Spike frequency error reading on version 3.1.x with SA44B

Hi Andrew,
I’ve sent the same message to your support e-mail address together with Spike preint screens. I will attache two of theme here too.
Let me explain again the used setup measurement.
An RF generator was connected to SA44B input throught a 50ohm 6dB fixed attenuator. This hardware setup was not changed throughout the test! The RF generator was set to transmit a fixed unmodulated CW frequency. That frequency was 10MHz and 150MHz. I chose two frequencies to show that the frequency offset is not constant but increase with frequency!
Now, what was changend in the measurements was just the Spike software version! The hardware connection was keept unchanged. The RF generator is miniVNA PRO and have very good frequency stability (less than 10Hz/hour).
As can be seen from the attachement, with Spike ver. 3.0.19 the RF signal has a sharp 10.000000MHz and with the latest Spike ver. 3.1.18, the same RF carrier was read as 10.001773MHz. Somehow the new version add about 1.773kHz to the real frequency and this is even worst as the frequency increase.

Also from the att. you can see that RBW=VBW and are the same for both versions (30Hz). The analisys mode was Sweep measurement without spurious rejection. Ans yes, I put the marker 1 just to be clear the frequency reading. I use the peak search function which is working well.
I think you missunderstood the problem! There is NO hardware frequency offset. So how can I verify the frequency offset with different software version?

Please let me know if now is clear.
If you need more information, please let me know.

Thank you for support.

BR,

Cristian Negru

Hi Andy,

Thanks for the advice. Problem solved.
I didn’t know that on the new version 3.1.x the ref. freq. should be adjusted.
For this, as Andrew mentioned, I’d used Utilities>Timebase Adjustment.

Thank you both of you!

BR,

Cristian Negru