Siglent SDS800X HD 12 bit DSO's

[tautech]

What is so nice about this feature?

Have accurate NIST time stamp on this low cost DSO

Best,

What use case could this be useful?

I would have been happy with an inaccurate time stamp, without having it to connect to a LAN.

General screenshot documentation.

System fault finding.

Single trigger capture of any event, locally or remotely where LAN connectivity becomes invaluable to provide a real timestamp.

[eTobey]

I would have been happy with an inaccurate time stamp, without having it to connect to a LAN.

A battery and RTC add extra cost to the BOM, while the LAN is built in anyway and hence NTP time comes for free.

I would be happy to pay one or two bucks for the battery, and would trade some other feature with a RTC, or pay 5 bucks mor for a RTC.

I am just dissapointed, that the power button stays lit, but the time is lost... I would not mind, if the time is lost, when the plug has been pulled.

I wonder if there is a way to set the time with USB.

[ebastler]

I wonder if there is a way to set the time with USB.

The documentation states that there is a SCPI command, :SYSTem:TIME. But sending SCPI via USB means using NI-VISA, I believe? That's a mouthful just for setting the clock...

Edit: What's wrong with a network connection? Or did you take the scope with you on your camper-van vacation after all? 

[eTobey]

What use case could this be useful?

I would have been happy with an inaccurate time stamp, without having it to connect to a LAN.

General screenshot documentation.

System fault finding.

Single trigger capture of any event, locally or remotely where LAN connectivity becomes invaluable to provide a real timestamp.

I have a csv file of a capture, but there is not a date or time in it. How can i make it include the date/time?

Edit: What's wrong with a network connection? Or did you take the scope with you on your camper-van vacation after all? 

Of course not! Its not something i like that much. Some times you just dont have a network. But a USB dongle could the a cructh solution for this scope.

[KungFuJosh]

Of course not! Its not something i like that much. Some times you just dont have a network. But a USB dongle could the a cructh solution for this scope.

You don't need the USB dongle. A WiFi bridge adapter is actually a much better solution. Primarily because the previously supported USB WiFi dongle was outdated WiFi. With a WiFi bridge adapter, you can get much better speeds depending on which one you buy, and you won't be limited to whatever USB dongle is chosen.

[ebastler]

A WiFi bridge adapter is actually a much better solution.

Well, an external bridge is "better" in terms of enabling higher WiFi bandwidth. But a dongle is better in terms of "less clutter, no extra cables, keeps the scope easily transportable".

Bandwidth is not really critical for the amounts of data one deals with, and the transmission speed of the scope, I'd say. To get network access at the bench, I am currently using an old 2.4-GHz-only router which I have repurposed as a bridge under OpenWRT. It does the job just fine, but isn't helpful when I need to carry the scope to another room occasionally.

It's a nice and compact scope, easily transported, so it would be nice if it had a compact and clutter-free Wifi solution too.

[KungFuJosh]

A WiFi bridge adapter is actually a much better solution.

Well, an external bridge is "better" in terms of enabling higher WiFi bandwidth. But a dongle is better in terms of "less clutter, no extra cables, keeps the scope easily transportable".

That entirely depends on what you choose to use as a bridge. There are tons of devices like this that you can attach with velcro or double-sided tape: https://www.amazon.com/dp/B01198RNYW

There are plenty of only slightly larger ones with better specs. Personally, my scope sits in one spot (on ethernet) and never moves, but to each their own. 🤷

[mawyatt]

Agree, it would be nice if WiFi were enabled. We use an RJ45 connection, but don't need to move the scope....yet.

In fact would be nice if all our TE were WiFi enabled, however as an unintended consequence could be additional RFI which may pose a problem with certain delicate measurements. We already see "effects" from our wireless router and phones.

Best,

[KungFuJosh]

That's one reason why I prefer ethernet. I love having everything connected, but I already have 1000 wifi devices in my house. I have plenty of stray noise and stray AC that I don't want to add more noise in my office. Though I'm not sad that my scope has built-in wifi if I ever need it.

If it won't overtax the 800's limited resources, I think they might eventually add support for a USB dongle, if for nothing other than to get Rob to stop asking. 😉

[BillyO]

A little off topic, but I have very little use for WiFi beyond truly portable devices like cell phones or laptops and those few stupid IOT things we have around here.  Even my laptops get connected to Ethernet if they are going to reman in position for more than a day.

[2N3055]

A little off topic, but I have very little use for WiFi beyond truly portable devices like cell phones or laptops and those few stupid IOT things we have around here.  Even my laptops get connected to Ethernet if they are going to reman in position for more than a day.

Same here.
I positively suffer WiFi only if I have to.

[Furna]

I received the SDS824X HD just yesterday ... until now vey happy.

Surprised to find it bound to a fixed IP address instead of DHCP and no MAC address label on the back of the chassis; not really an issue.
Just a couple of touch on the screen and switched to DHCP and easily found it's IP addres on the network.
From the web interface found the MAC address and configured a reserved IP address on DHCP server and FQDN on the DNS.
Obviously configured the DHCP to also respond with appropriate NTP server's IP addresses.
Updated the firmware to 1.3.6 (arrived with 1.3.3)
Compensated the 4 probes ... the most difficult task was to remove the color rings from the probes to replace them with the appropriate ones.
Unfortunately, even if DNS server is correctly set, the SMB client seems to not understand FQDN and want an IP address ...
After 2 and an half hours here it is my first picture taken from the scope.

I now have 2 points where maybe someone can help me:

- Maybe the question have been asked already but I was not able to find it and also searching the manual result in no luck ...
  Is there a way to have the "Default" button to set the probe attenuation to 10X? I will very rarely move the probe's switch to 1x and going to the channel menu every time to bring it back to 10x is annoying.

- I didn't understand the BNC accessory adapter provided with PP215 probe ... the manual says "above 200MHz" but it doesn' fit the probe mechanically ...

Help on first point is very apreciate.

Cheers

[2N3055]

In manual :

https://www.siglenteu.com/wp-content/uploads/dlm_uploads/2024/02/SDS800X_HD_UserManual_EN01A.pdf

Page 208.

You can save full setup of scope in config file. You can have many of those.
And you can set Default button to load one of those..
See "To default Key", page 209.

BNC adapter gets fit over tip and ground part of probe tip. You remove hook plug from top, to expose tip..

Nice article:
https://www.digikey.com/en/articles/understanding-selecting-using-passive-oscilloscope-probes

[eTobey]

What use case could this be useful?

I would have been happy with an inaccurate time stamp, without having it to connect to a LAN.

General screenshot documentation.

System fault finding.

Single trigger capture of any event, locally or remotely where LAN connectivity becomes invaluable to provide a real timestamp.

I have a csv file of a capture, but there is not a date or time in it. How can i make it include the date/time?

I could not find a option, to add a timestamp to the acquired waveform. It seems, this is not possible?

[awakephd]

I feel like I have read a posting somewhere in one of these threads regarding some "pickiness" with which USB drives the SDS800X HD series can use, but I can't seem to find it. Did I imagine this? I ask because I have two identical thumbdrives - identical as in they came out of the same package, show up with the exact same partitioning and formatting on my laptop, but the SDS804X HD that I recently got won't recognize one but does recognize the other (e.g., as a destination to save to). If anyone can point me to a solution or more info, I would be most grateful!

On edit: I apologize for not providing more information:

SDS804X HD
Firmware 1.1.3.6
Front-panel USB (haven't tried the back)

[tautech]

I feel like I have read a posting somewhere in one of these threads regarding some "pickiness" with which USB drives the SDS800X HD series can use, but I can't seem to find it. Did I imagine this? I ask because I have two identical thumbdrives - identical as in they came out of the same package, show up with the exact same partitioning and formatting on my laptop, but the SDS804X HD that I recently got won't recognize one but does recognize the other (e.g., as a destination to save to). If anyone can point me to a solution or more info, I would be most grateful!

On edit: I apologize for not providing more information:

SDS804X HD
Firmware 1.1.3.6
Front-panel USB (haven't tried the back)

Yeah well a few months back I sourced 5x cheap 8GB USB sticks primarily as they had LED's.
4 worked fine and one went in the trash.

Historically 8GB FAT32 with 4k clusters worked with any Siglent instrument however some 32GB sicks recently sourced from Siglent have been working fine with everything I have tried them in.

A Quick Format to FAT32 with 4k clusters in W10 File Explorer has always been a reliable starting point.

[awakephd]

Thanks, I'll give that a try. These are PNY brand - middle of the road, maybe? The package is marked 16G, but they report 32G. Hopefully they are legitimate, and Amazon did not sell me fakes ... !

[KungFuJosh]

These 32GB drives have worked on all my USB capable TE without issue: https://www.amazon.com/gp/product/B09FFK1QLR/

Those are currently installed in 4 Siglent devices, since I sold the 5th one recently.

[eTobey]

What use case could this be useful?

I would have been happy with an inaccurate time stamp, without having it to connect to a LAN.

General screenshot documentation.

System fault finding.

Single trigger capture of any event, locally or remotely where LAN connectivity becomes invaluable to provide a real timestamp.

I have a csv file of a capture, but there is not a date or time in it. How can i make it include the date/time?

Since i have not received an answer, i assume, that there is no way to have a timestamp in the csv file of a capture.

There is this "super nice feature", of a NIST time stamp, that has an accuracy of 10ns, but then i have to get out pencil and paper to note down the time a waveform has been captured? If this is true, then it deserves not one, but two facepalm smileys.   

Is there actually a case where this "accuracy" is used?

[eTobey]

I feel like I have read a posting somewhere in one of these threads regarding some "pickiness" with which USB drives the SDS800X HD series can use, but I can't seem to find it. Did I imagine this?

I can tell you, that the SDG1032X is picky too. 1 out of 4 sticks worked. I have to say, those were some kind of old 4gb-ish ones.

[ebastler]

There is this "super nice feature", of a NIST time stamp, that has an accuracy of 10ns, but then i have to get out pencil and paper to note down the time a waveform has been captured?

I don't think NTP time claims to be anywhere near that accuracy. Try 10 ms for time obtained via internet.

The time of saving any type of data (traces, setups, protocol decodes...) is captured in the file's time stamp. That works well for me -- helps me to identify the right file, put files in the right order etc. It would be nice if the time stamp could also be part of the auto-generated file names, to reduce the risk of messing with it when copying files around.

[awakephd]

I checked again on the two PNY USB2.0 32GB thumb drives (one that the scope recognizes, and one that it doesn't); they were formatted exactly the same, with 16K clusters. I reformatted both to use 4K clusters. And ... still the same results. The same one that the scope recognized before, it continues to recognize; the one that it didn't recognize before, it still doesn't recognize.

Maybe there is something finicky in the timing, such that slight differences in the internal components makes a difference in whether or not it works ... ?? Very odd, since USB has been around for so long - this seems like it ought to be a plug-and-play component. Unless, perhaps, to save a dollar, they chose not to use a standard USB interface IC, and instead are handling it in software ... In which case, one could hope for a software update to fix it. But then again, since some USB drives work, there may not be much sense of urgency to address this.

Ah, well - for now I have a USB stick that does work with the scope ... and I don't actually have anything that I need to store yet! Still experimenting and learning ...

[awakephd]

The time of saving any type of data (traces, setups, protocol decodes...) is captured in the file's time stamp. That works well for me -- helps me to identify the right file, put files in the right order etc. It would be nice if the time stamp could also be part of the auto-generated file names, to reduce the risk of messing with it when copying files around.

I have not yet experimented with saving .csv files from the scope, so I may be way off base in what I am about to say ... but FWIW I would not expect a time stamp to be included in a .csv file, since that could confuse whatever program is trying to read the data (a line of text that doesn't fit the .csv structure). I would think it more likely to have a time stamp included as part of the file name, much the way my Android phone names pictures that I take with its camera. Again, I am so new to DSOs that my comment might be way out of line - my perspective is based on writing and using other sorts of software that use .csv files to capture/transfer data (e.g., various sorts of database software).

[bateau020]

Just so to make it clear what happens with the .csv files: when using csv dump instead of image dump when pushing "save", the file name is like "seg{n}.csv" with n is a 5 digit number. The header in the csv file is like:

Code: [Select]

Record Length,Analog:400000
Sample Interval,5.000000E-10
Vertical Units,V,
Vertical Scale,+5.000000E-04
Vertical Offset,+0.000000E+00
Horizontal Units,s
Horizontal Scale,+2.000000E-05
Model,SDS824X HD
Serial Number,**REDACTED**
Software Version,3.8.12.1.1.3.6
Source,CH1
Second,Value
-1.0000000000E-04,+2.187500E-04
-9.9999500000E-05,+2.052083E-04
-9.9999000000E-05,+2.208333E-04
-9.9998500000E-05,+2.739583E-04
.....

This will also come with a "metadata file" called "C1.awg" that I imagine being for feeding into a AWG, with the following content:

Code: [Select]

sample_rate,2000000000.000000
amplitude,1
run_mode,countinuous
wait_trigger_type,zero
burst_count,1
burst_hold_type,Start Value
trigger_mode,button
timer_interval,0.001
trigger_edge,raise
interp_mode,zero_hold
length_strategy,length
increasing_strategy,interpolation
decreasing_strategy,decimation
file_type,SDS

segment_0,seg00001.csv,-0.000001,0.004259,400000,1,-1,next,0,0
segment_1,seg00002.csv,0.000300,0.002048,400000,1,-1,next,0,0


Not aware of these files following an industry wide standard, but it looks like it is a Siglent inter-device standard.

So: no dates or times, other than relative time.
I do think that adding dates in either the csv file or in the "metadata" file would be a nice addition.

[awakephd]

Just so to make it clear what happens with the .csv files: when using csv dump instead of image dump when pushing "save", the file name is like "seg{n}.csv" with n is a 5 digit number. The header in the csv file is like:

Code: [Select]

Record Length,Analog:400000
Sample Interval,5.000000E-10
Vertical Units,V,
Vertical Scale,+5.000000E-04
Vertical Offset,+0.000000E+00
Horizontal Units,s
Horizontal Scale,+2.000000E-05
Model,SDS824X HD
Serial Number,**REDACTED**
Software Version,3.8.12.1.1.3.6
Source,CH1
Second,Value
-1.0000000000E-04,+2.187500E-04
-9.9999500000E-05,+2.052083E-04
-9.9999000000E-05,+2.208333E-04
-9.9998500000E-05,+2.739583E-04
.....

This will also come with a "metadata file" called "C1.awg" that I imagine being for feeding into a AWG, with the following content:

Code: [Select]

sample_rate,2000000000.000000
amplitude,1
run_mode,countinuous
wait_trigger_type,zero
burst_count,1
burst_hold_type,Start Value
trigger_mode,button
timer_interval,0.001
trigger_edge,raise
interp_mode,zero_hold
length_strategy,length
increasing_strategy,interpolation
decreasing_strategy,decimation
file_type,SDS

segment_0,seg00001.csv,-0.000001,0.004259,400000,1,-1,next,0,0
segment_1,seg00002.csv,0.000300,0.002048,400000,1,-1,next,0,0


Not aware of these files following an industry wide standard, but it looks like it is a Siglent inter-device standard.

So: no dates or times, other than relative time.
I do think that adding dates in either the csv file or in the "metadata" file would be a nice addition.

As I said, I might be way off ... and clearly I was! If there is a header in the .csv file, then there is no reason it could not include a timestamp; same with the metadata file. Ah, well ...