Siglent SSA3021X Calibration Info lost with Firmware V1.2.9.1

[panman]

Myself and a fellow ham both have Siglent SSA3021X units both purchased late 2016. We recently noticed that both units read about 3.5dB low in absolute amplitude and this appears to be independent of RBW and power level (e.g., 0dBm, -20dBm, -40dBm, ...). Both units have gone through the same firmware upgrades since the unit was purchased. The screen snapshots below are from my unit and I notice the calibration date went missing in action with the V1.2.9.1 firmware upgrade. That leads me to believe the firmware upgrade destroyed the calibration data which could have contributed to the low amplitude readings. Unfortunately we never tried any absolute amplitude tests prior to the V1.2.9.1 firmware upgrade.

Has anyone else noticed the loss of "Calibration Date" from the system info screen with the V1.2.9.1 firmware update and if so, have you seen any degradation in absolute amplitude measurements?

As hobbyists, our test equipment is not traceable to NIST standards but we do see consistent results with multiple power meters and multiple signal generators.

73,
Ron, WB2WGH

[Orange]

I upgraded from 1.2.8.2, and all versions in between to the latest 1.2.9.2a, and have not experienced loss of calibration. I have checked my SA with a calibrated HP signal generator, and the thing is always within 0.5 dB. It changes a bit when warming up. Certainly not 3.5 dB

I'm beginning to suspect your signal source.

If you want you can check the file system in your analyser. The calibration files are located in
/usr/bin/siglent/firmdata0/calib and in /usr/bin/siglent/backup/calib

login with root/ding1234 via telnet to check

[panman]

Orange,

Sounds like your upgrade path is a little bit different than mine. My path was
-V100.01.02.07.07 (factory version)
-V1.2.8.1
-V1.2.9.1 (lost calibration date here, was there just before the upgrade and gone right after the upgrade)
-V1.2.9.2.a

A fellow ham, Doug, and I have identical units purchased within days of each other and both having the same upgrade path (he is still at 1.2.9.1). Both unit have the calibration date. Both of us have several signal sources and consistent results (about 3.5db low). Agreed, this does not rule out the signal source.

We also both have a low cost power meter (https://www.immersionrc.com/fpv-products/rfpowermeter/) that lends credibility that the sources are good (yes, qualitative and not quantitative).

Plan going forward:
- I contacted Siglent yesterday and reviewed the data with them.
- Tonight Doug and I will compare our Siglents against a calibrated spectrum analyzer we just got access to and document the results. I will update the forum with the results along with providing feedback to Scott at Siglent
-Scott will also research the loss of the calibration date and I'm hoping someone on the forum had similar calibration date results (my be upgrade path dependent).
- Based on your suggestion, I will snag a copy of the calibration file, hopefully later today.

73, Ron WB2WGH

[panman]

I'm beginning to suspect your signal source.

Last night we tested against a calibrated spectrum analyzer (Agilent E4405B) and found Doug's Siglent to be on the average of 2.99dB low and mine to be 3.24 dB low across 10 data points (0dBm and -40dBm at 1MHz, 10MHz, 100MHz, 1000MHz & 2000MHz). I have just provided the attached summary to Siglent with objective evidence including screen snapshots from the Siglents and camera images from the Agilent.

[panman]

If you want you can check the file system in your analyser. The calibration files are located in
/usr/bin/siglent/firmdata0/calib and in /usr/bin/siglent/backup/calib

login with root/ding1234 via telnet to check

I telneted to the Siglent and did capture a directory listing of the calibration data. The backup directory did not exist in my case. A copy of that directory is attached as I wanted to provide closure to your suggestion. I continue to work this with Siglent as a warranty issue so not going to dive into the calibration data. From the perspective of intellectual curiosity, is there any online info as to the content of these files?

I still have the sense that something happened to cal data with an upgrade. I think, going forward with any future upgrade, I will not only snapshot the SYS Info screen but will also snapshot the same test data (0dBm & -40dBm at 1MHz, 10MHz, 100MHz, 1000mHz & 2000MHz) with my uncalibrated signal generator both before and after any upgrade. In theory, any deltas between the before and after should be very small. This mitigates the need for a calibrated source.

73, Ron WB2WGH

[Orange]

Sorry Ron, the backup files are in /usr/bin/siglent/usr/backup/calib
The content of the directory seems complete, if I compare it to my system.

On my analyzer the System info page does also not show the calibration date any longer. I think this was a new 'feature' of the upgrade.
I find it hard to believe that calibration is lost with an upgrade. However your system is one of the first analyzer's that were sold, and perhaps something went wrong already in the beginning with firmware upgrades, you just did not noticed it.

Anyhow, the systems are way off in amplitude calibration, so it's best that they are re-calibrated again

ps. I notice in the systems a considerable loss at 2GHz, probably the cable 

[panman]

Orange, OK on the backup file location. I did get confirmation from Siglent they dropped the calibration date from the system info screen. And yes, way out of wack from an amplitude perspective. Still waiting back from Siglent on game plan going forward but hoping they will recalibrate as still under warranty. As to GHz tests, yes likely the cable as we did not care about cable losses in our test given same cable used with all 3 test devices. The cable was 2.5M long older "thin wire" Ethernet cable (50 ohm 10Base2). They work, just not well at GHz. Most of my stuff is HF.

Ron