Unequal Waveform Amplitude

[gyropilot]

Greetings,

I have a Siglent SDG1032X Signal Generator and a Siglent SDS1202X-E Oscilloscope which are both less than 6 months old.

I just noticed something interesting today...

When I have the signal generator produce a waveform with equal positive and negative amplitude, such as a sine or square wave, the oscilloscope will measure that waveform with sightly higher positive than negative amplitude.  For example, let's say I set up a simple sine wave having +500 mV and -500mV amplitude.  The oscilloscope will show this as roughly +500 mV and -480 mV (not actual measurements... just an example).

I've verified no signal offset or phase angle is set in the signal generator waveform.

I've run the self test on the oscilloscope with no change.

It doesn't seem to matter whether I use an 1x, 10x probe, or a 36" length of RG-58 coax (properly terminated of course) between the signal generator and oscilloscope.

Frequency and signal amplitude don't seem to matter, although the percentage of error changes with both.

Does this sound like normal behavior?  Is there anything else I should check?

Thank you,

John
AA7US

[DaJMasta]

Oscilloscopes aren't really precision voltage measurement devices, so it could be just related to the error in the front end amplifiers, but it's possible for a small amount of offset to be introduced in a lot of places.  Does the offset change if you look at a different vertical resolution setting?  Does the self test/self cal on the scope include zeroing any offset?  How did you verify the signal generator output, is it just that no offset is set, or did you use a multimeter with appropriate AC bandwidth?  Have you tried just AC coupling the scope and seeing if it goes away?


There are a lot of stages and it's small enough to likely be near the noise floor of the specifications of the two instruments, but it is certainly nice to know where it's coming from.

[vk6zgo]

Are you just going by the 'scopes readout, or have you compared the amplitudes on the actual display?
Is the 'scope set to AC coupling or DC coupling?

If, switching from one type of coupling to the other, the waveform moves vertically on the screen, you probably have a DC offset on the waveform.

I was going to suggest that you look at the gen output with a DMM on DC volts, but it may not always work, as some DMMs can give false readings in the presence of an AC waveform with a DC offset.

[tautech]

Does this sound like normal behavior? 

Yes.

There's at least three specifications at play here:
Warm up time as specified in manuals and datasheets, unless done this can affect channel offset and both DSO and AWG accuracy.
DSO vertical accuracy is typically just +3%
AWG amplitude accuracy +(1%+1 mVpp)

All these things add up to what you see. 

Small edits for spelling and meaning.

[gyropilot]

Does the offset change if you look at a different vertical resolution setting?

No... the offset remains when changing the vertical resolution.

Does the self test/self cal on the scope include zeroing any offset?

Unfortunately no.

How did you verify the signal generator output, is it just that no offset is set, or did you use a multimeter with appropriate AC bandwidth?

I've not verified the validity of either the signal generator or the scope.  I simply made sure there was no offset selected for the signal generator waveform..

Have you tried just AC coupling the scope and seeing if it goes away?

Yes and no change.

I should probably mention that both the signal generator and scope have dual channels.  I tried both and the offset is exactly the same no matter what combo I tried.  Also, when inverting the signal on the scope, the offset follows the inversion.

There are a lot of stages and it's small enough to likely be near the noise floor of the specifications of the two instruments, but it is certainly nice to know where it's coming from.

I'm new to test equipment and I don't think I have any way to properly check the signal generator's waveform.  I have a reasonably good handheld DMM, but I don't think it'll tell me the positive and negative component of an AC waveform.  I suppose I could try and build a test rectifier from some diodes to isolate the positive and negative components.

Thank you for the reply,

John
AA7US

[gyropilot]

There's at least three specification at play here:
Warm up time as specified in manuals and datasheets, unless done this can affect channel offset and both DSO and AWG accuracy.
DSO vertical accuracy is typically just +3%
AWG accuracy +(1%+1 mVpp)

All these things add up to what you see. 

Thank you!

John
AA7US

[gyropilot]

Are you just going by the 'scopes readout, or have you compared the amplitudes on the actual display?

I'm using the scope's built-in measurement function where it calculates and displays various signal parameters.  Besides seeing the offset displayed numerically, it's also visible on the scope's trace if you look closely.  It's small enough so it's not very noticeable with a quick glance.

Is the 'scope set to AC coupling or DC coupling?  If, switching from one type of coupling to the other, the waveform moves vertically on the screen, you probably have a DC offset on the waveform.

Either setting shows the slight offset.  If a DC component is present on the signal, the trace shifts appropriately on the scope screen, but there's still the offset to the AC positive and negative components.

I was going to suggest that you look at the gen output with a DMM on DC volts, but it may not always work, as some DMMs can give false readings in the presence of an AC waveform with a DC offset.

It's worth a try so I'll give it a shot tomorrow.

Thank you for the help,

John
AA7US

[MasterT]

,,,,, I have a reasonably good handheld DMM, but I don't think it'll tell me the positive and negative component of an AC waveform.  I suppose I could try and build a test rectifier from some diodes to isolate the positive and negative components.

John
AA7US

If your DSO has FFT, positive/ negative asymmetry easily to spot as 2-nd, 4-th, etc harmonics content, You also may use your PC sound card as AWG with pretty good parameters, or even small transformer connected to AC outlet - 100% there wouldn't be any offset.

[StillTrying]

You'll also have to have the waveform large enough on screen - about 6 divisions high to get any accuracy.