SIGLENT SDS1104X-E measurement problems

[Gaktyt]

empty, i can't seem to attach my 3 images correctly.

[Gaktyt]

I have some weird problems measuring with my new SDS1104X-E.

This is the signal pulse I'm trying to capture in full:


On the same capture as above; setting to 2V/DIV - this is the signal I'm expecting to see in the next capture, obviously with the missing peak.


However this is what I get:


For this capture I had to increase delay as it triggers at some preceding noise at this low trigger level.

The weird thing is that if I swap CH3 with either CH2 or CH4 and set them to 10VDIV both channels work as expected.  CH2 and CH4 also work as expected with the scope ref signal.

I have also tried to dump the data for the 2V/DIV in the hope that the correct data was there but just not displayed correctly on the scope - alas it was not.

What could be the problem?

[Gaktyt]

it seems that setting CH2 and CH4 to 2V/DIV alters the signal, and that the issue is not trigger related. At least that's my current theory.

Triggering on CH3 with CH2 and CH4 at 1V/DIV


Triggering on CH3 with CH2 and CH4 at 2V/DIV

[tautech]

Some suggestions if I may.
Nearly always use DC trigger coupling. Then the trigger level is visible.
AC channel coupling is only needed if a DC offset is present.
Slow timebase down which displays more data points and captured waveforms can then be magnified both horizontally and vertically from the capture.....this is the real power of a DSO.....post capture data inspection.
Use Trigger holdoff when the capture you need to see is after a previous event or adjust trigger level so to miss that event.
Attempt to keep the trigger position on the display so to be able to see the actual trigger event.

Further tricks are to work in Zoom mode with the primary display at a slow timebase setting, then much pre-trigger information is available in the capture and you can use horizontal position to view it in the zoomed display.

You have much more power than a CRO at your fingertips so practice some to learn how to get the best from it. 

[Gaktyt]

tautech, thank you for the tips, I appreciate it.

Good rule of thumb for the AC coupling, was in doubt bout that. And I will start using trigger holdoff. ;-)

I agree the post capture data inspection is the rel power of DSO, I use it for the same reason. However the scope seems to alter my signal which renders it useless for my measurements. I will have to build an shielded attenuator to keep the signal within 1V/DIV as I don't have other ideas and options time being.

[tautech]

When I look at the signal amplitude of your screenshots vs the channel input attenuation settings, at first glance all are using 10x attenuation but one channel 10V/div. 
Do you have one probe set to 1x by mistake ?

If by chance there is, the additional capacitance loading from a 1x probe setting may be impacting on the signal fidelity.

[Gaktyt]

Do you have one probe set to 1x by mistake ?

Unfortunately not. When fiddling with the probe it's very easy to flip the x10 button by mistake, and I have experienced several ruined days only to discover it was set on x1, so from then I'm very alert about the probe setting, even check it several times a day even if I haven't been touching the probes just in case the lab gremlins did something naughty.

[tautech]

OK, so the question still remains what is causing the signal to be altered to what you see ?

Scope settings or probing technique ? 

[Gaktyt]

Scope settings or probing technique ? 

If it's scope settings, then I have no clue what it could be.

I currently doubt (not excluding it) it's probing technique, because CH2 and CH4 are connected to two individual scientific grade precision passive current transformers and as such are not physically connected to the circuit. CH3 is connected to the circuit, but if I disconnect CH3, the problem persists on CH2 and CH3.

[tautech]

Scope settings or probing technique ? 

If it's scope settings, then I have no clue what it could be.

I currently doubt (not excluding it) it's probing technique, because CH2 and CH4 are connected to two individual scientific grade precision passive current transformers and as such are not physically connected to the circuit. CH3 is connected to the circuit, but if I disconnect CH3, the problem persists on CH2 and CH3.

OK, can we do a sanity check using just the probe compensation 1 KHz square wave on all channels @2V/div.
Use whatever channel you like for triggering and spread the channels apart so we can inspect each waveform.
Post a screenshot.
TIA

[Martin72]

Hi,

because CH2 and CH4 are connected to two individual scientific grade precision passive current transformers and as such are not physically connected to the circuit.

Remember, all the channels got the same gnd potential.
Plus: Do you note the correct polarity of the transformers, when connecting the probes ?

[tautech]

Hi,

because CH2 and CH4 are connected to two individual scientific grade precision passive current transformers and as such are not physically connected to the circuit.

Remember, all the channels got the same gnd potential.
Plus: Do you note the correct polarity of the transformers, when connecting the probes ?

Yes, just dawned on me too....ground loop ....but why is 1V/div OK ?

[Martin72]

First of all, he should set the coupling to DC, all channels.
Then it would be good if he explain, what he expected to measure - circuit description.
And showing a screenshot wit lowered timebase, so we can see more from the signals.
Nevertheless, it´s a strange behaviour.

[Gaktyt]

Thank you all for your kind suggestions:

The scope is powered through a isolation transformer- - no earth
The circuit is battery powered - no connection to mains or earth
All channels except CH4 are physically disconnected from scope.
CH4 connected to current transformer specs here https://www.pearsonelectronics.com/pdf/4100.pdf
CH4 set to DC

CH4 connected to CT via BNC to probe adapter


CH4 connected to CT via 50ohm coax


CH4 connected to CT via 50ohm coax with 50ohm termination


Changing to 2V/DIV makes it impossible to trigger due to preceding noise, no matter what I set holdoff time to, it triggers at the same noise in the beginning.

It should be noted, that there are no problems when the scope is not only in 1V/DIV, but it's also fine at 500mV, 200mV etc. it's everything above 1V/DIV that presents problems.

[2N3055]

What is the amplitude of the top of that pulse? The fast first edge. It seems that is waay up outside of screen.
If you are overloading the input that might explain wierd signal behaviour.

[Gaktyt]

What is the amplitude of the top of that pulse?

I can't measure it ;-)

If you are overloading the input that might explain wierd signal behaviour.

Yes that could be a plausible explanation, but then why does it also work with 500mV/DIV and 200mV/DIV?

[2N3055]

Use faster timebase and find pulse top.
Also, current transformers are specified to work on specific load on secondary. You have to have 50 terminator on it, so correct is to use 50 Ohm coax and termination on scope side and set ch to 1X.  No other config is valid.

[Gaktyt]

NOW MEASURING WITH A SIGLENT SDS2104X

The scope is powered through a isolation transformer- - no earth
The circuit is battery powered - no connection to mains or earth
All channels except CH4 are physically disconnected from scope.
CH4 connected to current transformer specs here https://www.pearsonelectronics.com/pdf/4100.pdf
CH4 set to DC

CH4 connected to CT via BNC to probe adapter


CH4 connected to CT via 50ohm coax


CH4 connected to CT via 50ohm coax with 50ohm termination


Compared to the SDS1104X-E (in post 01:46:13 pm) the difference in results is quite impressive, maybe even worrying.

Here is the signal in full length at 500mV/DIV

[Gaktyt]

Use faster timebase and find pulse top.

How will a faster timebase do that? I'm not sure I understand you.

Also, current transformers are specified to work on specific load on secondary. You have to have 50 terminator on it, so correct is to use 50 Ohm coax and termination on scope side and set ch to 1X.  No other config is valid.

As I read the specs for this specific CT, all three configurations are valid, the 50ohm termination is just to avoid reflections. But perhaps I'm wrong, so if I'm wrong please tell me where in the CT documentation you find it.

[Martin72]

Could you share a photo from your testsetup with us ?

[Gaktyt]

Could you share a photo from your testsetup with us ?

Sure, here you go:

[Martin72]

OK, thanks.

CH2 and CH4 are connected to two individual scientific grade precision passive current transformers and as such are not physically connected to the circuit. CH3 is connected to the circuit, but if I disconnect CH3, the problem persists on CH2 and CH3.

If your circuit is not top-secret, could you show the schematic from it and at which points you´re measuring ?

If I remember it right, we got a such a current transformer too and a 1104X-E, maybe I can nearly reproduce it at work.

[Gaktyt]

OK, can we do a sanity check using just the probe compensation 1 KHz square wave on all channels @2V/div.
Use whatever channel you like for triggering and spread the channels apart so we can inspect each waveform.
Post a screenshot.
TIA

[Gaktyt]

If your circuit is not top-secret, could you show the schematic from it and at which points you´re measuring ?

It's not top secret, it's mainly made of DIY components which took me weeks to fabricate, so you wouldn't be able to replicate it unless you had the time and machinery.

If I remember it right, we got a such a current transformer too and a 1104X-E, maybe I can nearly reproduce it at work.

I sincerely (in the original meaning of the term) appreciate your kind gesture, however I'm afraid it won't be this time. But thank you once again.

[Martin72]

Without further information, helping will remain a joint guessing game.
OK, let´s guess again…
As we don´t know the value of the current through the transformer, maybe he´s saturating or the inputs of the scope are going to clip.
Hard to believe you couldn´t get the full signal (Amplitude) on the screen, with 10x probes, 400V are the maximum.