Author Topic: Trying to figure out discrepancy between readings on DSO and analog oscilloscope  (Read 393 times)

0 Members and 1 Guest are viewing this topic.

Offline Hgspine

  • Contributor
  • Posts: 19
  • Country: us
Hello,

I have an older direct drive turntable that has a trim pot adjustment for sinewave voltage that drives the motor.  The turntable was made before DSO's were available.  The service manual tells which controller test point to use for the (analog) oscilloscope tip and where to connect the ground lead. SM instructions say to (with platter in stop mode, no sine wave to motor) adjust the baseline of the scope signal (horizontal trace) to the center of the graticule.  Then with platter spinning adjust trim pot so that the max and min voltages of the sinewave are equal.  Turning the trim pot only affects the negative portion of the sine wave.  Turning the trim pot clockwise results in a lower negative voltage of the sine wave, but positive voltage stays the same.  Sinewave generator is not perfect, slight variations of voltages between each cycle.  Hence trying to get a better result with DSO.

I thought that with a DSO, if I set the trim pot so the mean voltage was zero over a large number of cycles, that would mean average positive voltage would equal average negative voltage, and that would be a more perfect way to set sine wave voltages with the trim pot. Measuring done with a 2000 series Tek DSO, 20K measurement points setting, DC coupling, edge triggering, somewhere around 100 cycles per measurement gate.

So when I get the trim pot set so that the DSO reads a mean of zero volts (DSO meaurement math function reading) for the sine wave and then I check on the analog scope using the SM recommended instructions, the negative max voltage of the sine wave is about a box lower than the positive max voltage of the sine wave on the analog scope.

I was using 0V offset on the DSO for measurements.  Maybe I should have put a DMM on the test point and measured the voltage in stop mode and set the offset negative to that on the DSO?

Not sure how to do this correctly on the DSO. I don't understand the discrepancy between what I am seeing on the DSO and on analog scope measuring the same signal. I am assuming that I am doing something incorrectly with the DSO.

Thank you.
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3787
  • Country: au
Are both scopes set to DC coupling, and do both have the probes at x10 (10M impedence)?

I am not sure of the frequency of the sinewave, but you can probably just use a multimeter set to DC volts to check the offset - adjust the trimmer for 0V. You may need to go up one range - so if the sinewave is 10V p-p and the multimeter has a 20V DC range, try it on both the 20V and 200V ranges and if they disagree, trust the 200V range.

If the sinewave cycles vary from cycle to cycle, it could just be a triggering issue. The two scopes may be triggering at different voltages, so  the DSO may be triggering when the waveform is a bit more biased in the positive direction. Set the analog scope for a really slow sweep (say 1 sec/div) and see if the sinewave offset is changing during the sweep.

Richard
 

Offline Hgspine

  • Contributor
  • Posts: 19
  • Country: us
Richard,

Thank you very much for your reply. 

Tek probe is not adjustable, says 10x (also says 10Mohm/<12pf).  I used this probe with the Tek DSO.

With analog scope I used an adjustable probe that is adjustable for either 1x or 10x.  It was in the 1x mode. (ugh)

I connected the DMM to the test point just now.

If I measure AC voltages, Math function of DMM (randomly stopped by me at 566 measurements) was min voltage of +26.771mv, max voltage of +28.414mv, average voltage of +27.598mv

If I measure DC voltages (567 measurements) min voltage was -3.916mv, max voltage was +7.067mv, average was + 1.520mv

The DC measurements are a lot closer to the 0v setting I made with the DSO.

I think the DSO is processing a lot more measurements than the DMM.


 

Offline Hgspine

  • Contributor
  • Posts: 19
  • Country: us
Both scopes were set to DC coupling.  I tried AC coupling with the DSO, but in AC coupling mode, turning the trim pot made no difference in the mean voltage reading.

I assume the trim pot is adjusting the amount of DC in the sine wave.
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3787
  • Country: au
Richard,
I connected the DMM to the test point just now.

If I measure AC voltages, Math function of DMM (randomly stopped by me at 566 measurements) was min voltage of +26.771mv, max voltage of +28.414mv, average voltage of +27.598mv

If I measure DC voltages (567 measurements) min voltage was -3.916mv, max voltage was +7.067mv, average was + 1.520mv
Do you have a basic handheld DMM? It sounds like you were using a fast bench DMM. If that is the only DMM you have and you cannot slow down the sampling rate, you will have to add a RC averaging circuit before the meter - say a 1M resistor to the test point and a 1 to 10uF capacitor across the DMM. You want to be measuring the average DC volts, not spot measurements of the AC waveform.
 

Offline vk6zgo

  • Super Contributor
  • ***
  • Posts: 5118
  • Country: au
It sounds to me that there is nothing wrong with your 'scopes, but perhaps something wrong with your generator.
It shouldn't matter where you set the DC level of the signal on the screen, if the instructions say you can adjust the sine wave to be symmetrical around that voltage, it should do so.
It seems that it is not doing so.
 

Offline Hgspine

  • Contributor
  • Posts: 19
  • Country: us
I have two handheld DMM's.  The voltage readings refresh quicker on one than the other.  On the one that refreshes slower the voltages are bouncing around between -0.015 and +0.020 volts, a bit higher on the positive side, which is what I was seeing with the bench DMM.

So this is roughly close to to the 0V mean I got with the DSO, me thinks...

I recapped the power supply but the drive board has old caps.  Maybe a recap will not have the voltages bounce around so much?
 

Offline dmendesf

  • Regular Contributor
  • *
  • Posts: 61
  • Country: br
Math with digital scopes can be... Interesting. I have no expertise with Tek gear but with Keisight scopes you have 2 different options that seems similar but get different results: "average" (that averages the entire memory buffer) and "average N cycles (that averages over N cycles of the incoming waveform). The simple average over the entire buffer always gets bogus reports caused by the non-integer number of cycles inside it.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf