Siglent SVA1015X and SVA1032X 1.5, 3.2GHz Spectrum & Vector Network Analyzers

[maxi]

What has been your overall impression of the SVA1032X,  and what signal generator where you using for the posted test?

Thanks
[/quote]

I had an Advantest R3361B for many years and replaced it by the SVA1032X in 2021 - ordered all options. It is faster and has better resolution. The phasenoise could be better but correlates with it´s price-class. The phase-noise shown by user IM3 is strange - as you can compare with mine or that of user RoV. The signal-generator is a HP8657B and most signal-generators have better phasenoise characteristics than the Siglent Analyzers.
But Siglent is improving over the years but you have to pay for better specs.

[IM3]

Nice MAXI that you performed my measurement setup in the same way and that your SSA1032X has good phase noise performance.

There are good and bad examples of phasenoise indeed. that makes it more unclear.

I started checking my phasenoise because someone on EEVBlog reported his bad phasenoise. That's how it started for me.
I did the measurement as I published here.

I  went to the Dutch supplier with this problem and he presented it to Siglent Europa. A few months later this was discussed at the head office in China and it was stated that this was hardware related and could not be resolved.
My supplier then took my SVA1032X back and I subsequently bought an SSA3032X-R which Siglent China indicated had no problems.

I can see that my SVA1032X was really bad compared to some others. I have lost confidence in that RF design and have no regrets that I made the switch to the better RF design.

Whether this is exemplary will only become clear if more people perform the same measurement and post it here.

[EE-digger]

Discovered the following two items I wanted to leave somewhere regarding usage.

1. the SVA now supports saving in SNP, or S parameter format.  It saves as S1P.  This is very handy for bringing the VNA sweep of a network to an external program for massaging.  This is not in Siglent docs nor in firmware revision guide.

With a quality cable attached and calibrated to the end, attached a small antenna eval board for a response plot, with or without markers.  When saved as STA or CAL, neither one could load back the working state and screen appearance.

2. Enter the CSA save format.  This is not in Siglent docs nor firmware revision guide.  CSA saves calibration AND state combined.  When you load a CSA file, you will return to where you left off, with cal in place, entire setup in place including markers, etc. 

Also, if you find something that is not in Siglent docs, look at their manual for their higher end model, the SNA5032A.

If there is a thread dedicated to usage, please move this there.

[tautech]

Discovered the following two items I wanted to leave somewhere regarding usage.

1. the SVA now supports saving in SNP, or S parameter format.  It saves as S1P.  This is very handy for bringing the VNA sweep of a network to an external program for massaging.  This is not in Siglent docs nor in firmware revision guide.

With a quality cable attached and calibrated to the end, attached a small antenna eval board for a response plot, with or without markers.  When saved as STA or CAL, neither one could load back the working state and screen appearance.

2. Enter the CSA save format.  This is not in Siglent docs nor firmware revision guide.  CSA saves calibration AND state combined.  When you load a CSA file, you will return to where you left off, with cal in place, entire setup in place including markers, etc. 

Also, if you find something that is not in Siglent docs, look at their manual for their higher end model, the SNA5032A.

If there is a thread dedicated to usage, please move this there.

https://www.eevblog.com/forum/testgear/siglent-sna5000x-two-and-four-port-vnas-coming/

[EE-digger]

Nice looking units. 

And they do either directly load or import Agilent/Keysight cal kit files. With full editing and custom kits.  The basics of this should have been carried into the SVA. 

I can't understand why I don't have a line of people to buy my KS PNA for $6k 

[tautech]

Nice looking units. 

Totally next level and some.

And they do either directly load or import Agilent/Keysight cal kit files. With full editing and custom kits.  The basics of this should have been carried into the SVA. 

Reasonably sure they do but need get mine out to confirm ....when I get a chance, please bare with me....

Have you watched the TSP vid ?
https://youtu.be/6vHX0moZUFg

[EE-digger]

The Siglent F604MS says it's similar to the Agilent 85033E that I have.

I ran a cal and not only is it similar, it appears to be exactly the same 

Would you do me a favor and post the F604MS kit cal constants, for the male connector.  This would include OPEN, SHORT, LOAD and possibly THRU.  Critical numbers are the delay in ps or length,  C0, C1, C2, C3  AND L0, L1, L2, L3

While this doesn't fix the internal 85033E kit, it raises hope that the F604MS is an easy way to call up the required constants.  May get lucky and find the same for the N type kit (85032x).

[EE-digger]

On a converted SVA1032X, MARKER FUNCTION does not exist in Windows remote control mode.

If you establish markers for bandwidth of an antenna for instance, they will be visible on the remote screen, as will MARKER FN but as soon as you leave that soft menu, you can't get back to it.

Thank you to anyone who can verify and turn it in.

[tautech]

Damaged SVA1032X repair

Factory replacement PCB's are only available as a calibrated set and there are some differences to earlier HW as now both RF PCB's are sandwiched between a common shield.
 
Nevertheless to swap out/in is quite straightforward by removing all the RF PCB screws marked 1 plus one other short screw near the ribbon cable sockets.
The Data PCB has different length screws so care need be taken to lay them out in a manner so they can be restored to original positions.

The only parts that follow into reinstallation are the 5x SMA and SMB coax cables and the 2 ribbon cables plus the power switch, a silicone button with contact membrane on the underside of the Data PCB < easy to remove and refit.
And of course the screws.....many of them.

[tautech]

........... anyone spot which cable I missed reconnecting ?
And yes I had the rear tin work on for the PSU and it was immediately obvious which after powering ON.   

[tautech]

On a converted SVA1032X, MARKER FUNCTION does not exist in Windows remote control mode.

If you establish markers for bandwidth of an antenna for instance, they will be visible on the remote screen, as will MARKER FN but as soon as you leave that soft menu, you can't get back to it.

Thank you to anyone who can verify and turn it in.

Not sure I see what you do. 
SVA1032X screenshot from the webserver attached.

[DL4JY]

Siglent released a new firmware version - did someone apply this version to an upgraded SSA? Any issues ?
Version SVA1000X Firmware_V3.2.2.6.2R5 (Release Date 06.28.24 ).

merci

[tautech]

Siglent released a new firmware version
Version SVA1000X Firmware_V3.2.2.6.2R5 (Release Date 06.28.24 ).

21 MB
https://int.siglent.com/upload_file/zip/firmware/Spectrum_analyzer/SVA1000X_V3.2.2.6.2R5_EN.zip

Release notes
Notice:
This firmware must be upgraded from V3.2.2.4.0 or later. If your analyzer has an earlier version, please update to V3.2.2.4.0 first and then proceed to update to the latest version.

1. New Functionality:
 Add Picture Viewer
 Add Open Source Acknowledge
 Add Remote Control Mode, when entering, the interface displays system time, limit status, system messages, etc., shielding UI refresh. SCPI: DISPlay:WINDow:ENABle ON|OFF
 Save the configuration every 60 seconds, and load it when Power On is turned Last
 SA/VNA mode Limit Setup adds Fail to Save, which automatically saves screenshot and trace data when the Limit Test result is Fail. It can only be used when Fail to Stop is turned ON

2. Improvements:
 Preset optimization: When pressing the Preset button, it will first jump to the Preset menu instead of directly performing Preset; Touch Assist's Preset adds preset option
 Optimization System Info display content
 When taking screenshots, do not capture the message indicating successful screenshot
 Limit the length of numbers entered on the screen and keypad
 Add SCPI commands to read system messages:SYSTem:MESSage?
 Added screenshot function under the Help window
 Set the FFT menu to invalid when the SA mode cannot select FFT
 SA mode can open the Freq Counter for markers other than marker 1
 The SA mode peak table displays the peak of the current trace
 The SA mode peak table displays the peak of the current trace, rather than the current marker’s trace
 The dBc parameter range of the SA mode OBW is changed to negative numbers
 When Noise Marker is turned on in SA mode, set Detect Type to Average
 In SA/MA/EMI mode, Wait for Trigger is displayed only if Trigger is not triggered for 2 seconds.
 Add Display Line in SA/VNA/DTF/MA/EMI mode
 When using markers in SA/VNA/DTF/MA/EMI modes, display the marker info enlarged in the upper left corner of the screen
 SA/EMI mode display more information of limit fail and pass
 Return more marker information when using SCPI command :CALCulate#:MARKer#:Y? in SA mode
 The MA mode supports the SCPI command: TRACe#[:DATA]? to get trace data
 SCPI command for setting marker X in MA mode: TRACe#:MARKer#:X supports specifying time units

3. Solved Issues:
 Fix the problem of the inconsistency between the numerical accuracy displayed in SA mode Freq menu Ref Offset and the status bar on the left
 Fix the problem of Marker display unit error when Noise Marker is turned on in SA mode
 Fix the problem of the SCPI command TRACe#:MODE in the SA mode setting trace type, which causes a crash when the selected trace exceeds the range.
 Fix the problem of not adding Freq Offset to the marker x value in SA mode
 Fix the problem of Correction errors when switching between X-Scale Lin and Log in SA mode
 Fix the problem of inconsistent units and settings for values obtained by SCPI commands :DISPlay:WINDow:TRACe:Y:RLEVel? in SA/EMI mode
 Fix the problen where Sweep Time may be incorrect when set EqLPF to Auto in MA mode

[Martin72]

Hi,
I'm very much leaning towards getting an SVA1015X instead of using the well-known "trick" with the SSA3021Xplus.
But one of the things holding me back is that Siglent's affordable SA and VNA ranges are already quite old (2018).
Not that new successors will be released next year...

[TERRA Operative]

Siglent released a new firmware version - did someone apply this version to an upgraded SSA? Any issues ?
Version SVA1000X Firmware_V3.2.2.6.2R5 (Release Date 06.28.24 ).

merci

I applied it just now, seems to work fine.

[james38]

As I mentioned in the SSA3000x/plus Thread

What worries me a little is this function:
"Save the configuration every 60 seconds, and load it when Power On is turned Last".

Why does it have to be saved every 60 seconds?

My fear is that if it is saved in flash, the memory cycles can be greatly reduced.
Flash memories also have a limited number of writes.

I once had a similar problem with a router that was constantly writing to the flash
and then stopped working after half a year.
After that, the option was removed from the firmware because there were too many failures.

Has Siglent provide any further information about what this is all about?

[Dwaine]

Siglent released a new firmware version - did someone apply this version to an upgraded SSA? Any issues ?
Version SVA1000X Firmware_V3.2.2.6.2R5 (Release Date 06.28.24 ).

merci

I applied it just now, seems to work fine.

Same here.  Working great.

Dwaine

[tautech]

As I mentioned in the SSA3000x/plus Thread

What worries me a little is this function:
"Save the configuration every 60 seconds, and load it when Power On is turned Last".

Why does it have to be saved every 60 seconds?

My fear is that if it is saved in flash, the memory cycles can be greatly reduced.
Flash memories also have a limited number of writes.

I once had a similar problem with a router that was constantly writing to the flash
and then stopped working after half a year.
After that, the option was removed from the firmware because there were too many failures.

Has Siglent provide any further information about what this is all about?

All/any boot settings are from flash memory.

Not that I know exactly but I strongly suspect the FW settings save changes are to reduce/minimize flash memory wear, however this is not an ongoing problem as I have yet to encounter it with any of the SVA/SSA models.

[james38]

All/any boot settings are from flash memory.

Not that I know exactly but I strongly suspect the FW settings save changes are to reduce/minimize flash memory wear, however this is not an ongoing problem as I have yet to encounter it with any of the SVA/SSA models.

Maybe I misunderstood you, but I don't see that saving every 60 seconds reduces/minimizes writing to the flash memory.
Quite the opposite.

It is also listed as a "new functionality" and not a fix.
How were the parameters saved before, but only when they were changed or not?

Well, it is clear that there are no problems now.
This only becomes noticeable after a certain time when the flash cycles of the module are coming to an end.
I had this effect with a router before and after about half a year or three quarters of a year the module was suddenly defective.
Only then was the problem recognized and the firmware adjusted.

As I said, I am very skeptical about this and will not flash the update for the time being until something more detailed is known about this point.
The device is too valuable to me for that.

[tautech]

Maybe I misunderstood you, but I don't see that saving every 60 seconds reduces/minimizes writing to the flash memory.
Quite the opposite.

I'll see what I can find out.....if not immediately, in a few weeks when I visit HQ.

However, I personally deal with many NAND failures worldwide and have a multitude of recovery packages to repair problems with NAND management which are getting less by the day.
I can in no way see Siglent changes to SVA/SSA SW impact on NAND wear as they are very conscious of these issues.

I like you highly value SVA and it's my fav RF instrument yet have never had or seen problems with NAND wear in any Siglent analyzers.

[james38]

Thank you very much for your effort.
I think I won't be the only one who will be interested in this.

[steve1515]

I'm also interested to know more about the issue with writing to the flash every 60 seconds. This doesn't seem like a good idea.

[shabaz]

Just speculating, (meaning, I have nothing to base this on apart from typical engineer experience!) maybe it's a release notes issue, for instance, maybe the code saves the config to NVM periodically if there have been changes, i.e. if a user adjusts a knob, then a flag is set, and the config will be saved within 60 seconds of the change (or changes) being made; or, the config is compared and then stored within 60 seconds if there is a difference. Most developers are conservative, they won't want to be changing things (i.e. changing NVM contents) unless they have to.

[tautech]

I'm also interested to know more about the issue with writing to the flash every 60 seconds. This doesn't seem like a good idea.

Let's revisit the FW release notes:
Save the configuration every 60 seconds, and load it when Power On is turned Last

If boot is set to default ^^^ is not applicable to you.
However I prefer booting to last used settings and always have.
We can alternatively use Preset to return to Factory Default or customise Preset to recover our preferred settings.

All/any of this is done from flash memory.
I don't foresee problems.......

[james38]

Only Siglent itself knows how to interpret the sentence.

I interpreted it like this:

Save the configuration every 60 seconds.
Load it when Power On is turned Last

For me that means that it is always saved every 60 seconds.
Only if the parameter "Power On " is set "turned Last" will this permanently saving config be loaded.

And yes, all of that can happen in the flash,
but only if it is necessary or if changes are made to the config, but not across the board.
That is the point that worries me.

But let's see what new information we will hopefully get on this.
Maybe it is just poorly worded and meant completely differently.