I purchased a SparkPNT GNSS Disciplined Oscillator in May 2025. The 10 MHz output appears to no longer be disciplined by the GNSS.
When I first got the unit, the 10 MHz output was showing good frequency stability. The Allan deviation was 3e-10 @ 10 seconds and 1e-12 at 1e3 seconds. In June, we had an issue with the unit’s antenna connector, but after fixing the connector, we found that the 10MHz had became undisciplined, with an Allan deviation staying flat around 1e-11 from 1 second out to 3e3 seconds, and then drifting up at longer times.
The Mosaic-T says reports that it is operating normally, and it is getting position data from almost 20 satellites. I measured the stability of the Mosaic-T’s 1 pps output, and got good results that are similar to another Mosaic-T that I have access to (i.e., 1e-9 at 10 seconds and 2e-11 at 1e3 seconds).
Is there a software switch that controls the disciplining of the 10 MHz? or is there something else that I’m not doing right?
Can you please tell us how the antenna connector was broken? Did you drop the unit while the antenna cable was attached?
Can you please tell us what values you see for “Bias” on the OLED display? Are the values sub 1 nanosecond most of the time?
It would be helpful if you could: attach a USB cable to the CONFIG ESP32 port; install the CH340 drivers if needed; open a Serial Monitor or Terminal Emulator at 115200 baud; copy and paste some of the text that is output. The 11th column is the Bias. I would like to see how it changes over perhaps one minute. Please follow the link below if you need more information.
Have you made any changes to the PI (Proportional Integral) terms? (These are set through the serial menu. Details below)
The unit is no longer in its metal case. I am wondering if the temperature variations in the room are playing a part in this? The oscillator is the part shown in the second link below. Perhaps insulating it will help? If you have any anti-static foam, you could tape some over the oscillator and surrounding PCB.
The antenna connector broke off when the sma connector was being screwed on. The unit has never been dropped.
The bias on the OLED display always says “+0.000ns”.
Concerning connecting to the “CONFIG ESP32” port, I’ve installed the CH340 driver and followed the instructions in the manual many times, but the connection is always refused. I’ve used Tera Term and Putty as terminal programs. In my latest attempt to connect to the CONFIG ESP32 porth (which I tried just now), I received this message:
“Unable to open connection to COM11
Error 5: Access is denied”
I have not changed the PI terms.
I placed the cover back onto the case, but this has made no difference. Also, the unit is in a temperature stabilized room.
I have no idea how, but somehow I managed to connect to the “CONFIG ESP32” port. Once connected, I typed the “Reset all settings to default” command, and almost immediately the unit started working. For example, the Error light went out and the Lock light come on. The OLED display started to display a reasonable IP address, and a reasonable latitude and longitude. The bias is usually below 1ns.
I am measuring the 10MHz stability right now. I will be off sight tomorrow (Wednesday), but on Thursday I’ll check the stability and let you know what it looks like.
