Having issues with channel one on my Siglent scope, is this hardware?

[rwgast_lowlevellogicdesin]

Ok so I bought a Siglent 25Mhz 1022DL on the forum about 1.5 years ago. When I first got it I upgraded all the firmware bought quality american probes, and compensated them. I know at that time there were no issues with the scope. For about the last 8 months Ive mostly used it for its FTT functions in AC coupled mode, so im not even sure when this issue popped up. On a freshly calibrated probe channel one is not properly displaying the 100khz square wave on the output, the rising edge is awful. I hooked a probe to the 100khz output then set both of the channels for the same voltage/timing, here is what im seeing.

Probe on channel #2


Same probe same settings on channel #1


Now here is what channel #1 looks like when the waveform is scaled to fit in the y axis


I really can't afford to replace the scope... and im really hoping this is not a hardware problem but I hit the defualt setup button a few times and I even relashed the firmware. I use my scope for the things I need so I don't know all the nooks and crannies in scope software im hoping maybe I changed some stupid setting that isn't changing back... Thinking about this issue if its hardware I don't think it is anything to terribly wrong aybe just some bad caps or something in the front end screwing up the impeadence/filtering/compensation? The thing is Ive never used this scope for anything more than 5 maybe 12v, for the most part and ive never smoked anything while testing so im not sure why anything in the front end would be messed up.

So has anyone seen something like this before? Hopefully have some good news so I can get back to work!

[tautech]

Ch 2 looks like it's working as expected.
Weird, nothing comes to mind immediately as to why Ch 1 is behaving that way.

I'll point Siglent support to this thread and hopefully they'll offer some advice.

[singapol]

What I can see is ch2 is "armed" trigger so what is triggering it, whereas ch1 is normal "triggered" ?
The measurements in pic1 and 2 is different for Vpp and Vmax.

[mij59]

Try a time base setting of 5 usec, looks like aliasing.

[Arjan Emm]

Try a time base setting of 5 usec, looks like aliasing.

Both channels should show aliasing the same way, i'd guess?

[3db]

Have you tried both probes on the working channel ?

3db

[mij59]

Try a time base setting of 5 usec, looks like aliasing.

Both channels should show aliasing the same way, i'd guess?

Yes thats a bit strange, let ask Siglent.
I both cases the frequency measurement  is wrong.

[StillTrying]

Assuming the waveform is the scope's 1khz 3V p-p compensation signal, and one channel's X10 is wrong, then all the frequencies and voltages are correct!

Low pass digital filter set on ch1?

[rwgast_lowlevellogicdesin]

Ive tried both probes, on both ports same thing.

As far as the digital LP filter, I believe this may be the problem. A while back I was designing a 7 pole 30mhz LP and I went futzing around with the digital LP filtering in the scope to see how well it worked. The thing is when I hit ch1 the third page of the menu is for digital filtering and it is all turned off, maybe there is a setting somewhere else I dont know about or something. Iv'e been through all the menu's but I know the scope has and X,Y view and I haven't seen that either so I know there is some settings somewhere I'm missing. Hitting the return to defualts button and reflashing the firmware doesn't wipe all the settings I know this because me measurement menu has never changed from my set up. Is there a soft reset or something?

So this obviously isn't a problem commonly seen on a DSO, and most likely not hardware? Oh and it passes all the self cal and self testing.

[tautech]

Just to be sure, in the OP you mentioned 100 KHz, don't you mean 1 KHz ? 

If you're using the probe cal output, then what's on the display doesn't match the channel settings.....3 V amplitude should be 6 divisions @ 500mV/div.
It seems to me that you haven't set the input attenuation to match the probe's attenuation, check that will you.


Edit
Another trick is to push the Trigger level adjust knob, that autosets the Trigger level to 50% of the waveform.
Also, check the Ch 1 BNC for rigidity just in case the mount is broken, I've not seen this on Siglent's, only on other scopes and it might explain an unexpected waveform.

[tautech]

What's the DSO's SN #, we don't need the last 4 digits but the ones prior to them signify the build date, it might sill be under warranty.

[nctnico]

Isn't there an 'auto setup' or 'default' button on the scope? I'd try that.

[rwgast_lowlevellogicdesin]

I have pressed the auto set button a few times.

Im sorry somehow channel 1 was not at 10x, and I did mean 1Khz not 10Khz. Here is a picture of the waveform at 500mv 10x setting and the Auto 50% Trigger button pressed.



The SN# sds00001130545, that would be great if it was under warranty but it's an older model so im sure im not that lucky.

[tautech]

0113 signifies Jan 13 build but have you got proof of purchase and exact date ?

I'll get a similar scope out in the next day or so and see what I can do to replicate your fault.

[edpalmer42]

On a square wave, when you lose the horizontal parts of the signal, that means you've lost your low frequency response.  If you lose the vertical parts of the signal, you've lost your high frequencies.  If you have a signal generator, check your low frequency response to see where your -3dB point is.  That might help determine whether you're looking at a software glitch or maybe a cracked solder joint.

Ed

[rwgast_lowlevellogicdesin]

I leant my Sig gen out to a buddy(it's only 2mhz), I could sweep the scope with a sound card gen to find the -3db point right? I mean it should be well under 1khz based on the wave form, idk if a sound card can output under 2khz.

As far as proof of purchase, I bought it working last year on the forum, from a member in the eu. It was bought for one of his clients who bought a Tek and never used it. So does not having the original purchase date with proof mean I'm out of luck?

[tautech]

As far as proof of purchase, I bought it working last year on the forum, from a member in the eu. It was bought for one of his clients who bought a Tek and never used it. So does not having the original purchase date with proof mean I'm out of luck?

I was just trying to investigate with you IF you had the original purchase documents, obviously not.
Even though we've established a Jan 13 build, I often get new units from the factory that are already some months old, that is they're produced in batches, held as stock then shipped only then to be probably held in stock again by the distributor before sale to the user.
Get the picture  some units that I might have on hand are 12 months or so since their build date before sale to a end user. I warranty Siglent units for 3 years from date of purchase.

So if there's any chance you can find out the original date of purchase and from which dealer AND if it's a HW failure there might be hope for a warranty claim. 

Then again, if this DSO suits your requirements we might be able to arrange a replacement MB from Siglent if we don't pursue a repair.

[Anks]

I would expect a bad coupling cap. Its acting like a high pass filter. Just how it looks to me.

[StillTrying]

Random thoughts time.(is there any other type)

While you're playing it might be worth keeping both probes and both channels on X1, to eliminate one variable.

Is it the same shape with AC/DC ch1 coupling.

View the 1khz on both channels at the same time, just in case ch1 is loading the signal somehow.

Enable the digital filter on ch1 and try some settings to see if they change the waveform, set a LP of a few Mhz, if the filter settings have been corrupted somehow rewriting new values might fix it, including the on/off.

Try capturing some fast rising edges, just from a 1.5V battery would do, ch1 only seems to rise about 200mV before it has a problem.

200hz to 5khz 1V square waves from a sound card might help to give a clue what it's doing.

Well I did say they were random!

[tautech]

Ive tried both probes, on both ports same thing.

As far as the digital LP filter, I believe this may be the problem. A while back I was designing a 7 pole 30mhz LP and I went futzing around with the digital LP filtering in the scope to see how well it worked. The thing is when I hit ch1 the third page of the menu is for digital filtering and it is all turned off, maybe there is a setting somewhere else I dont know about or something. Iv'e been through all the menu's but I know the scope has and X,Y view and I haven't seen that either so I know there is some settings somewhere I'm missing. Hitting the return to defualts button and reflashing the firmware doesn't wipe all the settings I know this because me measurement menu has never changed from my set up. Is there a soft reset or something?

So this obviously isn't a problem commonly seen on a DSO, and most likely not hardware? Oh and it passes all the self cal and self testing.

I had a good crash around in the UI and menus of a new SDS1kCML+ unit while I was doing a presale check, I got all sorts of strange waveforms from the probe cal output using the inbuilt filters and other settings but I could not replicate anything that looked like yours. Defaults etc all worked as expected too. So I strongly suspect there's something in your Ch1 signal path faulty but I do have a MB here and I'll have a look at the input stage for you........

[rwgast_lowlevellogicdesin]

Sorry iv'e been a little busy doing Halloween activities with my daughter.  Thank you for checking around the software for me, I will get a message to the original buyer today to see if he has any proof of purchase or any info on the date/distributor the scope was bought from.

My digital filtering settings wont go above a few hundred Khz and my Acquire menu shows 2.5Msps (with one channel only) even though its a 500Ms scope, is any of this normal? I printed out some copys of the manual a while back I need to bust out but I would think set defualts should have fixed any goofy software settings :/. Ill also run some tests with my soundcard and post pics ASAP.

[tautech]

From an old thread by rf-loop:
https://www.eevblog.com/forum/reviews/siglent-sds1000-7'-oscilloscope-anatomy/

These look exactly the same as a MB I have here and my pics wouldn't be as good so I linked these.

MB


Underside


Ch input



I'm not sure if there's any reverse engineered schematics for the SDS1k channel inputs, someone following may know of them and hopefully provide you with a link.

[rwgast_lowlevellogicdesin]

Damn, I dont think I would even know where to begin there are so many discrete smt components, especially if your not sure what values they are suppose to be in the first place. Ill usually rip anything apart and give it a try, but im a bit scared to screw up my scope and on top of that I wont have a scope to help trouble shoot anything on the mobo. I have some old rack mount HP analog but it is only 500khz so I dont think that will help me at all.

Do you know what it would cost me to order a new mb, or if maybe I could upgrade to a better bandwidth scope with a drop in mobo replacement?

[tautech]

Damn, I dont think I would even know where to begin there are so many discrete smt components, especially if your not sure what values they are suppose to be in the first place.

Do you know what it would cost me to order a new mb, or if maybe I could upgrade to a better bandwidth scope with a drop in mobo replacement?

I'd hope that it would be possible to follow the signal path with the good channel on your DSO, but a schematic is needed as these are a multilayered PCB. Values, don't worry about them, I've got your back with a MB on hand.  This one has a corrupt LCD output but the CH inputs are good.


Until we get hold of Siglent (weekends  ) I couldn't say. They've been very helpful to forum members in the past, so let's just give it a bit more time to see what can be arranged.
Where in the US are you ?
At some point we'll probably hand over to some of the Siglent folk at Ohio to help get this sorted for you.
I can't say (don't know) what stock of parts they have on hand or if bits would need to be sourced from the factory.
Sit tight for a few more days then we'll know more.

[rwgast_lowlevellogicdesin]

Im in southern california about 100 miles from LA near Palm Springs. Thanks for everything, a lot more than I expected when I opened this thread.