Hack of Sigllent spectrum analyzer ssa3021X?

[dembi]

Hey guys!
I'm about to buy SSA3021X Plus. Do I understand correctly, the coax cable rework inside is required to self auto-calibration so there's no need to calibration kit and manual calibration?

[tautech]

Hey guys!
I'm about to buy SSA3021X Plus. Do I understand correctly, the coax cable rework inside is required to self auto-calibration so there's no need to calibration kit and manual calibration?

Welcome to the forum.

Not quite, the coax is missing to restrict simple VNA conversion from a SSA3kX+ to a SVA model.
A Cal kit is still required for precision measurements.

[dembi]

Thank you for such quick response. There's a promo for this SSA3021X Plus till the end of 2025 so I'm considering buying it.

Not quite, the coax is missing to restrict simple VNA conversion from a SSA3kX+ to a SVA model.
A Cal kit is still required for precision measurements.

Siglent provides multiple calibration kits. I found this one https://www.siglent.eu/product/2165861/siglent-f503fe-4-5-ghz-n-female-type-vna-calibration-kit. Would it be ok? Should I get another one? I saw some people have written about calibration kits with a bridge. Should I get this one too? I found this kits quite expensive. May I use other, cheaper, yet with same quality?

The big problem is, and this isn't mentioned, that the CAL file is unique to each device and that ultimately hacking it will give you a bad VNA, while if you buy a real SVA you'II have a much beter VNA at your disposal.

So, shouldn't it be calibrated with a calibration kit?
If this is not the case, isn't the original calibration info backed up with the command and later restored:

Code: [Select]

cp -R /usr/bin/siglent/usr/vnacalikitdata /usr/bin/siglent/usr/mass_storage/U-disk0/SA-vnacalikitdata

[willie.from.texas]

That is a cal kit for use with the VNA. As far as I know, there is nothing special about that kit except for the cost. Maybe it comes in a nice box. I initially used the cal kit that came with my NanoVNA, but ordered an N-type cal kit. There is a kit available on Amazon rated to 3GHz for $20 U.S., that I am sure would work for most anything the VNA is capable of measuring.

[tautech]

That is a cal kit for use with the VNA. As far as I know, there is nothing special about that kit except for the cost. Maybe it comes in a nice box. I initially used the cal kit that came with my NanoVNA, but ordered an N-type cal kit. There is a kit available on Amazon rated to 3GHz for $20 U.S., that I am sure would work for most anything the VNA is capable of measuring.

Yeah but……
To do the job properly you should be calibrating a VNA as close as possible to the DUT to remove any adapter or cabling effects.
Most don’t want to work with N type as SMA is generally much nicer to use and this is why many use SMA Female Cal kits.

[KungFuJosh]

Don't buy the crappy and/or overpriced cal kits, just get the LibreCAL and search this forum for how to use it with the Siglent.

[tautech]

Don't buy the crappy and/or overpriced cal kits, just get the LibreCAL and search this forum for how to use it with the Siglent.

Or get something tested and decent from member hendorog:
https://vnalab.net/collections/all

[KungFuJosh]

Or both.

For me, I would get the LibreCAL. I already have N and SMA standards (nothing special), but one of the big reasons the Siglent interests me is for the LibreCAL autocalibration. Which is also already tested by a forum member that set it up.

[dembi]

Just did the crossmod from SSA3021X Plus to SVA1032X + all options enabled without any issues. At least on the software side of things, haven't put the jumper in place yet. Still debating whether or not I'll put proper SMA-connectors in place or just hardwire some semi-rigid coax in place. Anyway, you guys are awesome! 

UPDATE: I ended up just soldering some coax in place. All works as intended after calibration.  Also, I upgraded the FW to version 3.2.2.5.1R1 (official download from Siglent) and everything is still enabled and working.

My SA came with exactly the same version of software and hardware. I've managed to enable 3032 and all options successfully. @BalticLab, could you tell me which version of the firmware for SVA1032X did you use to crossflash? Was the 3.2.2.4.0 your first try?

I saw some reworks with the coax in the previous posts. What connectors did you use, guys? I saw the footprint for the connectors is different than usual. Where did you buy so short coax cable with plugs?

[marcopolo]

Hi,

I updated my SSA3021X+ by following the Mandor procedure:
https://www.eevblog.com/forum/testgear/hack-of-sigllent-spectrum-analyzer-ssa3021x/msg5746107/#msg5746107

The VNA works correctly compared to a nanoVNA.
However, I have a problem with the spectrum analyzer part:
I have a defect that repeats every 1.25MHz.



This can be seen more clearly by connecting the tracking generator directly to the input:



Second problem: the output level of the tracking generator is incorrect:





If anyone has any ideas for solving these problems, I'm all ears
Thanks.

[Gerd_02]

Hello, I'm new here as a user, but I've been reading along for a long time. I find this forum, especially the SSA3000x section, very interesting and informative.

I have an original SSA 3032x Plus, with all the available options unlocked via a Python script. I haven't performed a VNA upgrade.

I replicated the settings shown in the screenshots and haven't found any errors on my device except for the sawtooth pattern. The sawtooth pattern is also present on my device with extreme settings, every 1.26 MHz, but I think that could be due to the ADC's sampling window. However, it's not visible, or barely noticeable, in most measurements, so it's not a problem for me.

[willie.from.texas]

Interesting. I measure something very similar but with a higher amplitude. My instrument is the SVA1032X converted from the SSA 3021X Plus.

[Gerd_02]

Yes, that's really interesting. I took the noise level screenshot back in 2024. Unlike yours, mine was taken without a signal and with a vertical resolution of 1 dB/div. I've now taken another one with TG activated at -20 dBm and 0.1 dB/div, using exactly the same settings as yours. It looks different on my instrument compared to yours.

[Gerd_02]

I'm sorry, the attachment failed to load.

[dembi]

Guys, could anyone point to the SMA connector part number which fits into the PCB? I tried 2, one with too big raster/spacing, another one with too big diameter of the signal pin.

[S13]

Because of that exact problem i just put an SMB connector in there:
https://mou.sr/4sQ8Ulg

And the mating part:
https://mou.sr/4qG4mwx

And then some RG316 cable in between.