rest of the info is just noise.

Can these cheapie GPSDOs "indicate lock," etc, or would I be missing out on necessary functionality, even for my minimal needs, if I don't get a unit that can speak Heather?

Can anyone offer advice on which cheapie units they have had success with, etc?

I appreciate your input,

Kirk, NT0Z

P.S. I SHOULD just bite the bullet and buy a Bodnar GPSDO and be done with it, but I'm trying to buy other goodies, too, so saving money is key...

[EC8010]

I bought two GPSDO. The first was a cheapy 10MHz only from ebay; it works fine (for my needs, at any rate, which seem similar to yours).

https://www.ebay.co.uk/itm/175372737601?itmmeta=01HRN6BBY45ZGT320EXP7A8HTS&hash=item28d5079c41:g:OygAAOSwh6Vi6z5B&itmprp=enc%3AAQAIAAAA4M1Aia5X4rEdXaw7s6Vb30ruywauCAsV2eVjV0X9CH6pq2C9QE64e48y0%2BC3%2BJLICmp%2BfHtaJO7lbHcFXiHjnKbHi8QI4UP41ozQfjTxw9AY3JI%2BYnJ2NAQ36Kbo0HRSpMqw0%2BnV%2BT7vdIT50jjsGu6ppZIMYpQDjmvSGZ7HBO2KKfnvElld95N9jH5fWiNv0M2EiiopzTimat20yePubkWQ2ajUV5OemQ1nT0XevVR2j%2Bj0itTg1RzVIyKLDjbgQQe5aSwHUMuswz0HJxHDPp%2Fz7A6SgibR%2F8VTDCq5fdop%7Ctkp%3ABk9SR6a_rabFYw

Then I bought a Leo Bodnar that would do 10MHz and 96kHz on two separate outputs. A square wave output makes it easier for other equipment to lock accurately because most PLLs operate on a square wave internally. The cheapy GPSDO indicates when it has achieved lock (takes at least ten minutes), albeit not terribly intuitively. For best results, a GPSDO should be left permanently powered, so that is arguably unimportant. It is, after all, a crystal oscillator, and needs to settle. Something that did surprise me was that my oscilloscope initially refused to lock to the Leo Bodnar, yet locked happily to the cheapy. It turned out that the oscilloscope's Ref. In didn't terminate the cable and reflections were upsetting the square wave. A through termination solved that problem.

A distribution amplifier is the proper way to feed a number of destinations. But it's clear that some devices expect you to loop through. Like you, I have an RF oscillator that will lock to a 10MHz signal. Haven't yet tried it with a square wave.

No idea what Lady Heather is.

[GigaJoe]

for such application , I use  a well isolated OCXO in Styrofoam , and do adjust it against gpsdo once in a while ( usually twice per year , or less )
the cheap one  - https://www.ebay.com/itm/266588217507  (it AC CUT)
it constantly ON , no any movement or reposition,   (gravity affect oscillation freq)   the result after  6-8 month like:  9.99999999XX Mhz
sufficient for 8 digit counter .....

recently got this one - no clue how well it work
https://www.ebay.com/itm/333559659893

so maybe ocxo approach , with some periodic sync, will work for you as well  ...

[ddosegov]

Buildd one from here https://www.eevblog.com/forum/projects/budget-gpsdo-a-work-in-progress/

9A4BP

[GigaJoe]

for someone who not much familiar with  coding ..etc  . it a challenge

[trobbins]

for someone who not much familiar with  coding ..etc  . it a challenge

Which GPSDO is a challenge wrt coding?  I don't 'code', and prepared the GPSDO in post #8 link for about $30 worth of ebay purchasable parts.  You do have to install some free software, and use USB to RS232 like adaptors, so it helps if you aren't a PC or IT newb, but you have to do that anyway to monitor GPS connectivity with cheap GPS modules/antenna.

[vindoline]

If you are willing to DIY, this came out great:
https://www.eevblog.com/forum/projects/yet-another-diy-gpsdo-yes-another-one/msg5064550/#msg5064550

[NT0Z]

My GPSDO and my low-budget TCXO are always powered on.

Lady Heather is a PC program that monitors and controls a whole bunch of GPS/GPSDO/Freq Standard modules, mostly from 5-15 years ago (maybe newer?). It talks to my GPSDO via serial port to report on status, correction voltages, reference lock, etc. People way smarter than me wrote it and give it away for free. Yay!

I would be interested in building a GPSDO, but my time is limited and I'd rather spend it on the things that having a working GPSDO enables, like restoring vintage ham gear, building HF radios, etc.

If the supplied details were JUST A TITCH MORE COMPLETE I wouldn't hesitate, but I haven't programmed a PIC or an EPROM for 35 years, would prefer to work in an Arduino environment because I am a clueless digital newb, etc. My 45 years building circuits has all been analog/RF, so if I were to undertake the GPSDO build now it would take lots of time and resources, and chances are good that I'd end up, at least initially, with TWO non-working GPSDOs instead of just one (which I already have)   

A kit build would be much easier, but I'm not sure if any are available.

The DIY GPSDO referenced is a fantastic thing, but as is it's still a bridge too far to be practical for me. Unless there's a kit or semi-kit?

Regards,

Kirk, NT0Z

[GigaJoe]

right ....  sooo ...  hm ...

option 1 -
check this guy --- https://www.ebay.com/usr/macman1972a
the store has some  complete option based on DIY discussion ...  ( don't like price )

option 2 - ( my preference )
https://www.ebay.com/itm/133264349995
reclaimed board , no need any coding for a basic run ;  low jitter , quite precise , just need stable +5 2.5A
receiver not so sensitive and GPS only

[NT0Z]

Thanks for the specific links. Much appreciated!

--Kirk, NT0Z

[ddosegov]

Lady Heather is a PC program that monitors and controls a whole bunch of GPS/GPSDO/Freq Standard modules, mostly from 5-15 years ago (maybe newer?). It talks to my GPSDO via serial port to report on status, correction voltages, reference lock, etc. People way smarter than me wrote it and give it away for free. Yay!

Can it flash Trimble Lassen IQ GPS modules? Is there any firmware update for Trimble Lassen IQ? Few years ago there was a lot of very expensive  GPSDO/NTP equipment that had to be tossed away due to gps week rolover.

[NT0Z]

No idea about Lady Heather and flashing a Trimble module...but some folks on the "time nuts" list (google it) might. I think thats's where the software dev also hangs out.

Regards,

Kirk, NT0Z

[fourfathom]

If you want something that just works, not too expensive, and has reasonably good performance, it's hard to beat a Bodnar.  The hold-over during loss of GPS isn't as good as the fancy units, and there is some minor phase wander, but for my own needs (general measurement and frequency-stabilized transmitters and receivers used in ionospheric research) it's an excellent solution.

[Wallace Gasiewicz]

I built a GPS Disciplined OCXO out of stuff in my Junk Box. Including a Brick PS and an old GPS unit.  But I still had to buy some stuff.....     
It took quite a while and it works fine, the GPS signal disciplines the heater in the OCXO.   
After all the time and effort, I think I would have been better off buying a Bodnar....

[enut11]

My GPSDO and my low-budget TCXO are always powered on.

Lady Heather is a PC program that monitors and controls a whole bunch of GPS/GPSDO/Freq Standard modules, mostly from 5-15 years ago (maybe newer?). It talks to my GPSDO via serial port to report on status, correction voltages, reference lock, etc. People way smarter than me wrote it and give it away for free. Yay!

I would be interested in building a GPSDO, but my time is limited and I'd rather spend it on the things that having a working GPSDO enables, like restoring vintage ham gear, building HF radios, etc.

If the supplied details were JUST A TITCH MORE COMPLETE I wouldn't hesitate, but I haven't programmed a PIC or an EPROM for 35 years, would prefer to work in an Arduino environment because I am a clueless digital newb, etc. My 45 years building circuits has all been analog/RF, so if I were to undertake the GPSDO build now it would take lots of time and resources, and chances are good that I'd end up, at least initially, with TWO non-working GPSDOs instead of just one (which I already have)   

A kit build would be much easier, but I'm not sure if any are available.

The DIY GPSDO referenced is a fantastic thing, but as is it's still a bridge too far to be practical for me. Unless there's a kit or semi-kit?

Regards,

Kirk, NT0Z

I built this GPSDO from a partial kit and it works. No programming skills required.
https://www.eevblog.com/forum/projects/budget-gpsdo-a-work-in-progress/

[5U4GB]

option 1 -
check this guy --- https://www.ebay.com/usr/macman1972a
the store has some  complete option based on DIY discussion ...  ( don't like price )

option 2 - ( my preference )
https://www.ebay.com/itm/133264349995
reclaimed board , no need any coding for a basic run ;  low jitter , quite precise , just need stable +5 2.5A
receiver not so sensitive and GPS only

Nick Sayer on Tindie also has a bunch of GPS timing stuff, USB GPS timing module, full GPSDO, and an FE-5680A GPS discipline module which he's been updating continuously, just noticed that it's now quad-constellation.  He's also really good to deal with in terms of support, which you may not get from eBay sellers.

[NT0Z]

Nick appears to be taking a break until the middle of May.

--NT0Z

[NT0Z]

Okay. I did some poking, tapping, and freezing...

Lady Heather says the OCXO is GPS locked.

On pin 2 of the MAX4016 (inverting input 1) I have a 10-MHz signal that measures about 22 mV P-P with a 10X probe. So is that actually 220 mV? There is no output on pin 1, however (the corresponding op-amp output pin. This pin feeds the 74ACT244 chip, which feeds the output connectors via an SMD cap or two. There is "jittery stuff" on the output connectors, but that's it.

After applying freeze spray I can tap on either chip -- but mostly the larger 74244 -- and get short periods of BIG 10-MHz output.

In this area there are the two chips and a few SMD caps, but nothing underneath the PCB in that area.

There are two "larger" chip caps from the MAX4016 input pin that are paralleled to GND. I could solder a wire to that cap/rail and feed a MMIC amp to boost the signal, bypassing whatever is going on with the MAX4016 and the 74244?

I could apply a bit of liquid flux and heat the solder pads for the legs on the chips and the ends of the caps with my hot air tool in case one is intermittent?

If I need to replace either chip, should I start with the MAX4016? It has an input signal but no output. Vcc is 5-V dc, and Vee seems to be GND.

Replacing these parts isn't rocket science, but it's new to me, and my eyesight isn't helping...

Ideas?

Thank you,

Kirk, NT0Z

SORRY -- POSTED THIS TO WRONG MESSAGE, BUT YA NEVER KNOW!

[ddosegov]

Ideas?

Nice hi-res picture or schematic would be helpful, but amp giving nothing on output despite there is signal on input looks like troublemaker.

[SCSKITS]

When looking at a GPSDO, look for plots of frequency difference over time and Allan deviation at the frequency of interest.
In other words basic TimeLab measurements.

Look at the OCXO used. The larger double oven OCXOs are best but have a higher cost and require more power. The the mid size units (about 1.5" by 1") are single oven (a crystal heater), usually without insulation, but are good enough for most use in a GPSDO. The small (about 1" square) <$5 OCXOs are the worst of the lot but can still be disciplined by the GPS.

There are some GPSDO articles out there where they use measurement of the control voltage converted to frequency as their determination of stability. The DMM used can hide some variations as most meters have a long integration time that can smooth out some of the noise. This method also does not account for variation internal to the OCXO.

If Phase noise is important for your application, look for that also. Unfortunately I currently lack the equipment for phase noise measurement, working on it.

I recently made a couple of GPSDO boards using UBLOX M8 and M9 GPS receivers, an EPLD for the dividers and phase detector, an analog PLL filter, and a 5V OCXO.
The phase detector output to the filter is buffered by a CMOS gate powered from a precision reference. Both square wave and sine wave outputs are provided.
The output frequency difference is at the limit of the test equipment I have. Frequency difference and Allan Deviation plots are in the listing.
The GPSDO was tested using an Agilent 53210A frequency counter with an HP Z3801 GPSDO external reference.

Listing is here: https://www.ebay.com/itm/156042688976.

ed

[jjoonathan]

I'm in this market as well, looking to spend maybe a few hundred bucks but not much time, I'm just looking for something that works. Based on the many recommendations I'm sorta leaning towards the Bodnar GPSDO, but I have one big hesitation: does it really not have NMEA output? That's... sort of a big deal for me. I have major trust issues with "PLL Locked" LEDs. Maybe I've just spent too much time digging around in old broken equipment, but I've just seen far too many PLLs give an "all good  " when all was not in fact good. Seeing a bunch of satellite info and correct computation of time & location makes me much more comfortable. I wouldn't expect this to be a rare viewpoint, but it sure seems to be... does everyone here have 5 GPSDOs and equipment to compare them so missing NMEA on one of them is no big deal? Or does nobody else have trouble trusting the LED? Why is it so abnormal for me to want to see NMEA data?

[fourfathom]

I'm in this market as well, looking to spend maybe a few hundred bucks but not much time, I'm just looking for something that works. Based on the many recommendations I'm sorta leaning towards the Bodnar GPSDO, but I have one big hesitation: does it really not have NMEA output? That's... sort of a big deal for me. I have major trust issues with "PLL Locked" LEDs. Maybe I've just spent too much time digging around in old broken equipment, but I've just seen far too many PLLs give an "all good  " when all was not in fact good. Seeing a bunch of satellite info and correct computation of time & location makes me much more comfortable. I wouldn't expect this to be a rare viewpoint, but it sure seems to be... does everyone here have 5 GPSDOs and equipment to compare them so missing NMEA on one of them is no big deal? Or does nobody else have trouble trusting the LED? Why is it so abnormal for me to want to see NMEA data?

No, the Bodnar does not deliver NMEA data, and no, you are not alone on wishing it did.  In one of my applications I use a cheap GPS chip/module on my board to get NMEA as well as 1pps, and an external Bodnar for the reference clock.  I will eventually try to combine these functions, but for now using the Bodnar saves me the time and effort of implementing a GPDSO.

Is anyone aware of an inexpensive GPS chip that simultaneously does NMEA, pps, and high-frequency clock for a GPSDO reference?  This doesn't have to be time-nut quality, just good enough.

[PCB.Wiz]

Is anyone aware of an inexpensive GPS chip that simultaneously does NMEA, pps, and high-frequency clock for a GPSDO reference?  This doesn't have to be time-nut quality, just good enough.

Many allow the 1pps to change even up to some MHz on a NCO basis, but I've never seen one with two output pins.

[SCSKITS]

The LEA-M8F (if I recall correctly) has a 30.72MHz output and a programmable output.
Some of the others may also have two programmable outputs.
The 8, 9, and 10 series parts can save the settings in FLASH, the 5, 6, and 7 series need a battery to backup the settings.
The parts with USB are easy to configure using the UBLOX configuration program. I would assume the serial port on the parts is similar.

The BODNAR GPSDO (with two BNC outputs) is a UBLOX MAX-M8Q module and a SI5328C jitter correction chip controlled by a PIC with USB.
The SI5328C provides the two outputs, but the second output must be integer related to the first.
Of all the SI chips, the SI5328C has the lowest frequency loop filter for the PLL.

ed

[fourfathom]

Is anyone aware of an inexpensive GPS chip that simultaneously does NMEA, pps, and high-frequency clock for a GPSDO reference?  This doesn't have to be time-nut quality, just good enough.

Many allow the 1pps to change even up to some MHz on a NCO basis, but I've never seen one with two output pins.

Yes, I'm involved peripherally with some designs that run the clock out pin way up in frequency (to drive a GPSDO), but in addition to the frequency reference I also need the 1pps time reference.  Or *some* absolute time reference that's more precise than the arrival of an NMEA sentence.  Of course I also need lat/lon/day/hour/min/sec/status.  I can get this from my old thunderbolt box, but I want it on an inexpensive module I can have JLCPCB put on a board for me.  I'm afraid I will end up with two chips, one for time and the other set up for frequency.