"CALIBRATE above 5.34GHz" throws a "Signal level too low or not on frequency" error #bug_report
¿ªÔÆÌåÓýHi Joe You already got an answer from Eric, as the frequency must be 5.34GHz within 1KHz. However, the predefined harmonic offset is after a firmware update set to 10.5 and is not far away to what is created during the calibration above 5.34GHz, so is not essential except you want to measure above 5.35GHz with best possible accuracy. However, there is a simple work around as if you enable ultra mode as choose center frequency 5.34GHz and span 100MHz. Then you apply a signal from a signal generator on one half or one third frequency, and tune the signal frequency so you see a signal 10MHz lower=5.33GHz and 10MHz higher 5.35GHz. The levels shall be equal on either side of 5.35GHz and if not the tune the leveloffset harmonic via a terminal command e.g. change the leveloffset harmonic from 10.5 to 10 as a start, and see if that was in the right direct for the level at 5.35GHz. The level at 5.33GHz does not change. Kind regards Kurt ??????? ? Fra: [email protected] <[email protected]> P? vegne af josephlevy via groups.io ? Hi Erik. |
¿ªÔÆÌåÓýHi Joe One detail I forgot to mention. In config set repeat to X10 to stabilized the marker level fluctuation Kind regards Kurt ? Fra: [email protected] <[email protected]> P? vegne af josephlevy via groups.io ? Hi Erik. |
|
Using the 10Mhz cal output and one of the cheap Chinese MAX2870 PLL's sold at the different places (Amazon link only to show the model) modified to accept an external clock of 10Mhz gives me a 5.34Ghz signal with a frequency offset over a GPS of 430hz after a 20 min warm up of both, the generator and the TinySA ultra
I had to amplify the 10Mhz signal by 35db since the PLL needs between 0 and 10dbm clock but gain is cheap. So, if someone has one of those, it can easy be done even without having a GPS for accuracy. In order for it to accept the external input the internal oscillator has to be disabled (removal of a 0 ohm resistor on mine). I use it with an external GPS signal but wanted to see if it could be done using the cal output. https://www.amazon.com/Generator-Stability-Frequency-MAX2870-23-5%E2%80%916000M/dp/B08YD2F8FF/ref=sr_1_54?crid=3T3X8OPI0FNI6&keywords=max+2871+signal&qid=1674069883&sprefix=max+2871+signal%2Caps%2C83&sr=8-54&ufe=app_do%3Aamzn1.fos.18630bbb-fcbb-42f8-9767-857e17e03685 |
|
Good morning,
The 35db gain is needed when using the calibrator output of the TinySA as the reverence clock. The calibrator output is around -36dbm and needs to be boosted to a level acceptable for the clock input of the PLL if used that way. Both of my TinySA ultra cal outputs are accurate enough to produce a 5.34Ghz signal that is less than 500hz of frequency. That was the gist of the test to find out if the Tiny can be used instead of a GPS. I wanted to give those that do not have a GPS a solution to generate the required signal. I used a MAR-6 followed by a MAV11 with a 75 ohm pot at the output for level adjustment, Its just 10Mhz, close to DC so it is not that critical and the IC can easy handle the slight mismatch. Of course any combination of MMIC or transistors can be use to amplify the signal. Btw, I installed a switch in my PLL case so it can be switched between internal and external clock. The PLL is accurate over all but there are some division ratios it has problems with and the output frequency is off. I have not experimented how the choice of the clock frequency influences this but, for 5.34Ghz with a 10Mhz GPS clock is is spot on showing 0hz deviation on my HP microwave counter.? John |
Kurt Poulsen
John Cunliffe W7ZQ