New Sigrok/PulseView hardware support (Siglent SDS HD, Rigol DHO...)

[dpenev]

Thanks Furna,
I have  followed the procedure I have described two messages back and the log attached.
 

...
sr: scpi_tcp: Successfully sent SCPI command: ':TDIV? '.
sr: scpi: Got response: 'DAT2,#9000000000 AVEDESC1.00E+075.00E-04', length 40.
sr: session: bus: Received SR_DF_HEADER packet.
sr: scpi_tcp: Successfully sent SCPI command: 'WAV:MAXP? '.
sr: scpi: Got response: '5.00E-04', length 8.
sr: scpi: get_int: non-integer rational=500/1000000
sr: session: Could not start siglent-sds device tcp-raw/192.168.1.6/5025 acquisition.
...

'DAT2,#9000000000 AVEDESC1.00E+075.00E-04' looks a concatenation of few responses.

5.00E-04 is my current timing resolution 500us/div

WAV:MAXP?  if sent separately to the oscilloscope returns 10000000 but I don't see it in the log.

 

[Furna]

Thanks Furna,
I have  followed the procedure I have described two messages back and the log attached.
 

...
sr: scpi_tcp: Successfully sent SCPI command: ':TDIV? '.
sr: scpi: Got response: 'DAT2,#9000000000 AVEDESC1.00E+075.00E-04', length 40.
sr: session: bus: Received SR_DF_HEADER packet.
sr: scpi_tcp: Successfully sent SCPI command: 'WAV:MAXP? '.
sr: scpi: Got response: '5.00E-04', length 8.
sr: scpi: get_int: non-integer rational=500/1000000
sr: session: Could not start siglent-sds device tcp-raw/192.168.1.6/5025 acquisition.
...

'DAT2,#9000000000 AVEDESC1.00E+075.00E-04' looks a concatenation of few responses.

5.00E-04 is my current timing resolution 500us/div

WAV:MAXP?  if sent separately to the oscilloscope returns 10000000 but I don't see it in the log.

Can you try with a different resolution? I was succesfull with 10mV/div with SDS824X HD

See also
https://www.eevblog.com/forum/testgear/new-sigrokpulseview-hardware-support-(siglent-sds-hd-rigol-dho800-)/msg5605601/#msg5605601
With SDS2104X+ it is not possible to go lower than 10mV/div and fredo_ is looking at it.

[dpenev]

Seems not related with Voltage resolution. I have tested with 1V/div and still similar behavior.
| am attaching another log for someone to check. I have a feeling that it is a timing issue in the the communication.
Assuming that the software is working fine for someone else with SDS2000X+
In my case both the PC running PV and the scope are in the same LAN connected directly in the same router, 1,2m LAN cables.

Pinging 192.168.1.6 with 32 bytes of data:
Reply from 192.168.1.6: bytes=32 time<1ms TTL=64
Reply from 192.168.1.6: bytes=32 time<1ms TTL=64
Reply from 192.168.1.6: bytes=32 time<1ms TTL=64
Reply from 192.168.1.6: bytes=32 time<1ms TTL=64

[fredo_]

Hi @dpenev and thank you for your investigations !

Seems not related with Voltage resolution. I have tested with 1V/div and still similar behavior.
| am attaching another log for someone to check. I have a feeling that it is a timing issue in the the communication.
Assuming that the software is working fine for someone else with SDS2000X+
In my case both the PC running PV and the scope are in the same LAN connected directly in the same router, 1,2m LAN cables.

No it doesn't seem to be a timing issue, the problem comes from the response of the scope to the ACQ:POIN? command.
You scope gives a ".9" reply while the Sigrok (and the documentation) expects something of the form "1.25E+08".
This command returns the number of points of the current waveform. Can you check the memory depth of the scope and maybe the value returned by ACQ:POIN? SCPI command form web control ?

[fredo_]

Oh and @dpenev, you may want to try and reboot your scope (including removing the powder cable).
It happened to me a few times: I started having very strange SCPI responses from the scope, and after a full reboot, everything went back to normal  .

[fredo_]

I couldn't set the vertical sensitivity lower than 10mV/div.
Both scope can go down to 500µV/div so it seem that PulseView
doesn't support it even it is displayed in configure channel top menu.

That I should be able to fix, I'll take a look.

That would be great !

Hi there,
Here is a new version that should fix the sub 10mV vertical range issue for Siglent models : Windows (64bits) installer (zip password is 'sigrok').
08/26/2024 Update
Siglent SDS : fixed support for sub 10mV vertical ranges

[Spike]

I couldn't set the vertical sensitivity lower than 10mV/div.
Both scope can go down to 500µV/div so it seem that PulseView
doesn't support it even it is displayed in configure channel top menu.

That I should be able to fix, I'll take a look.

That would be great !

Hi there,
Here is a new version that should fix the sub 10mV vertical range issue for Siglent models : Windows (64bits) installer (zip password is 'sigrok').
08/26/2024 Update
Siglent SDS : fixed support for sub 10mV vertical ranges

Updated macOS builds available; see first post for link.

[hansibull]

Amazing work! What are the odds of getting your fork merged into the official Sigrok repo? I have been using Sigrok's libserialport library, and I even found a nasty bug that affects macOS. I clearly demonstrated the issue and even submitted a PR, but It hasn't gotten any attention from the developers.

[fredo_]

Amazing work! What are the odds of getting your fork merged into the official Sigrok repo?

Thanks !
Well I created the two PRs yesterday, now that things seem quite stable:
https://github.com/sigrokproject/libsigrok/pull/245
https://github.com/sigrokproject/libsigrok/pull/246
So wait and see !...

[Furna]

Regarding LA acquisition, I spent quite some time trying and understand how to have pagination work (data blocks are limited to 5Mpts, over that memory depth, data has to be paginated) but I came to the conclusion that the SDS2000X HD has a bug that prevents pagination to work. The result is that the first 5Mpts datablock is replicated several times on LA channels :-/
I ended up sending a mail to Siglent support to report the issue. Not sure if/when I'll get a response and this will be fixed...

Just a quick update on this , I got the following  reply from Siglent support :

This may be a bug in the software. Our R&D department is analyzing this problem. I will inform you as soon as I have the result.

Wait and see !...

Any update from Siglent ?

[Furna]

Amazing work! What are the odds of getting your fork merged into the official Sigrok repo?

Thanks !
Well I created the two PRs yesterday, now that things seem quite stable:
https://github.com/sigrokproject/libsigrok/pull/245
https://github.com/sigrokproject/libsigrok/pull/246
So wait and see !...

Let us know if we can sponsor your PR

[fredo_]

Any update from Siglent ?

Yes I had a first reply from the support team with a code snippet showing a working example of pagination on digital data, but this was not relevant since it was paginating only on a few hundred of bytes.
I replied explaining that the problem only appeared for 5Mpts+ frame, but they did not come back to me so far (this was mid-august).

Let us know if we can sponsor your PR

For now I'm trying to reach out to project committers on the dedicated Discord channel...

[hansibull]

I really hope your PRs gain some traction from the Sigrok maintainers. They cover the most popular Rigol and Siglent scopes currently on the market, making Sigrok relevant for many new users.
I used your Sigrok fork to capture data with a Siglent SDS2000X HD and decode an SPDIF signal, something no scope has built-in decoding for.

[Spike]

Amazing work! What are the odds of getting your fork merged into the official Sigrok repo?

Thanks !
Well I created the two PRs yesterday, now that things seem quite stable:
https://github.com/sigrokproject/libsigrok/pull/245
https://github.com/sigrokproject/libsigrok/pull/246
So wait and see !...

The reviewer of the siglent PR on which your work is based was github user fenugrec.  I’d be shocked if that isn’t the very same fenugrec user as here on the EEVblog forum.  Perhaps you could reach out here and see if you can get things moving a little sooner.

[wiretail]

Just wanted to add a hearty thanks for this.  One of the first things I tested with my new SDS804/824.  Works great!

Doing analog->digital convert->custom python decoder's direct from the scope, , is wonderful.  (Wherever it might be -- with a tailscale network, it's handy for remote investigation too.)

[fredo_]

Just wanted to add a hearty thanks for this.  One of the first things I tested with my new SDS804/824.  Works great!

Doing analog->digital convert->custom python decoder's direct from the scope, , is wonderful.  (Wherever it might be -- with a tailscale network, it's handy for remote investigation too.)

Thanks for the feedback @wiretail !
If you have some spare time, you might want to check the other project I'm working on: https://www.eevblog.com/forum/testgear/ngscopeclient-ultimate-acquisition-software-protocol-decoder-debug-tool/
We need testing from SDS800 owners ;-)

[fredo_]

Regarding LA acquisition, I spent quite some time trying and understand how to have pagination work (data blocks are limited to 5Mpts, over that memory depth, data has to be paginated) but I came to the conclusion that the SDS2000X HD has a bug that prevents pagination to work. The result is that the first 5Mpts datablock is replicated several times on LA channels :-/
I ended up sending a mail to Siglent support to report the issue. Not sure if/when I'll get a response and this will be fixed...

Just a quick update on this , I got the following  reply from Siglent support :

This may be a bug in the software. Our R&D department is analyzing this problem. I will inform you as soon as I have the result.

Wait and see !...

Any update from Siglent ?

I just received a confirmation from Siglent support that they could reproduce the bug and that they are working on a fix ! 

[Furna]

I just received a confirmation from Siglent support that they could reproduce the bug and that they are working on a fix ! 

Thank you for the update!