Siglent SSA3000X and SSA3000X-Plus Spectrum Analyzers

[videobruce]

Am I missing something here, or it that the ultimate 'bug'? 
If something like that appeared on my unit I would create a real fuss!   

[hendorog]

Am I missing something here, or it that the ultimate 'bug'? 
If something like that appeared on my unit I would create a real fuss!   

I'm assuming you are not joking, so to clue you in:

Porting doom to a linux box is an ancient and worthy activity

The SSA is just a linux PC with a bit of RF hardware added. So someone has ported Doom to run on it.

IIRC the same has been done to Keysight and Rigol devices  - although the Keysight was CE not linux.

[videobruce]

Correct, I was not joking.
A. I don't do vidiot games having close to zero interest in that lame pastime (same goes for toy idiot phones)
B. I didn't consider this a "Linux PC" in spite of it running Linux (which I kinda forgot since that isn't a day to day function).

"Ancient"?? 

[hendorog]

Correct, I was not joking.
A. I don't do vidiot games having close to zero interest in that lame pastime (same goes for toy idiot phones)
B. I didn't consider this a "Linux PC" in spite of it running Linux (which I kinda forgot since that isn't a day to day function).

"Ancient"?? 

No worries, enjoy your day Bruce.

[tautech]

New firmware for SSA3000X models.

Version 1.2.9.3a
7.5 MB
http://old.siglentamerica.com/USA_website_2014/Firmware&Software/firmware/SSA3000X_1.2.9.3a_EN.zip


Changelog
1. This firmware must be upgraded from V1.2.8.1 or later. If your SSA3000X has an earlier version, please
    update to V1.2.8.1 first and then proceed to update to the latest version.
2. Fixed the OBW - Transmit Freq Error bug
3. Fixed the LOG scale –Correction display error
4. Fixed some marker and display text bugs
5. Add “UNCAL ”warning when setting changed after TG normalization
6. Add “Select All” in File-Operate menu
7. Add peak table “Sort By Freq/Ampt“ menu in Peak-Search Config
8. Optimize Freq Offset to 10 GHz, Scale/Div to 0.1 dB
9. Optimize several SCPI commands, such as *OPC? ，ESR? , etc.
10. Optimize .sta file limit parameters, Optimize .csv file parameters

[JPortici]

New firmware for SSA3000X models.

Version 1.2.9.3a
7.5 MB
http://old.siglentamerica.com/USA_website_2014/Firmware&Software/firmware/SSA3000X_1.2.9.3a_EN.zip

number 7 alone is worth it

[rf-loop]

New firmware for SSA3000X models.

Version 1.2.9.3a
7.5 MB
http://old.siglentamerica.com/USA_website_2014/Firmware&Software/firmware/SSA3000X_1.2.9.3a_EN.zip

number 7 alone is worth it

Yes

And same works also now with SDS1004X-E  FW 6.1.33,   FFT (peaks sort by level or freq, limited max 10 peaks)

[tv84]

New firmware for SSA3000X models.

Version 1.2.9.3a
7.5 MB
http://old.siglentamerica.com/USA_website_2014/Firmware&Software/firmware/SSA3000X_1.2.9.3a_EN.zip

https://www.eevblog.com/forum/testgear/hack-of-sigllent-spectrum-analyzer-ssa3021x/msg2452119/#msg2452119

[john95]

SSA3032X TG output shut off after being on several hrs even tho it says its on. Turned off TG than on again but still no output. Reset (Preset) no help, had to power cycle unit in order to get TG to work again.

SSA3032X Firmware ver 1.2.8.3

I have a brand new ssa3021x, just couple of days old. and today, after playing with TG for an hour, TG output was gone even though display said it was on. power cycle it again and same problem happened again 20 min later. do you find any solution?

here is the info:
SW1 1.2.9.2.a
SW2 20180708-1
SW3 000000E1
HW  0F.03.00

[tautech]

SSA3032X TG output shut off after being on several hrs even tho it says its on. Turned off TG than on again but still no output. Reset (Preset) no help, had to power cycle unit in order to get TG to work again.

SSA3032X Firmware ver 1.2.8.3

I have a brand new ssa3021x, just couple of days old. and today, after playing with TG for an hour, TG output was gone even though display said it was on. power cycle it again and same problem happened again 20 min later. do you find any solution?

here is the info:
SW1 1.2.9.2.a
SW2 20180708-1
SW3 000000E1
HW  0F.03.00

Hmmmm
Could you please install Version 1.2.9.3a firmware mentioned just a few posts back.

Also please document any/all settings changes that precede the loss of TG.
That is, if there's an undiscovered bug we need as much info as possible for the factory to replicate it and find a fix.
TIA

[TurboTom]

Right from the start (after I "improved" my SSA that is) I wasn't too happy with the performance of the tracking generator. Once the machine has warmend up, the level drops below -7dB close to 3GHz. That is the case even though I modified the level calibration of the TG, but the calibration range just isn't wide enough. While the SSA is still cold, the situation is somewhat better, it drops by round about 5dB in this case.



Of course, normalization takes care of this during measurements that permit to normalize the instrument without having to change the settings afterwards. Always being nosey, I decided to take a look if there are any hardware tweaks possible that may ameliorate the situation. I decided to take a closer look at the TG module, and there, only behind the TG mixer so the repercussion on the first local oscillator signal in the main RF module would be neglible / non-existant. So I first started to take a close look at the quite complex distributed element filter directly after the mixer:



(Photo probably borrowed from Dave's "teardown collection")

I vectorized the graphics in my CAD system and used the edges to produce a "proper" drawing of it. Using the mounting flange of the N connector as a reference (it's specified to be 1" square), I scaled the drawing to the proper size. The size of the stripline to the N connector, which should have an impedance of 50 Ohms and is approx. 1mm wide, together with the assumption that the PCB material is some Rogers R4003 hybrid multilayer laminate (since for economical reasons, I don't expect Siglent used PTFE based material here), a copper thickness of 35µm and a dielectric thickness of 0.406mm and an ε_r of 3.38 appear correct (those parameters are both stock values from Rogers).

Finally, I calculated the inductances and capacitances of the elements (there are nice online tools available for this; google is your friend) and entered the schematic of the distributed element filter in LTSpice. Actually, I considered my approach to be not much more than a "wild guess", the more I was surprised of the result, see here:



At 4GHz, the attenuation of the signal is 50dB+ ! 
So even though this approach didn't take me closer to a solution of my TG improvement attempt, I'm more than impressed how accurate characterizations of such delicate circuitry are possible with "household" methods. I tweaked some of the component values of the filter in the simulation to find its performance only gets worse...

Time to think about different solutions to get the TG straight  . An interesting approach it was anyway, I think  .

Cheers,
Thomas

[Performa01]

This topic got me curious and just for fun I designed a 3.6GHz low pass filter based on the same structure as in the SSA3000X. If you look at the component values, my approach quite obviously uses very different design parameters:


LPF_01_CD

Here is the frequency response up to 4GHz:


LPF_01_FR

And here’s the response up to 3.6GHz, so we can see the details of the pass band:


LPF_01_PB

I think it’s funny how I get a prominent dip at 2.45GHz, which looks a bit like the mirrored version of the original dip at 2.95GHz. Of course, the dip is negligible at some -0.25dB in my simulation. Yet I think the tendency is clearly visible and with non-ideal components and higher tolerances (and maybe different design parameters) it might turn into a problem.

Apart from that, this problem should not be typical. Specification for amplitude flatness of the TG is +/-3dB.

Hasn’t there been a delay on delivery of the first SSA3032X because Siglent had to do a revision of the tracking generator for its unsatisfactory output level flatness?  I would not be surprised if the circuit boards produced up to that point had not been thrown away but used up for the low cost 2.1GHz SSA3021X instead.

[tautech]

Hasn’t there been a delay on delivery of the first SSA3032X because Siglent had to do a revision of the tracking generator for its unsatisfactory output level flatness?

No, while the TG flatness is early units was not good later HW versions were revised to improve flatness but supply AFAIK was never affected. The Ohio GM made posts about this HW change earlier in this thread.

I would not be surprised if the circuit boards produced up to that point had not been thrown away but used up for the low cost 2.1GHz SSA3021X instead.

Possibly IDK or even reworked for better flatness. AFAIK there are screen shots of early HW TG flatness here if you hunt them out, however the issue didn't effect typical operation after Normalization.
Of course the recent addition of a UNCAL warning when adjustments have been made after normalising alerts the user to correct operation of normalising after appropriate settings have been made and not before.

[videobruce]

Regarding the latest F/W. Any issues so far (in spite of nothing reported) and any of those 'fixes' amount to anything?
I see Siglent still refuses to acknowledge, let along address the BW/Sweep issue.

[hfleming]

Does anybody have any idea how to recover a SSA3021x that is stuck on the boot screen?

[tautech]

Does anybody have any idea how to recover a SSA3021x that is stuck on the boot screen?

I'm sure the forum community can help but we need a bit more info than what you have provided here:
https://www.eevblog.com/forum/testgear/hack-of-sigllent-spectrum-analyzer-ssa3021x/msg2589123/#msg2589123

I suggest you pick one thread or the other and sort the problem in it only.
There are smart clever SSA owners that can help.

For dealers, are there boot recovery files for SSA ?

[tautech]

Regarding the latest F/W. Any issues so far (in spite of nothing reported) and any of those 'fixes' amount to anything?

No issues, all new units come with 9a installed.

I see Siglent still refuses to acknowledge, let along address the BW/Sweep issue.

From my investigations it's a known issue and a big rewrite of code to fix so will take time.
FYI SVA models are not affected.

[hfleming]

Does anybody have any idea how to recover a SSA3021x that is stuck on the boot screen?

Keeping the “system” button pressed whilst booting fixed the issue

[videobruce]

I see Siglent still refuses to acknowledge, let along address the BW/Sweep issue.

From my investigations it's a known issue and a big rewrite of code to fix so will take time.
FYI SVA models are not affected.

Thanks for that, that is the 1st I had a somewhat straight answer (including Siglent) as to the problem that is well over a year old. But, if the newer model doesn't have the problem, they obviously figured out what/where the problem is, one would logically think that would/could be re-written here also.

[charliedelta]

Anyone have a recommendation for a good transient limiter?  Since I will be mainly using by Siglent for EMC  work something with  high pass filter would be nice.

[tautech]

I see Siglent still refuses to acknowledge, let along address the BW/Sweep issue.

From my investigations it's a known issue and a big rewrite of code to fix so will take time.
FYI SVA models are not affected.

Thanks for that, that is the 1st I had a somewhat straight answer (including Siglent) as to the problem that is well over a year old. But, if the newer model doesn't have the problem, they obviously figured out what/where the problem is, one would logically think that would/could be re-written here also.

We hope so too and it's only recently that I've had a direct line to the SSA/SVA team.

As the BW/Sweep issue can be manually worked around it's not the highest priority fix unfortunately but here's hoping when the development of the new OS in SVA for the 3.2GHz model is released programmers will have time to revisit SSA and sort this bug out. I'm not sure if the SVA code can be ported back into the SSA for a quick fix so we'll just have to let them sort it out when they can.
Sorry for the PITA as for now you'll just have work around it with the manual corrections you've been doing.

Anyone have a recommendation for a good transient limiter?  Since I will be mainly using by Siglent for EMC  work something with  high pass filter would be nice.

What pads/attenuators do you have already ?
You have 51dB of internal attenuation you can dial in and the SSA inputs are 50V max rated so they already have a DC blocking cap for input protection.

When I got mine I was scared of frying the front end so I got the accessories kit:
https://www.siglentamerica.com/accessory/ukitssa3x/

TBH I rarely use the 10dB pad included now with the confidence gained from using SSA for a while.
Default has 20dB attenuation preset into settings and just some time with a short piece of wire tuning in local radio stations using Demod and an earpiece can get you used to SSA and its sensitivity.
You'll quickly learn to move WiFi devices and cellphones well away when dialing in any moderately sensitive settings.
My advice for the new SSA owner is to use Preset occasionally to return the instrument to known safe settings until you become familiar with the SSA and spectrum analyser usage in general.
Be thoughtfully careful and you'll be OK.

[Performa01]

Anyone have a recommendation for a good transient limiter?  Since I will be mainly using by Siglent for EMC  work something with  high pass filter would be nice.

I don't have it, but this one looks really decent:

https://emcsupplies.com/collections/rf-accessories/products/tbfl1-transient-limiter

[charliex]

haven't used my ssa 3021X pretty much since i got it in 16 or so, booted it up and the screen looks like the contrast setting is too high, backlights on too bright, and the display is washed out and hard to read, tried the factory reset etc,  turned it on and off and all that, to no avail, anyone seen it? am i missing a contrast setting somewhere or did something just go wacky inside.

the easy pc software works, though the new version i just updated too seems to be worse than the one i had before but i can use it that way.

cheers

[tautech]

@charliex
We'd need a pic to properly see what you describe please.
I see you're in Las Vegas so the guys in Ohio might be your closest service center if there's a warranty issue.
Fingers crossed it's still covered.
Get that pic up and I'll point Siglent guys to your post.

[charliex]

super hard to read on screen. i think its just out of warranty since i think it's just over the 3 year mark since i picked it up. literally second time i'd used it , heh . also about to go to vegas for defcon (moved back to LA), shame i can't really bring it, ah well , hopefully the pc software does everything the device itself does. (still has the screen protector on it i noticed)