Siglent SSA3000X and SSA3000X-Plus Spectrum Analyzers

[rch]

I don't think it is a bug.  The port is previously indicated in TCPIP but I don't know how to handle that.  Wireshark shows that well. And VISA works, too.

If you're connecting via telnet you need to either to send the commands you want to use over telnet, or run a program which resides  on the scope via telnet or make a new connection with a different program from your computer.  In the latter case, if it uses the same port as telnet you need to close telnet first. The new program will make a new connection with a different return port.  I don't think this is a bug.

[Bicurico]

Hi,

I wasn't home, so here a more complete answer.

I used Wireshark to sniff the network traffic between my PC and the SSA3000X, when using EasySpectrum, which requires NI VISA.

The SCPI communication starts at port 111 and/or 897 on the PC side.
However, any reply from the SSA3000X goes to port 4xxx. This port is different on every new connection and is told to the PC through what I imagine is a lower level protocol.

The problem I have with this, is that I haven't found a way in VB .net to read that incoming port number. So I don't know which port I should listen to.

Using Wireshark, I can see the raw TCP/IP packets and yes, there is the incoming port number. But how do I access it? No idea!

NI VISA supports such protocol and I managed to test communication under VB .net using NI VISA (hint: you need to install VISA with all options - somehow I missed the newer .Net runtimes on the first install).

But it kind of sucks having to install 600MB of software to then run a 200kB VB .net executable!

So what I am asking Siglent: please make SCPI work with a single fixed port number! The standard port number would be great, which is 5025.

Make it work in both directions and to test it, one could then just use "telnet 192.168.1.2 5025" to access the SSA3000X (which has IP 192.168.1.2 in this example).

As it is right now, you cannot telnet, because while you can SEND SCPI commands on port 111 or 897, the telnet session will NOT receive any incoming data on those ports! You can still verify the truth of what I am saying, because doing a telnet to the SSA3000X on port 111 and typing "*IDN?" will cause the device to show "REMOTE" on the screen and lock all keys except the ESC key (used to close the remote connection).

You won't see and reply, though, as the SSA3000X will reply to a different port.

Using one single port is perfectly doable and Rigol does that (at least with the DS1054Z I own: I can telnet, type commands like "*IDN?" and do get a propper reply). However, Rigol chose to use port 5555.

I have another professional device with SCPI support, that behaves like the Siglent. Again, that is not a bug and is according to TCP/IP specification. But it does prevent people like me to use it without NI VISA.

Hope this clears all up!

Regards,
Vitor

[nctnico]

First use a telnet client (the telnet command) and then write code! A component which sets up a tcp/ip connection should do the trick.

[Bicurico]

@nctnico: please read what I have written. YOU CANNOT USE TELNET FOR SCPI WITH THE SSA3000X, which is why I was asking Siglent to change the behaviour, despite the current implementation not being a bug.

Please try yourself and you will see what I mean.

Regards,
Vitor

[rf-loop]

First use a telnet client (the telnet command) and then write code! A component which sets up a tcp/ip connection should do the trick.

Is is better that you start first reading TCP/IP Instrument Protocol Specification VXI-11.
But good luck with windows telnet client.

It is totally other game  IF  Siglent implement telnet or what ever  based simple 1 host - 1 device SCPI communication channel for some simple PC2SSA use for some hobby fun. 

At this time there is VXI-11 and there communication need do as VXI-11  is specified.

Here very short example (and note for @ntcnico: without any detected bugs!)
There can also see ports and how it build link.

[nctnico]

The instruments I have seen so far (including my Advantest spectrum analyser and Keysight multimeter) support SCPI over telnet (which is super simple to do) but for the SSA3000X it seems you have to implement the complicated VXI-11 protocol then. It seems Siglent did things the hard way by implementing a protocol which isn't even widely used.

Edit: where I'm hinting at: Bicurico has made some very nice software for a cheap 'Ebay' spectrum analyser. People like that can create an ecosystem around an instrument which can provide some additional sales. For that to happen communication with an instrument has to be simple without needing a specific version of the many megabyte VISA package from NI or having to implement complicated protocols which basically transport SCPI commands.

[tautech]

Anybody seen this and can offer advice to a new SSA3kX owner ?
https://www.eevblog.com/forum/testgear/ssa3021x-spur-345-mhz/

[tautech]

Recent question about 50 and 75 input impedance selection:
https://www.eevblog.com/forum/testgear/hack-of-sigllent-spectrum-analyzer-ssa3021x/msg1103377/#msg1103377

[alexo]

In the screenshot underneath the spectrum analyzer is "looking" at it's 50 ohm terminated input between 0 and 2Mhz. The first 600Khz or so are very high. Is this normal ? Shouldn't be flat(ish) from 9Khz ?

[TurboTom]

The shape of the noise looks like the phase noise component of the 0Hz peak (here the 1st LO will have the same frequncy as the first IF). I haven't got my SSA available right now so I cannot verify your finding, but the Rigol DSA815 doesn't show this behavior (see attachment).

You may check this closer by setting center frequency to 50kHz. If there's a "dip" of the noise around 20kHz and towards higher frequencies, the noise increases again, the I'm almost certain that it's phase noise.

Cheers,
Thomas

[alexo]

Here it is...

[DL4RAJ]

In the screenshot underneath the spectrum analyzer is "looking" at it's 50 ohm terminated input between 0 and 2Mhz. The first 600Khz or so are very high. Is this normal ? Shouldn't be flat(ish) from 9Khz ?

As Turbo Tom says it is the phase-/sideband noise of the LO at zero Hz.
I've measured the  sideband noise with -98.36dBc/Hz at 10kHz offset,
see https://www.eevblog.com/forum/testgear/hack-of-sigllent-spectrum-analyzer-ssa3021x/msg1065022/#msg1065022
which means around -83dBc/Hz in 300Hz RBW.
When you look at the plot of my SSA below you'll see that the noise is about this level around 10kHz.The noise of your SSA is a bit higher due to normal sample variations.
Regards

[alexo]

Thank you very much. I like this particular machine a lot as it is very quiet and almost spur free in particular around 500Mhz, the zone I am interested in. I really hated the thought of having to have it replaced or repaired.

[videobruce]

I hope no one is in a hurry, there won't be any until March 1st according to Siglent;
.

all our units been selling fast for this item and we checked with the warehouse in SOLON OHIO, and they have been experiencing delays due to the Asian holidays.

Many items delay since China Holiday
 Ruby Wu
Siglent Technologies America Inc.
Ruby.wu@siglent.com
440-3985800x103

[Chig]

I've spoken with TrioTest in Australia too, they are in much the same boat - they have *one* demo unit (fully unlocked for all features), but no stock of any new units for another ~4 weeks.


Chig.
(long LONG time reader, first time poster...)

[tautech]

I've spoken with TrioTest in Australia too, they are in much the same boat - they have *one* demo unit (fully unlocked for all features), but no stock of any new units for another ~4 weeks.


Chig.
(long LONG time reader, first time poster...)

Welcome to the forum.

Thanks for that, just asked the factory if they have stock and will come back with an edit. they are under pressure to match demand.
For Kiwi's I have one 3021X plus my 3032X demo on hand.

[videobruce]

Two things, the 1st is consistent:

Using the Tracking generator; if the 'Units' (under Amplitude) are set to anything other than dBm, the Normalize function will not work. If you remember, on the Rigol, when the Units are set to something other than dBm, theirs will automatically change the setting back to dBm. On this it doesn't, it just won't allow a normalize function to work even thou the TG function is 'on'.


There is a problem powering off my unit, but I don't have a handle on it as of yet, there doesn't seem to be a pattern. Even after I did a reset, the problem manifested itself the 1st powerup/down cycle. The next time, it powered off ok. 

More than once, I had to hold the power button down for as long as four seconds before it powered down. There is never a problem powering back up. I'm trying to narrow it down, but haven't been able to. I noticed this right off the bat upon receipt and initially though it had a delayed off where one had to hold the button purposely.

[videobruce]

On a side note, there is a F/W update on their site available released today (2/23/17);
http://siglentamerica.com/prodcut-gjjrj.aspx?id=5113&tid=227&T=2

1. This firmware must be upgraded from V1.2.8.1 or later. If your SSA3000X has an earlier version, please to V1.2.8.1 first and then proceed to uplate to the latest version.
 2. Add socket/telnet-scpi, you can operate remote control through LAN port without NI-VISA installed. Please check SSA3000X Programming Guide for more details.
 3. Change two limit lines in different colours.
 4. When quit from TG, stay in“Auto”sweep.
 5. Lower some spurs when FFT.
 6. Improve the stability.

[Bicurico]

Hi,

This new FW is actually the answer of Siglent to a request I made in this thread (they didn't do it for me, of course - at least I wouldn't think so):

It allows SCPI commands through a regular telnet connection. This opens doors to people like me to program custom applications that use the SSA3021X remotely.

Without any promise, imagine this: a software similar to my "VMA Simple Spectrum Analyser" (made for the cheap ADF4350, ADF4351, MAX2870 range of devices, like the NWT4000 - discussed in a different thread) in a special version for the SSA3021X, offering all the "advanced" measurements, but based on measurements of a propper spectrum analyser!

More info about my software: http://vma-satellite.blogspot.pt/2016/12/vma-simple-spectrum-analyser-download.html

Thumbs up for Siglent!

Regards,
Vitor

[tautech]

Hi,

This new FW is actually the answer of Siglent to a request I made in this thread (they didn't do it for me, of course - at least I wouldn't think so):

It allows SCPI commands through a regular telnet connection. This opens doors to people like me to program custom applications that use the SSA3021X remotely.

Without any promise, imagine this: a software similar to my "VMA Simple Spectrum Analyser" (made for the cheap ADF4350, ADF4351, MAX2870 range of devices, like the NWT4000 - discussed in a different thread) in a special version for the SSA3021X, offering all the "advanced" measurements, but based on measurements of a propper spectrum analyser!

You will see more of this in Siglent products in the near future....but it's best left to be discussed in other threads when it happens. 

[dpenev]

Hello,

I am after accurate noise measurements.
I have applied flat White Gaussian noise -120dBm/Hz to the analyzer.
I have done two measurements with and without pre-amplifier and I get about 0.35dB difference.
See attached picture. Probably this PA offset can be fixed?
Anyone with experience with the instrument calibration?

Thanks
Dimitar

 

[dpenev]

Hello,

Another things that puzzle me.

My setup:
-120dBm/Hz flat white Gaussian noise
I have tried to measure using "Noise marker" and setting few RBWs.
Each time I keep the VBW same as RBW and average traces enough to get low variance (0.5dB)
Detector is "Average"
RBW=10KHz(100 avg traces)  Noise marker=  -120.1dBm/Hz (trace variance +/-0.5dB)
RBW=3Khz (100 avg traces)    Noise marker= -120dBm/Hz     (trace variance +/-0.5dB)
RBW=1KHz (100 avg traces)    Noise marker= -120dBm/Hz     (trace variance +/-0.5dB)
RBW=300 (3 avg traces)          Noise marker= -117.8Bbm/Hz  (trace variance +/-0.5dB)
RBW=100 (1 avg traces)          Noise marker= -118.3Bbm/Hz  (trace variance +/-0.5dB)

For the RBW 300Hz and 100Hz my measurement is about 2dB too strong. Do you think it is a bug?
Should I prefer big RBW and many trace averaging or small RBW and less  trace averaging?

I also have compared the dBm readings on the screen at the marker and what the "Noise marker" calculates
for example I see my trace at -95dBm with the RBW=300Hz  and I expect power density -95-10*log10(300)=-119.77 dBm/Hz
but the noise marker calculates it to  -117.8 dBm/Hz (2dB too strong)
It is interesting how the analyzer is calculating the "Noise marker"?

Things are getting even more confusing as I noticed that what I see on screen differs from the regular dBm marker with sometimes 1dB
See the attached picture. So I am not sure I should believe the trace and grid or the marker.

Any comments on all this is welcome.

Thanks
Dimitar
 

[nctnico]

Aren't you looking at the SA's own noise level at the higher RBW settings? Using a -120dBm noise source may not be the best way.

[Bicurico]

http://vma-satellite.blogspot.pt/2017/02/siglent-ssa3021x-new-firmware-p0802.html

 

Regards,
Vitor

[tautech]

http://vma-satellite.blogspot.pt/2017/02/siglent-ssa3021x-new-firmware-p0802.html

 

Regards,
Vitor

Exactly. 

@ dpenev
Any comments of measurement should always be made with the # firmware version in use.
The latest (8.2) was released yesterday.
Please comfirm the one you have installed.