10MHz GPSDO by "BH3SAP"

[mawyatt]

Without analysis, wide guess would be 3~4pF.

Best

[KungFuJosh]

I received another GPSDO from another seller. It looks like it doesn't matter who you order from. They're all getting the same junk stock with the bad wiring and Bowei OCXOs. 

[Kurt_!]

I didn't have to solder anything....

[KungFuJosh]

I didn't have to solder anything....

Did you check the wire under the blue pill?

[Kurt_!]

Yes, nothing soldered. Works very well with the latest update. Mine doesn't have any pins for RX or TX.

[KungFuJosh]

Yes, nothing soldered. Works very well with the latest update. Mine doesn't have any pins for RX or TX.

There's no red wire connecting the blue pill on the bottom side?

[fredo_]

Yes, nothing soldered. Works very well with the latest update. Mine doesn't have any pins for RX or TX.

There's no red wire connecting the blue pill on the bottom side?

Don't you have something like that @Kurt_! ? https://github.com/fredzo/gpsdo-fw?tab=readme-ov-file#flickering-screen--impossible-to-get-the-gpsdo-to-lock

[Kurt_!]

Yes, I did, but I left it as it is. It works perfectly!

[fredo_]

Yes, I did, but I left it as it is. It works perfectly!

Okay !
What model is your OCXO ? It's an ISOTEMP ?

[Kurt_!]

What's his name...the other one

[ME]

My units time readout would freeze,tony albus has a fix,if its of use , It fixed mine.

[ME]

One more question:
Pins PA0 and PA1 each have a LED connected to ground with a resistor.....

PA0/PA1 as follows
@fredo_ (GitHUB)

"
Can those pins sink enough current to drive an led say 10milliamps?.
GPS Lock and GPSDO Lock Outputs
Pins PA0 and PA1 can be used to drive leds showing GPS Lock and GPSDO Lock status.

PA0 will be pulled low as soon as the GPS module is locked and PA1 will be pulled low when PPB mean value goes above the set threshold.
"

I put a driver PCB here so as not to overload the MCU output. like: pict1

[KungFuJosh]

My units time readout would freeze,tony albus has a fix,if its of use , It fixed mine.

Yup, that's important especially for those units without the ISOTEMP OCXO.

You should read more of this thread- there's a better custom firmware available from fredo_.

[ME]

mine had this issue as well,after moving the wire its fine and is doing service as my bedside clock.

[KungFuJosh]

mine had this issue as well,after moving the wire its fine and is doing service as my bedside clock.

Which OCXO do you have?

[ME]

mine has the isotemp but i brought it nearly a year ago.

[KungFuJosh]

Nice. I bought a few ISOTEMP OCXOs to put in the newer ones with the junk Bowei OCXO. When I have some time, I have a fresh GPSDO I'm going to setup with an ISOTEMP OCXO, and the trimmers I mentioned previously to tune the output waveform. I'm hoping I can improve it over my other unit.

[Kurt_!]

Hello!
So, the device is running very well for me now!
With the firmware (0.16) and OX256B.
I didn't have to re-solder anything.
I'm very happy.
Thanks to the contributor.

I think once the device runs for a day, the frequency will be very stable, and there's no need to buy another overpriced device for the hobby!

I hope the translator translated everything correctly?

[KungFuJosh]

The wiring issue is interesting. The difference between the two pins is the OCXO Input and Output on the Blue Pill's MCU, which has a little 16MHz crystal connected there already.

Pin 5 is the Oscillator In, and Pin 6 is Oscillator Out on the MCU. So, what's actually happening when people think it works with the 10MHz OCXO connected to Pin 6? Is the 10MHz OCXO passing signal backwards through the 16MHz crystal? Is that even possible? Or is the 10MHz OCXO not actually connected?

Thanks,
Josh

[mawyatt]

With OCXO signal connected thru the RC network to the Blue Pill Osc output, this signal is fighting with the Osc output. Usually this OCXO signal wins and things work as expected, however it's possible for this to hiccup every once in awhile and cause a glitch as was originally discovered which led to the finding of the incorrectly wired crystal wire from the OCXO RC network. With the crystal wire connected to the Blue Pill Osc input, then this completely overrides the crystal control and the Osc assumes the OCXO signal without any hiccups!!

Best 

[KungFuJosh]

I figured it was something like that. Sounds like people should correct that wire even if it appears to work when wired incorrectly.

[Kurt_!]

I re-soldered it. It works!
I'll keep testing it and if I notice a difference, I'll report back....

Ps.: I had already tried it like I did with Tony and then soldered it back...

[KungFuJosh]

I bought a few ISOTEMPs to replace the 2nd Bowei. One of them is slightly better on the duty cycle, but they're all basically the same, which I guess is good. The ringing is obviously the cables, which is fun to screw with by adjusting the cables.

[Kurt_!]

I have to test it first. 5 cti + 2 vectron...
I'm currently testing Neo 6, 8, 9, and 10max modules...

PS: and software

und ein ATGM336 laufen alle auch mit 115200 Baud

[KungFuJosh]

Messing with another one of these.

280 is the Bowei in the unmodified stock GPSDO circuit.

281 is the Bowei out of circuit.

282 is an ISOTEMP out of circuit that I'll be replacing the Bowei with.

Thanks,
Josh