Square signal level > 20ms not looking good - Siglent SDS800X HD

[eTobey]

I have compensated my probe, but when looking at longer signals, they dont look that good. The probe is directly plugged into the BNC of the signal gen (SDG1032) with a proper adapter for it.

Is this a bad probe or what do i do wrong?


Edit:
Found some good information:
https://blog.teledynelecroy.com/2018/10/putting-probes-in-perspective.html
The following articles/post can be found on the dates list on the right.

[tggzzz]

I have compensated my probe, but when looking at longer signals, they dont look that good. The probe is directly plugged into the BNC with a proper adapter for it.

Define the signal: voltage, frequency, source impedance.
Define the probe: type.
Define the "proper adaptor".

Yes, I'll probably regret responding to this.
But maybe, just maybe, the OP will learn how to ask a question that can be answered.
https://entertaininghacks.wordpress.com/library-2/good-questions-pique-our-interest-and-dont-waste-our-time-2/

More likely: I am too optimistic.

[Manul]

At such low frequencies, probe compensation does not really matter (it's for high frequencies). I would say that what you see is an actual signal, in other words, it really happens. Does not look like probe's or scope's frequency response issue. We don't know enough about your setup.

[thm_w]

Only perform the adjustment and calibration using the front cal signal as explained in the manual.

At such low frequencies, probe compensation does not really matter (it's for high frequencies). I would say that what you see is an actual signal, in other words, it really happens. Does not look like probe's or scope's frequency response issue. We don't know enough about your setup.

There are LF and HF probe compensation adjustments: https://www.picotech.com/library/application-note/how-to-tune-x10-oscilloscope-probes
Most likely the probes OP is using only has LF adjustment.

But I agree, assuming the probe is properly compensated its probably the actual signal.

[schmitt trigger]

If you can borrow someone else's probe, you could verify whether the signal source is faulty.

Another possibility is that the scope channel itself is faulty. Have you tried the test on a different channel?? I know this is a remote possibility, but it will take you less than a minute to perform this test.

[David Hess]

If the input signal starts out significantly off of the display, then you are likely overloading the oscilloscope.  When the operating point of the signal amplifiers is changed, then thermal effects on the order of milliseconds will affect the signal.

[MrAl]

I have compensated my probe, but when looking at longer signals, they dont look that good. The probe is directly plugged into the BNC with a proper adapter for it.

Is this a bad probe or what do i do wrong?

Hello there,

Unfortunately, the only way to be absolutely sure if it is the scope, probe, or signal generator, the only way to know for sure is to get a different scope or different frequency generator and test using that.  If a different (manufacturer) generator or different scope shows the same thing then it's the scope or probe.  You can always try a new probe they are not that expensive.

You may not have either of these extra pieces of test equipment though so the only thing you can do is think about the quality of the generator, or perhaps just get some low-cost electrical parts like a 555 timer and generate a pulse with that and see what comes of it.

If you are using a homemade circuit already to generate the pulse, that could be the culprit, but without more information it's hard to say.

[BillyO]

Generally, it looks fine to me.  +/- 50mV on the leading edge of a typical probe compensation under typical conditions does not set off any alarms.  However, I don't have all the facts.

Could you let those of us that do not follow every minute of your amazing life (this is not a realty TV show!) the following:

1) The scope.
2) Has it been hacked?
3) The signal source (generator, frequency, risetime, etc..)
4) The probe
5) Any other information that might possibly be pertinent.

[eTobey]

I have now compensated the probes again (see picture "compensated on scope 1" and "compensated on scope 2"). The CH1 is the original probe the scope came with, and the CH4 probe is a Testec "TT - LF 312".

The signal does not  look any better (picture "still not that good").

I also made sure, that the connection of the other probe (CH4) is not affecting this issue too much. Picture "ugly" shows how it looks with that ugly connection, and picture "ugly fixed" shows it with a proper BNC adapter.

The setup can also be seen at the picture "setup".

The scope is a Siglent SDS800X HD which is not hacked, but "opened" via licence key. The generator is a Siglent SDG 1032X. I made sure, that the inputs are not overdriven (its not visible on the pictures).

[shapirus]

Now try to switch voltage ranges on the scope, to the point where the relay will click (and adjust the signal level if necessary), and you'll very likely see yet another change in the vicinity of the rising/falling edge.

I've never been able to compensate a probe to get an ideal waveform display in every condition. Yes it works fine if the waveform is zoomed out sufficiently: when 5-10 periods fit on a single screen, then it's a perfect square wave. Things change when you start to zoom in.

That's why I stop at "good enough" for the probes. It works for practical purposes, considering all the other limitations of passive probes, but it still does not explain the fine effects of it. I am pretty sure there are good articles on what probe compensation actually does, and they can be used to get a better understanding.

To answer the "is it the probe or is it the signal?" question* you need to compare what you see with the probe with what you see without it: connect the signal generator, assuming it has 50Ω output impedance, using a coax cable with a 50Ω feed-through terminator at the scope input, that will allow you to see the signal with as little distortion as possible. Of course, internal 50Ω termination would be much better, but your scope doesn't have it. External terminator should be fine except very close (few ns) to the edges.

*but not the "is it the scope?" question: that is yet another variable.

p.s. You do understand that a compensated probe is only compensated relative not only to a given scope, but also to a given input of that scope (and as I mentioned above, also, it seems, to a given vertical scale range), don't you? The capacitors and wiring on each input can differ sligthly, so they will require a different amount of probe compensation.

[CosteC]

I would try to observe signal from generator directly, without probe but with 50 Ohm termination at the oscilloscope and with 50 Ohm cable.

This will tell if signal from generator is as rectangular as we expect it. It may not.

Then I would check if generator risetime is better than oscilloscope calibration output. Generator again, I would use with 50 Ohm termination rather than Hi-Z mode for probe termination.

Finally, what accuracy is guaranteed by Siglent for this oscilloscope? Maybe it is fine "as is"?

[eTobey]

I am pretty sure there are good articles on what probe compensation actually does, and they can be used to get a better understanding.

Too bad google has become useless. Thousands of beginner tutorials, that do not go deeper. And then there is a lot of AI shit generated out there.

Do you remember those good old times, where there only were links on sites, and no google? Good thing was, it mostly was selected (good) ones.

To answer the "is it the probe or is it the signal?" question* you need to compare what you see with the probe with what you see without it: connect the signal generator, assuming it has 50Ω output impedance, using a coax cable with a 50Ω feed-through terminator at the scope input, that will allow you to see the signal with as little distortion as possible. Of course, internal 50Ω termination would be much better, but your scope doesn't have it. External terminator should be fine except very close (few ns) to the edges.

See pictures.

p.s. You do understand that a compensated probe is only compensated relative not only to a given scope, but also to a given input of that scope (and as I mentioned above, also, it seems, to a given vertical scale range), don't you? The capacitors and wiring on each input can differ sligthly, so they will require a different amount of probe compensation.

Yes i do, but for every scale? Have to test this...

[eTobey]

Some more tests:
BNC cable direct into scope with 50 term (yellow), and probe with coax tip adapter into into a y-adapter on the generator.

I also took a 12V battery, to "create" some signal with less variables (i guess). Not having a stable reading on it for 20ms does not feel so great to me?! How would, if there is, that specification of such a thing be called?

[shapirus]

Try dot mode display.

[CosteC]

Yes i do, but for every scale? Have to test this...

Well..
This is very good question.

Answer is: Depends on construction of front end. If input changes impedance depending of attenuation, which is entirely possible, then YES - probe compensation will be correct on ONE gain setting or for example two. Depends.

WHY?
For example at 1, 2, 5 mV/DIV input may have 10 pF capacitance. Then relay clicks to other divider ratio for 10/20/50 mV/DIV and at this setting input has 12 pF... changes between 1/2/5 can be done in software. This is only example of course.

[bdunham7]

This issue has been noticed and discussed regarding the older SDS1000X-E scopes and this looks like the same thing.  Sharp transitions at low repetition rates (like a 100Hz square wave) will result in these distortions on what should be the flat section of the waveform.  It is not strictly an overload recovery problem as it can be demonstrated with signals that are entirely on-screen.  IDK about the input capacitance changing with vertical scale changes, but that would be another source of distortion if it is happening.

[pdenisowski]

I am pretty sure there are good articles on what probe compensation actually does, and they can be used to get a better understanding.

Too bad google has become useless. Thousands of beginner tutorials, that do not go deeper. And then there is a lot of AI shit generated out there.

As someone who has made a probe compensation video : how much deeper / what additional topics would be helpful? 

It's a serious question: I would be very happy to do a deeper dive, I'm just not sure what I would add. 



Edit:  I should mention that towards the end of the video there is a slide on troubleshooting compensation issues

[pdenisowski]

The scope is a Siglent SDS800X HD which is not hacked, but "opened" via licence key. The generator is a Siglent SDG 1032X. I made sure, that the inputs are not overdriven (its not visible on the pictures).

Sorry if I missed it somewhere about, but is there a reason why you're not using the probe compensation outputs on the front of the scope?

[shapirus]

Sorry if I missed it somewhere about, but is there a reason why you're not using the probe compensation outputs on the front of the scope?

This brings up another question: is there any difference between using the scope's built-in probe compensation output vs an external source? Sometimes what's produced by the scope's one may be not the squarest wave possible.

Another interesting question: does the probe compensation adjustment waveform have to necessarily be a 1 kHz signal?

[shapirus]

As someone who has made a probe compensation video : how much deeper / what additional topics would be helpful? 

It's a serious question: I would be very happy to do a deeper dive, I'm just not sure what I would add. 

Good vid. One small (if it's small) issue is that it doesn't cover the difference (in terms of schematics) of having the variable cap before or after the coax.

A good follow-up video would include a howto guide of building an equivalent circuit and running a SPICE simulation of a 10x (and 1x as well!) scope probe. A simulation is helpful in understanding how the probe loads the DUT and how it distorts the original signal before it is displayed on the scope screen, and also what difference can be expected between a probe and a 50Ω-terminated (internal or external) coax.

We already know where the resistors are, where the capacitors are, now we can build a netlist having them, and add an LTRA element between the probe tip and the scope input (with the variable cap being on one or the other side of the LTRA). The difficult part is that we need to know the actual R, L, and C values that typical probes have to run a meaningful simulation.

[David Hess]

This brings up another question: is there any difference between using the scope's built-in probe compensation output vs an external source? Sometimes what's produced by the scope's one may be not the squarest wave possible.

An external source will work fine for probe compensation.  I sometimes use an external source because the BNC output can be connected coaxially to the probe tip reducing noise.  Some older oscilloscopes had a BNC output for their compensation output which is especially useful.

Another interesting question: does the probe compensation adjustment waveform have to necessarily be a 1 kHz signal?

No, it can be a different frequency.  1 kHz is commonly used because the probe calibration output is also a frequency and voltage reference.

Answer is: Depends on construction of front end. If input changes impedance depending of attenuation, which is entirely possible, then YES - probe compensation will be correct on ONE gain setting or for example two. Depends.

I do not know how newer oscilloscope get consistent input impedance without trimming.  Older oscilloscopes have a pair of capacitive trimmer for each high impedance attenuator.  One adjusts the compensation, and the other adjusts the input capacitance, so all attenuators on one channel can have the same input impedance, and channels on one oscilloscope can have the same input impedance, and all channels on multiple oscilloscopes can have the same input impedance.  This allows probes to be moved between channels and instruments without adjusting compensation.

Some more tests:
BNC cable direct into scope with 50 term (yellow), and probe with coax tip adapter into into a y-adapter on the generator.

I also took a 12V battery, to "create" some signal with less variables (i guess). Not having a stable reading on it for 20ms does not feel so great to me?! How would, if there is, that specification of such a thing be called?

I am not sure what you are asking here, but a "reference level pulse generator" like a Tektronix PG506 is used for testing settling time and transient response.  The design is a little different from a common pulse generator and produces a minimum of aberration even at high speeds.  The probe compensation output is good enough for adjusting probe compensation, but may or may not be better than this.

[eTobey]

As someone who has made a probe compensation video : how much deeper / what additional topics would be helpful? 

Thats a very good video. I have one question on this: Why is a variable capacitor on the prope tip unpractical?

It would be interesting to know (if the probe is the cause), why a signal takes 20ms to become stable. To me it seems it comes from the probe, because touching a battery (with additional cap) looks the same. Also on my testec, i can not get a perfect square, since there always will be a short overshoot at the beginning, or a tiny and longer flat drop after it.

The scope is a Siglent SDS800X HD which is not hacked, but "opened" via licence key. The generator is a Siglent SDG 1032X. I made sure, that the inputs are not overdriven (its not visible on the pictures).

Sorry if I missed it somewhere about, but is there a reason why you're not using the probe compensation outputs on the front of the scope?

Why would you think, that i did not use the comp outputs of the scope?

[G0HZU]

If it helps, see below for a screenshot from my old Infiniium 54825A scope (500MHz) with an Agilent 10073C x10 scope probe (compensated) fitted to channel 2.

A 2.9V pulse from an HP pulse generator is being measured with the scope and I've set the scope to 100mV/div and channel 2 is set up for a x10 scope probe.

You can see the pulse is still quite flat even when zoomed in at 100mV/div.

The pulse generator is an HP 8002A (made in 1972).

[pdenisowski]

Thats a very good video. I have one question on this: Why is a variable capacitor on the prope tip unpractical?

Thanks!  It's due to size and also because it's a good idea to keep a reactance away from the measurement point.

Sorry if I missed it somewhere about, but is there a reason why you're not using the probe compensation outputs on the front of the scope?

Why would you think, that i did not use the comp outputs of the scope?

Sorry, maybe I misunderstood you when you said:

The probe is directly plugged into the BNC of the signal gen (SDG1032) with a proper adapter for it.

I don't have an SDS800X HD, but from the pictures I've seen it looks like it has the "typical" probe compensation signal and ground connectors on the front (see attached).  These are the points I always use for probe compensation.  Again, my apologies if I mis-read your post or if I'm just asking a stupid question

[G0HZU]

Here's a similar test using a Tek MSO4104 scope and another Agilent 10073C x10 probe. In this case, the scope didn't make it easy for me to show the pulse top on 100mV/div.
So I set the pulse to about 4.5V and set the scope to 500mV/div.

It still looks quite flat. I rarely use this scope as the UI is horrible so I'm probably not getting the best from it.

I have compensated it for the Agilent 10073C x10 probe.