Siglent SSA3000X and SSA3000X-Plus Spectrum Analyzers

[nctnico]

@Tautech: On the Siglent SDS2000 I owned the BMP screendumps where already an issue and thus brought to Siglent's attention over 2 years ago.
Besides that: I wrote I did some tests and hinted on how Siglent can fix the JPEG output quickly.

Defaulting to calling me a troll isn't going to work. Better come up with good arguments or just accept the truth hurts.

[Bicurico]

Hi,

I think the screenshot file format has been discussed more than plenty in this thread and it is making the thread longer to go through for no real reason. I know that it is in the nature of forums to deviate and discuss in full length certain topics, but I would like to sum it up, so we could move on to other subjects related to the SSA3000X series.

Attached is a picture I edited showing the Siglent BMP and JPEG output side by side. Zooming in, one can see the typical JPEG artefacts, resulting from lossy compression.
The BMP picture has no artifacts, since every pixel is stored 1:1 in the file.

File size of both pictures:

BMP = 1.801 KB
JPEG = 82 KB

It now occurs to me that I could actually have freezed the spectrum to save two identical screenshots in either format, but I think even so, the results are clear to see.

Sum up:
1) Siglent allows to either store screenshots in BMP or JPEG format (at least in firmware 07.07)
2) There is no configuration for the JPEG format
3) It is not possible to store in PNG format (lossless compression)
4) IN MY OPINION the JPEG image is perfectly fine to use - I would be happy to use the pictures in reports and magazine articles. Artifacts can be seen if one searches them, but the relevant information of the screenshot is fully usable
5) Storing a BMP takes considerably more time, which can be a nuissance (JPEG = almost instantly, BMP takes like a few seconds)
6) The file size of BMP is IN MY OPINION irrelevant, since any 3 Euro buy a 4GB flash memory capable of storing thousands of screenshots
7) Would it be nice to add PNG? Yes. Can *I* live without it? Yes: JPEG quality is perfectly OK for me and I do write for publications and if required I can store in BMP or through EasySpectrum on PC (which in my case is nicer because I don't have to go through the USB stick)

Hopefully this helps clearing this up for good.

Regards,
Vitor

[tautech]

@Tautech: On the Siglent SDS2000 I owned the BMP screendumps where already an issue and thus brought to Siglent's attention over 2 years ago.
Besides that: I wrote I did some tests and hinted on how Siglent can fix the JPEG output quickly.

Defaulting to calling me a troll isn't going to work. Better come up with good arguments or just accept the truth hurts.

Again you fail to read or understand previous posts or blatantly refuse to believe what I have said. 
https://www.eevblog.com/forum/testgear/siglent-ssa3000x-spectrum-analyzers/msg1012253/#msg1012253
https://www.eevblog.com/forum/testgear/siglent-ssa3000x-spectrum-analyzers/msg1012293/#msg1012293

So moves ARE afoot to address this issue and further discussion on the file type issue is only 

[pascal_sweden]

Siglent is working on it!

And yes, nctnico, you can always go for a mail in rebate and trade your GW-Instek for a Siglent

[Muxr]

Siglent should really implement PNG. It's the perfect format for saving instrumentation screenshots. It's also patent free. https://tools.ietf.org/html/rfc2083

[tautech]

Siglent should really implement PNG. It's the perfect format for saving instrumentation screenshots. It's also patent free. https://tools.ietf.org/html/rfc2083

We know.
To quote some content of a high positioned Siglent employee I got in an email today (yes Sunday here)

I think we'll have PNG in the next FW release. I've been talking to them about it.

Patience grasshoppers. 

[pascal_sweden]

There should be plenty of open source libraries available for Linux related to PNG

Siglent can probably even pick from several ones!

[bson]

RF-Loop has pointed out that PNG is the best file format for this, since it’s lossless and the resulting files are tiny nevertheless.

Let's just be clear about one thing in general - PNG is not inherently lossless.  For continuous-tone images it has to map the tonal values to palette entries, which is a lossy operation that results in dither.  In fact, it's horribly poor for photography and you'll easily see dither and blotchy skin type artifacting as it applies a histogram-based mapping and dither.  It's *only* lossless when there are fewer discrete colors in the image than the selected PNG palette, so if you convert an instrument graphic with a 3-bit palette to a 4-bit PNG color map you're golden.  In that case it's lossless.  If you map a 10-bit palette to a 4-bit one there will be decimation and dither and the results are butt ugly.  It's *only* lossless in this particular use case of saving instrument screenshots and graphs, one it's absolutely perfectly suited for.

[GEuser]

In the 15 year old floppy drive scope i have the choices are TIFF and BMP , does this 3000x have a speaker in it? , they could set it up so when it saves a image to USB it also plays the sound of a Floppy at work (you do not know what you are missing out on!) ..

[KE5FX]

RF-Loop has pointed out that PNG is the best file format for this, since it’s lossless and the resulting files are tiny nevertheless.

Let's just be clear about one thing in general - PNG is not inherently lossless.  For continuous-tone images it has to map the tonal values to palette entries, which is a lossy operation that results in dither.  In fact, it's horribly poor for photography and you'll easily see dither and blotchy skin type artifacting as it applies a histogram-based mapping and dither.  It's *only* lossless when there are fewer discrete colors in the image than the selected PNG palette, so if you convert an instrument graphic with a 3-bit palette to a 4-bit PNG color map you're golden.  In that case it's lossless.  If you map a 10-bit palette to a 4-bit one there will be decimation and dither and the results are butt ugly.  It's *only* lossless in this particular use case of saving instrument screenshots and graphs, one it's absolutely perfectly suited for.

You're confusing .PNG and .GIF.  .PNG supports true RGBA color.  (It's true that it's still not a great format for photographs, of course.)

There's very little room for debate on this stuff.  Manufacturers should support .BMP and .PNG and call it a day.

[et328]

All in all, I do like the device very much. It is a great step up in regards to the best of my field meters with spectrum function.

But it can't do many of the things that field meters do. I'm also interested of the Siglent but I also want DVB-S/S2/T/T2 signal functions. So looks like it's devices like Rover HD Tab 7 or Promax HD Ranger that offer those functions with decent spectrum function. Too bad that those are way more expensive than the Siglent. I own two Emitor devices already.

[tautech]

All in all, I do like the device very much. It is a great step up in regards to the best of my field meters with spectrum function.

But it can't do many of the things that field meters do. I'm also interested of the Siglent but I also want DVB-S/S2/T/T2 signal functions. So looks like it's devices like Rover HD Tab 7 or Promax HD Ranger that offer those functions with decent spectrum function. Too bad that those are way more expensive than the Siglent. I own two Emitor devices already.

Your profile gives no hint to where you are. 
There're Siglent dealers all over and many likely have a demo model to have a try with.
The last one I sold was carefully checked out by a forum member before he bought.

[et328]

Your profile gives no hint to where you are. 

Added the flag. So I'm from RF-Loop country... 

[rf-loop]

All in all, I do like the device very much. It is a great step up in regards to the best of my field meters with spectrum function.

But it can't do many of the things that field meters do. I'm also interested of the Siglent but I also want DVB-S/S2/T/T2 signal functions. So looks like it's devices like Rover HD Tab 7 or Promax HD Ranger that offer those functions with decent spectrum function. Too bad that those are way more expensive than the Siglent. I own two Emitor devices already.

SSA3000X can not do this at all.

But, also vice versa, Emitor Combolook  can not do what SSA3000X can if we talk about RF spectrum analyzer.

Just like my ice scream machine can not cook coffee.

Promax HD ranger is much more expensive "travel TV" but no any idea what it is if think RF spectrum analyzer.

Just, different tools for different needs.

[Bicurico]

All in all, I do like the device very much. It is a great step up in regards to the best of my field meters with spectrum function.

But it can't do many of the things that field meters do. I'm also interested of the Siglent but I also want DVB-S/S2/T/T2 signal functions. So looks like it's devices like Rover HD Tab 7 or Promax HD Ranger that offer those functions with decent spectrum function. Too bad that those are way more expensive than the Siglent. I own two Emitor devices already.

Hello,

I was not implying that the SSA3021X was REPLACING any of my field meters! I was just saying that the possibilities of the SPECTRUM ANALYZER is far beyond the ones implemented in TV field meter!

If you need to do measurements for TV, CATV and SAT installations, you need a field meter, not a pure spectrum analyzer.

Yes, they are more expensive, but then they do have many special functions:
- DVB-S/S2/T/T2/C demodulator
- MPEG-2/4 CODEC
- Measurments: BER, MER, Constellation, Echo, MER vs Carrier, etc.
- Spectrum analyzer (configured to show transponders, hence normally with RWB of 1MHz)
- battery for on the field operation
- ...

Let me know if you need opinion on what device to purchase: I own devices from Rover Instruments, Spaun, Deviser, KWS, Kathrein, Emitor, etc. (unfortunatly so far no Promax).

Because I have all these field meters, I wanted to have a pure spectrum analyzer for special cases and to compare spectrum analysis performance on modern field meters. Some new devices feature a pretty good spectrum mode (like the Deviser S7000 - http://www.tele-audiovision.com/TELE-audiovision-1309/eng/deviser.pdf).

Example: while you can see all QPSK/8PSK transponders on the spectrum function of all these field meters, you fail to see some extra transponders which are not QPSK/8PSK. Believe it or not, but some very narrow band signals exist, which you can only see if you reduce RWB to 10-100kHz. I still cannot demodulate them, but knowing they exist, measuring their signal power and bandwidth, allows me to then try to analyze them by other means (i.e. TBS satellite card with CrazyScan or hooking up a radio scanner with ZF output for subsequent PC based analysis).

Regards,
Vitor

[Bicurico]

Hi,

Here is an example: the first pictures shows the spectrum of Astra 19.2E as seen by the Emitor Satlook Color (which I presume you have).

The next three pictures show the SAME spectrum as seen by the SSA3021X. Notice how far I can "zoom" into the spectrum, revealing the tiniest signals!

Still, the Emitor will do measurements, demodulation, constellation diagram, show MER and BER and even analog and DVB-S picture. Totally different device class.

Regards,
Vitor

[et328]

Here is an example: the first pictures shows the spectrum of Astra 19.2E as seen by the Emitor Satlook Color (which I presume you have).

The next three pictures show the SAME spectrum as seen by the SSA3021X. Notice how far I can "zoom" into the spectrum, revealing the tiniest signals!

Thanks for the info and pics. I have Combolook Color HD S2T2 and Satlook Micro. How fast is the spectrum display (of Astra) on the Siglent? Can you set it to almost realtime with resolution similar to Satlook? Or is it always way above Satlook Color's spectrum performance?

[Bicurico]

Here is an example: the first pictures shows the spectrum of Astra 19.2E as seen by the Emitor Satlook Color (which I presume you have).

The next three pictures show the SAME spectrum as seen by the SSA3021X. Notice how far I can "zoom" into the spectrum, revealing the tiniest signals!

Thanks for the info and pics. I have Combolook Color HD S2T2 and Satlook Micro. How fast is the spectrum display (of Astra) on the Siglent? Can you set it to almost realtime with resolution similar to Satlook? Or is it always way above Satlook Color's spectrum performance?

Hi,

I know EXACTLY why you are asking this question, so let me give you a longer answer, to make it eventually interesting to others.

The Emitor Satlook Digital series of field meters feature indeed a pretty fast realtime spectrum. This comes very handy for aligning a satellite dish, as you can immediatly "see" the satellite (much before a tuner can actually lock onto the satellite) and by the shape of the spectrum you can, with some experience, guess what satellite it is. Hands down, this is the best possible aid to align a satellite dish and fine tune it.

However, the Emitor products don't exactly stand out for precision or full set of measuring features. This is OK at the price they are sold. And again: I still use my Satlook Digital NIT (the one with the black and white CRT) as the primary device to tune my dishes, as it is the least valueable device to be carried to the roof. And the spectrum does what is needed.

BUT: the spectrum that you see, is just a representation of what is really happening (like with any spectrum analyzer) and in this case, the representation was made to be fast with a low resolution.

Cheap Chinese meters take this to a whole new level and actually show extrapolated spectrum frames with random noise inserted, just to show an apparently realtime spectrum! Disconnect the cable and the spectrum lives on for another 3-5 seconds... 

When it comes to a dedicated spectrum analyzer, it is not the speed that matters! I had to learn this myself.

Why? Because you can actually set the SWEEP speed to whatever value you like (within some limits)! But, if you set it too fast, the ADC won't have time to actually see anything usefull.

RBW, VBW and SWEEP go hand in hand and depend on the selected parameters.

Think of it like this:

The RBW is like the apperture letting in the signal. The smaller it is, the more detailed a signal you can distinguish.
SWEEP is the speed with which you will run the RBW "window" across the SPAN. Make it too fast and you "didn't have time to properly look into the apperture".

Finally, coming to your question: if I set the SPAN to 1GHz and use a RBW of 1MHz, then yes, the Siglent is as fast or even faster than TV field meters, including the Emitor.
If I want to set the SPAN to 3GHz and use a RBW of 100kHz with a suitable VBW (think of it as a filter to keep the noise level down by averaging the reading of each "RBW window" during sweep), then the refresh rate goes down. However, you will see things otherwise invisible.

Last example: Look at a picture with 800x600 pixels and then at the same picture with 1920x1080 pixels. No question which has more detail. That is what RBW is all about and apart from the Deviser S7000 no other field meter allows to toggle this filter. Some allow to select fast/very fast spectrum or something similar, which means you can select from two predefined RBW values. But the remaining parameters are still preconfigured to show analog or digital TV transponders (again, the Deviser S7000 is an exception and I don't know about the Promax range of products).

I think that the SSA3021X is not an option for you, if you are into dish alignment: remember that the SSA3021X does NOT provide power for LNB or 0/22kHz & 14/18V & DiSEqC support. I am using the Loop Trough port of a satellite receiver to get the satellite spectrum images. Also, the lack of internal battery makes it evident that you cannot use it on the roof.

So again, I didn't buy it to replace any of the field meters, but to learn, experiment, do DXing and to evaluate the quality of the spectrum analysers in modern TV field meters I get to test.

Hope this helps a little.

Regards,
Vitor

[TurboTom]

Today I had a little time for "Sunday Tinkering" while the "girls" are still to return from holiday and it's too hot to do anything else (the basement is my preferred location under these conditions...) 

So I made a return loss bridge to use with the Siglent analyzer in VSWR mode. Some of the more difficult parts to find is very thin 50Ohm quality coax cable to make the balun to couple the bridge to the measurement port. Actually I found a seller for 1.2mm PTFE coax but then I came across an offer for adapters that connect an SMA socket to one of those ultra-tiny PCB clip-on connectors (for WiFi or cell phone equipment). But the best about this was that it contained about half a meter 0.8mm PTFE coax and cost close to nothing (at least compared to the amounts to be spent for a "proper" RLB -- it must have been arond 4€). So I quickly found some scrap double sided FR4 (ages old but the photoresist still working), mixed some developer and etching solution and tried if I still can "cook" PCBs "the old way"... 

To be honest, I planned that already some time before and slected some toroids for matching AL values out of a bunch.

And look at the result and how it performs, even without a shielding enclosure (this will be milled some time in future, first have to top up my stock of 10mm aluminium plate). The screenshot is from an SMA WiFi antenna connected to the test port -- doesn't seem to be the worst quality.

Cheers,
Thomas

[Performa01]

So I made a return loss bridge to use with the Siglent analyzer in VSWR mode.

Nicely done! I love that!

Now it would be interesting to see the usable frequency range – could you show us the return loss with a quality 50 ohms terminator instead of the antenna?

[TurboTom]

Well, I did some more testing with the RLB, as-is, without any tweaking or shielding so far. Since most of my RF stuff is DIY or otherwise of questionable origin / condition, all the figures are to be taken with a grain of salt...

I did the following tests: The RLB is connected to the SSA via a pair of N/SMA adapters and 15cm long semi-rigid SMA/SMA cables, one straight and one 90° connector each. I salvaged these cables from a "Superfilter" (professional GSM gear), so they should be of reasonable quality. The first test was to normalize the TG with a straight-through SMA(F) adapter instead of the RLB and then to take a spectrum of the RLB with the test port open, shown in the first screenshot attached.

Unfortunately, the SSA somehow works erroneously on other traces if TG normalization is used so the following scans are taken without normalization. It appears the SSA occupies one of the traces for the normalization data but I didn't look at this in detail for the moment, I just disabled this function.

This takes us to the second screenshot. The orange trace shows us again the RLB with the test port completely open. For the magenta trace, I attached a 15cm semi-rigid cable to the test port, once again with an open end (open SMD SMA(F) socket installed). The cyan trace shows the same configuration, but this time with a reflective short (shorted SMD SMA(F) socket installed). And finally, the green trace was taken with a DIY terminator (SMD SMA(F) socket with 2* 100Ohm 805 resistors).

The resonance peaks changed significantly if I only touched the screen of the semi-rigid cable at the test port, so I guess proper shielding of the RLB will improve the situation.

Then I decided to keep the orange and the green trace and test two other 50 ohm terminators that I've got - the magenta trace of the third screenshot shows a Barry 50W power terminator which came with 23cm of rigid microwave cable premanently attached to it (nice beryllium oxide ceramics device with the proper warning labels attached...). I soldered a 90° quality SMA connector to the open end. The cyan trace in the third screenshot shows a MiniCircuits SMA(M) terminator (MCL-ANNE-50+ DC-18GHz).

Finally, I thought it might be interesting to compare the DIY RLB to a MiniCircuits ZEDC-15-2B (1MHz...1GHz) directional coupler, which is shown in the fourth screenshot. The orange and the magenta traces originate from the DIY RLB whereas the cyan and the green ones are from the MiniCircuits coupler. The orange and the cyan traces were taken with the test port open while for the magenta and the green traces, the aformentioned MCL-ANNE-50+ terminator was directly attached.

I'm quite curious what effect shielding (and maybe a little bit of tweaking) of the DIY RLB may have... Definitely still room for improvement, but as a starting point not too bad I would say! 

Cheers,
Thomas

[Performa01]

I'm quite curious what effect shielding (and maybe a little bit of tweaking) of the DIY RLB may have... Definitely still room for improvement, but as a starting point not too bad I would say! 

Shielding is a must, but other than that, I cannot see any significant improvement and the performance is quite impressive. I take my hat off to you and your fine effort!

Your RLB clearly outperforms the MiniCircuit device for frequencies >100MHz by quite a margin.

Your RLB appears to be very usable up to at least 2.5GHz. Now the intriguing question remains: how low can it go in frequency? 1MHz like the MiniCircuit? If so, you most definitely have a winner here. 

(and if not, it's still great work!)

[TurboTom]

The low frequency performance of my DIY RLB is superb and outperforms the MiniCircuits coupler considerably! See the following two screenshots. Also, a big plus, shielding is not an issue here  . Trace A is reflective open, Trace D reflective short. Traces B and C are with 50 Ohm termination, B with the MiniCircuits terminator, C with the (apparently more accurate) DIY version. I would say the usable range starts at 65 kHz.

What I really think needs improvement on the SSA is how "frozen" traces are processed. They scale properly if level-related adjustments are done, but if frequency-related parameters are changed, they just stay where/as they are. No stretching/shrinking or shifting at all so they are completely useless if they are supposed to be compared to a newly recorded trace that (partly) covers a similar span. On the DSA815 this is solved much better!

Cheers,
Thomas

[Performa01]

The low frequency performance of my DIY RLB is superb and outperforms the MiniCircuits coupler considerably!

Great news, congratulations! 

To be honest, I’ve almost expected that – your construction looks like it has the potential for a very good low frequency performance. So in the end, you’ve built a really useful tool!

Regarding the reference traces I have to agree with you. References are of rather limited use if they don’t shift/scale with the current settings accordingly. Still I would not do it like I’ve seen it on the DSA815, i.e. I would not want the reference trace to extend beyond the existing data. If we increase the span, well then the reference trace should be only visible for that part of the screen where data is available.

Ideally, we would be able to zoom into a reference trace as well, i.e. going to narrower spans and even lowering the RBW. Unfortunately I cannot see a reasonable way to implement this, so it’s not going to happen I’m afraid

The only realistic option is to compress/decimate the reference data for wider spans and spread/interpolate them for narrower ones.

[Bicurico]

 

Congratulations!

Can/will you provide some schematics and BOM, so that one could try to recreate your design?

Regards,
Vitor