# conundrum

Discussion in 'General Electronics Chat' started by kokusaikeirin, Nov 26, 2013.

1. ### kokusaikeirin Thread Starter New Member

Nov 26, 2013
9
0
Greetings! I'm currently attending an EET course and ran into a situation where, when measuring across two resisters connected to a signal generator set at frequencies above 80khz, there is a slight phase shift and a small change in symmetry on the alternations, the larger being on the negative alternation. Any ideas?

2. ### wayneh Expert

Sep 9, 2010
13,627
4,422
Compared to what? Measured how (I'm thinking artifact)?

3. ### kokusaikeirin Thread Starter New Member

Nov 26, 2013
9
0
Using both a digital and analog oscilloscope, with channel one across the source and channel two across the resistor tied to the source, there is a phase shift a little less than 90 degrees. We changed generators, Scopes, resistor types, and probes with no change to results

4. ### kokusaikeirin Thread Starter New Member

Nov 26, 2013
9
0
IM sorry across 2 resistors in series

5. ### crutschow Expert

Mar 14, 2008
16,551
4,464
Sounds like you are measuring the same point. Post a diagram of your setup exactly.

6. ### kokusaikeirin Thread Starter New Member

Nov 26, 2013
9
0
Not sure how to do that. I'll try to describe it more clearly. Two resistor in series, one probe at source, the other placed at junction of two resistors(both 39k). One resistor is dropping only 20 mv when there is 250mv applied. HOWEVER there is a phase shift and the signal waves are not symmetrical. Sorry I'm still new to electronics and forums

7. ### kokusaikeirin Thread Starter New Member

Nov 26, 2013
9
0
Can I post photos?

Nov 26, 2013
9
0

Nov 26, 2013
9
0
10. ### WBahn Moderator

Mar 31, 2012
20,247
5,758
Why do you mention "above 80kHz" and not just say "at a frequency of 330kHz"?

What size are the resistors? One looks like 5.1MΩ, is that correct? Is the other one the same size? If so, notice that you have an attentuation by a factor of about 13 instead of 2.

If so, your scope probes have resistances comparable to this and hence your measurement is being greatly influenced by the resistance and capacitance of the two probes.

Replace your resistors with even 1MΩ and you should see a significant change toward what you expect. If you go to 100kΩ then you should see something that has only a slight phase shift.

Apr 16, 2011
272
105
Try switching your scope probes between x1 and x10 (adjust vertical sensitivity to bring trace back to viewable size).

I suspect that you will see less phase shift on x10 than on x1. This will show that the scope probe capacitance is forming an RC filter with your resistors.

x10 probe has less capacitance and higher impedance than x1, and thus affects the circuit under test less.

At least I think this is the main effect happening here. I am too lazy to do the maths to prove it

Last edited: Nov 26, 2013
12. ### kokusaikeirin Thread Starter New Member

Nov 26, 2013
9
0
Both resistors are 39kΩ. I mention 80khz because as soon as I raise the frequency, these abnormal effects start to occur. I will try different values to see if that affects anything.

Last edited: Nov 26, 2013
13. ### kokusaikeirin Thread Starter New Member

Nov 26, 2013
9
0
Thanks for the reply. We adjusted the scope probes to x10 and the results didn't change from what I recall (no longer in class). I'll bring that up tomorrow.

14. ### crutschow Expert

Mar 14, 2008
16,551
4,464
Your appear to be observing the difference between theoretically perfect components and the real world. A large part of engineering is often identifying that difference.

15. ### WBahn Moderator

Mar 31, 2012
20,247
5,758
Did you compensate the probes?

16. ### WBahn Moderator

Mar 31, 2012
20,247
5,758
Duh. Brain fart. I don't know what I was thinking.

I would not expect such a dramatic effect, particularly the huge attentuation that you are seeing. What happens if you reverse the probes? Reverse the resistors? Swap channels? Use the probes and scope from a different station?

And be sure to compensate the probes each time you use them (or change which channel they are connected to).