Siglent SSA3000X and SSA3000X-Plus Spectrum Analyzers

[luma]

Has anyone documented a workable solution for bandwidth upgrades on a SSA3021X Plus?  I note one other user has claimed success but I'm not finding any public process for making that happen.  Is this a thing on the Plus or only on the not-Plus version?

[durec]

Hack of  SSA3021X Plus with SW 2.2.1.2.5:


After hack, SN is XXXXXXXXXX. Perhaps someone solve it and let us know.
All Options are permanent.


- using SSA3000X_Plus_telnet.ADS file on USB stick, run firmware upgrade of SSA
- telnet to your SSA on port 10101.  No password required.

- make backup:
   cd /   
   cp -R /usr/bin/siglent/usr/backup /usr/bin/siglent/usr/mass_storage/U-disk0/SA-backup
   cp -R /usr/bin/siglent/firmdata0 /usr/bin/siglent/usr/mass_storage/U-disk0/SA-firmdata0

- remount filesystem as rw:   
   mount -o remount,rw /dev/ubi2_0 /usr/bin/siglent/firmdata0
   
- cd /usr/bin/siglent/firmdata0
- in file NSP_sn_bandwidth.xml erase <license> .. </license> part
- in file NSP_trends_config_info.xml swap SSA3032X and SSA3021X in lines <product_type_1> and <product_type_2>

- mv nsp_data_b1 nsp_data_b1x   

- cd /usr/bin/siglent/usr/backup   
- mv nsp_data_b nsp_data_bx
- cd /
- sync   
- logout

reboot - turn off / on


That's all.

[1design]

I am working on repairing an Agilent N9340B, upon opening up the RF section, it looked vaguely familiar. I never opened one up before and wondered why does it looks familiar...law and behold the Siglent SSA3000.

Sigilent did not design their own SA, they just ripped off a 10 year old Agilent design, together with all the distributed filter structures, layout and obsolete IC and complete frequency plan.
If they called it a N9340C-desktop version, it would sound more on point The only things they changed are the tracking generator and using discrete PLLs. They probably struggled with reverse engineering the VCOs and decided that the off the shelf option is cheaper and with similar performance.

Buyer beware, most of the RF components used in the SSA3000 are obsolete(HMC307, HMC488 etc), even though this is sold as a recent design. This makes it expensive and much trickier to repair, even thou the design is full SMD and should be very easy to maintain, if modern components were used.

N9340B:


SSA3000:
https://www.flickr.com/photos/eevblog/27749729501/in/album-72157669490076552

[KE5FX]

N9340B:
(Attachment Link)

SSA3000:
https://www.flickr.com/photos/eevblog/27749729501/in/album-72157669490076552

That's fascinating.  An entirely different mentality at work.  You'd be thrown out of any EE school in the US for doing that, but in China, you get paid.  Who's doing it right? 

[Elasia]

If its a proven design, frankly they are.. why reinvent the wheel? And whos to say no when the government backs them? They are just following the law of laziness and least effort to get the job done and left to innovate elsewhere.. and slash the cost

Kind of interesting though... i guess this makes it an HP / Rigol bastard child?  I'm kinda tempted to take my Plus version open and take some pics to compare with now

[Bicurico]

Is it really THAT similar? Isn't the design conditioned by physics?

And of course they opened existing devices for some reverse engineering... Everyone does that.

I don't see any problem here.

Regards,
Vitor

[TurboTom]

...
Kind of interesting though... i guess this makes it an HP / Rigol bastard child?  I'm kinda tempted to take my Plus version open and take some pics to compare with now

That would be quite interesting. Not only regarding the question of the N9340B heritage but also to understand how much it has in common with the also recently released SVA models.

[KE5FX]

Is it really THAT similar? Isn't the design conditioned by physics?

And of course they opened existing devices for some reverse engineering... Everyone does that.

I don't see any problem here.

Regards,
Vitor

They cloned some of the microstrip filters with near-photographic precision.     That's not "reverse engineering."

Earlier, I was chatting with a friend who's an HP retiree.  He pointed out that certain Japanese T&M companies were behaving just as blatantly in the 1980s, if not more so, and cited some specific examples.  I was actually pretty surprised to hear that.  The Japanese certainly went through a "copy everybody else" phase, but from what I saw, they didn't resort to outright plagiarism.  And they always seemed to put an interesting spin on whatever idea or product they were copying, something the rest of us could learn from.

[Bicurico]

The thing I was trying to point out: can you do the microstrip filters with different shape and size or are they otimized in shape and size for the given frequency range?

And yes, opening something up and using whatever information is called reverse engineering. You can call it cloning, plagiarism, whatever. I do agree with you.

Regards,
Vitor

[2N3055]

They have similar block diagram. They are nowhere near to be "photographic copy".  They are different enough that both design had to designed and calculated separately. Especially at high frequencies, even fraction of millimetre of difference will mean new calculations and simulations.
Also there are limited amount of tools for design, that will converge to similar looking parts. Also, as said, most of planar microwave components are geometrically defined as such. They have to look the same to function...

[TurboTom]

The one thing that makes me wonder is that Siglent used components for designing the SSA3000X series of which at least some must have been listed EOL or even obsolete already at the time of design. If these are identical with the N9340B (codes not clearly readable in @Elasia's photo), I would call this strong evidence.

Maybe the release of the "Plus" model after only a few years into production of its predecessor was required due to the availability situation of these components?

[Sighound36]

What about the new LeCroy range of bench equipment I suspect they are using this Siglent range or the later RTSA version?

[Elasia]

The one thing that makes me wonder is that Siglent used components for designing the SSA3000X series of which at least some must have been listed EOL or even obsolete already at the time of design. If these are identical with the N9340B (codes not clearly readable in @Elasia's photo), I would call this strong evidence.

Maybe the release of the "Plus" model after only a few years into production of its predecessor was required due to the availability situation of these components?

Wrong person mate ;p

[TurboTom]

Sorry, confused you with @1design -- was about to leave and somewhat in a hurry... Apologies to you both 

[1design]

Sorry, confused you with @1design -- was about to leave and somewhat in a hurry... Apologies to you both 

Apology accepted, it happens to the best of us 

As you mentioned, it was surprising to see that Siglent decided to use EOL, and by now, obsolete parts in their design.

There are many ways to skin a cat when it comes to distributed filter design, what we see hare is blatant copying. No engineering went into that.
They did spin their own couplers and a few discrete filters, including the 10.7MHz IF section.

They did so little engineering that they even copied the 20dB fixed att + HMC307. There are better ways of doing it today, if they redesigned that, they would have been able to design it in a different way with modern components.

To me, it is more sad than anything else, the N9340B works, but it is far from being cutting edge design. It was a bottom of the barrel unit when released, it should have not been the reference for a design released 10+ years later. This is not a PXA level instrument.

[Elasia]

A few pics

I'll remove the plates tomorrow

[Elasia]

more pics

[TurboTom]

Wow, now we're talking! The tracking source subsection is connected to the main RF block by three wires. Which means that there is a signal fed back to the latter. And this makes me assume (btw something that I predicted already a considerable time ago) that the "Plus" SSA contains the full-blown SVA hardware! You may compare the positions of the internal SMB connectors with the scenes of Shahriar's brilliant review of the SVA1032X. If a hack may actually "liberate" that function and if the instrument is already factory calibrated on the "reflected" channel is another question. But it's very likely that the hardware is identical. In this case, the SSA3021X Plus is a real steal!

[tv84]

Who wants to try?   

[Elasia]

Who wants to try?   

Was wondering if you would chime in..

Have you done a comparison between their firmware files? Was also wondering if one was a fork of the other or if they are the same base and its just the main program with differences?

[tv84]

I think it's pretty much the same...

[Elasia]

Ladies and gentlemen... the million dollar question has been answered on what this platform was based on..

More pics to follow

[tv84]

I always suspected... 

[Elasia]

Same exact boards, even silkscreen version markings

Not going to post more pics, just watch TSP's video

[1design]

So, they did go in and replaced all of the obsolete parts, both the HMC307 and HMC488 were replaced with modern counterparts.
For the second mixer, they selected the HMC213...which again, is already obsolete 
I used this part extensively between 2010 and 2012, it is a good mixer, nevertheless...not for new designs