scope and generator question

[snoof]

So i have my sine wave generator hooked to chan2 (as well as the amp input) and chan1 is hooked to my dummy load.  As i turn the amp up the chan1 sinewave changes(as it should), but also the the chan2 input sinewave chages as well, it starts to distort but not in the same way as chan1's wave.  This doesn't seem right to me.  I've tried "Ying off" the sinewave a few different ways, and it always distorts along with the output.  Any thoughts??  I should also add that I've tried two different sig generators, one of which has a buffered output.

[sluckey]

What amp? Exactly how do you have the sig gen connected to your scope and amp?

[snoof]

well, I've tried this on both a cathode biased 2x6l6 amp, and a cathode biased SE EL84 amp.  The sig gen is plugged into the amp input jack.  I've tried the headphone out on my laptop running a sig gen program, w/ the L chan feeding the amp, the R chan feeding the scope.  I've also tried it with a standalone batt powered home brew sig gen(buffered out), and tapping the input jack to feed the scope.

[sluckey]

I don't think you would have that problem using a good quality sig gen such as HP 204D Oscillator or HP 3310B Function Generator. (Those are just examples. There are plenty of other good sig gens too.) And I don't really understand how just turning the volume up on a typical amp would cause the input signal to distort even on your basic sig gen.

[snoof]

And I don't really understand how just turning the volume up on a typical amp would cause the input signal to distort even on your basic sig gen.

me either 

[tubesornothing]

Post  a schematic - indicate where you inject the signal and where you are measuring it.   Without this info its hard to guess what is going on.

[snoof]

It does this behavior with multiple amps, multiple sig gens and multiple "y-ing" and injection points.  I'll try it out on my standard Fender PR later today.  It almost seems as if there is some crosstalk in the scope chans, but I've tried each sig gen into it's own chan on the scope, and the prob does not occur.

[sluckey]

Which scope? How are the switches set?

[Merlin]

Does the wave start to compress on one side? If so then that's normal- you're just seeing grid current flowing in the input valve as you start overdriving it, causing the input signal to clip.

[snoof]

Bottom trace is sig gen, top is amp on dummy load...







Fender SF Bandmaster head
vol on 3
vol on 3.5
vol on 4

output on sig gen is constant


B&K 2125A scope
source - chan2
vert mode - dual
coupling - auto
both probes on A/C setting
chan1 - 1vpd x10
chan2 - 50mvpd x10

laptop sig gen, stereo out, Lchan amp, Rchan scope.  Same results with batt powered buffered homebrew sig gen.

[sluckey]

output on sig gen is constant

But your pics show the bottom trace is approx 1 division, then approx 1.5 divisions, and finally over 3 divisions. That's not very constant.

laptop sig gen, stereo out, Lchan amp, Rchan scope.  Same results with batt powered buffered homebrew sig gen.

You're not actually monitoring the same signal that you're injecting into the amp with that setup.

Can't you use a probe to monitor the sig gen signal at the exact same point you're injecting the signal into the amp?

[snoof]

output on sig gen is constant

But your pics show the bottom trace is approx 1 division, then approx 1.5 divisions, and finally over 3 divisions. That's not very constant.

that's the problem!  I'm not altering the output in any way.  It's happening by itself.  I am only turning up the vol on the amp.

laptop sig gen, stereo out, Lchan amp, Rchan scope.  Same results with batt powered buffered homebrew sig gen.

You're not actually monitoring the same signal that you're injecting into the amp with that setup.

Can't you use a probe to monitor the sig gen signal at the exact same point you're injecting the signal into the amp?

Yes, and I have done that, the results are the same.  Probe plugged into low input jack(so in parallel with the reg input jack).  I gotta be doing something wrong here    I guess i'll have to throw down for a good sig gen and see how that behaves.

[Fresh_Start]

What are you using to trigger the oscilloscope?  Usually you set it on "Auto" and have it trigger off Channel 1.  If that's the case, you need to connect the signal generator's output to Channel 1, not Channel 2.  Forgive me if I misunderstood your posts.

A drawing showing exactly how the signal generator, the amp and the oscilloscope are connected would work wonders.

Chip

[snoof]

trigger is set to chan2 which is where the sig gen is plugged in.  Setting is auto.

[sluckey]

I don't think you're doing anything wrong. I believe a quality sig gen will solve that problem.

EDIT... Just one more thought. Do you have your sig gen ground and scope ground connected together and also connected to the amp ground?

[snoof]

EDIT... Just one more thought. Do you have your sig gen ground and scope ground connected together and also connected to the amp ground?

Yes indeed.

[Fresh_Start]

This unit seems to be working fine for me and I don't think the cost is out of line:  signal generator

I'm no expert on these things but, as I said, it's working well so far.

HTH

Chip

[snoof]

yep, i've seen that one at AES, and the price isn't that bad.  I'm gonna troll ebay for a bit and see what i can find.

[sluckey]

For the price of that AES sig gen, you can have a pristene HP204D or HP3310B (or similar). If you're patient, you can get either of those for under $40 shipped. I got my fully functional 204D oscillator for $25 +shipping. Had to scrub it a bit to look decent. I also got two non-working 3310B function generators for $15. Fixed both simply by cleaning the power switch. Kept one and sold the other on ebay for $30. I'm not that lucky all the time though, but I figured surely I could make one out of two. I was willing to gamble $50 even for a total loss.

[snoof]

i found a 209 on ebay for $55, so we'll see what happens when it arrives.

[sluckey]

I think you'll be happy with it. Did you get the clean looking unit that comes with the original manual? If not, here's a manual...

http://www.hparchive.com/Manuals/HP-209A-Manual-sn-818.pdf

Also, if you need to replace a knob, I had good luck with a large hole paper punch and a light grey colored tube box. I just put a dab of white glue on a 'chad' and dropped it into the knob. Fit very well. Just an idea.

[snoof]

yep, that's the one i got.  Good idea on the knob insert.

I borrowed a goldline sig gen from work, and tried it out.  The sinewave (paralleled between chan2 and amp as above) did still increase just a bit, and distorted just a hair, not nearly as much as it did with other two sig gens I tried.  Merlin states above that the sig gen wave will deform a bit, but is the slight increase in amplitude normal??

[tubesornothing]

Whats normal is a slight decrease in amplitude as you turn up the circuit, that's just because the gain stage starting to load the signal.  Also as Merlin points out, if you are overdriving the gain stage, you will see distortion in the wave due to grid conduction.  Rarely do you see this on input though - it takes a strong guitar signal to do this.

[snoof]

Hmm.  I'm def seeing a slight raise in amplitude when running scope chan1 from the dummy load, and scope chan2 from the sig gen.  I guess my fear is that something is amiss with the scope (just bought it, supposedly calibrated).

[sluckey]

Give this a try... Connect everything just as you have been doing all along. Now totally disconnect your scope probe from your D/Load. Turn the amp volume up. Does the input signal distort and increase now?

[snoof]

nope.  it's the interaction between the two that's somehow causing it.

[sluckey]

nope.  it's the interaction between the two that's somehow causing it.

OK. That's good to know. What scope do you have? Tell me about your scope's vertical inputs and switching modes. What input modes do you have available and which one are you using? I'd love to see a pic of the vertical section, ie, input jacks, volts/div switches, mode selector switches, etc.

There is one other thing that may be an issue. I'm guessing your dummy load is connected to the OT secondary? Is one side of the secondary connected to the amp chassis/ground? If not, then the signal at the D/Load is floating and you need to use a ground clip with your probe. If the signal is floating, that could be causing some weird symptoms like you're seeing.

[snoof]

There is one other thing that may be an issue. I'm guessing your dummy load is connected to the OT secondary? Is one side of the secondary connected to the amp chassis/ground? If not, then the signal at the D/Load is floating and you need to use a ground clip with your probe. If the signal is floating, that could be causing some weird symptoms like you're seeing.

Hmm.  I've tested various amps, I'll find one that the OT secondary is def chassis grounded and try that amp and see whats what.  I'll try to shoot some pics of the scope tonight.

[snoof]

well, i took the same test rig i use at home to work and used a scope there to run the same test.  Got very little(but still some) deflection of the sig gen signal.  Hmm.  It seems that my scope has a prob??  One this amp, the OT secondary is def chassis grounded.

Edit:  If I lift the grnd on my scope probe, all the deflection/distortion (in the sig gen wave) goes away, nice clean trace.  Am I an idiot and this is how i'm supposed to do it??  I thought the procedure was to read ACROSS the dummy load?

[sluckey]

Edit:  If I lift the grnd on my scope probe, all the deflection/distortion (in the sig gen wave) goes away, nice clean trace.  Am I an idiot and this is how i'm supposed to do it??  I thought the procedure was to read ACROSS the dummy load?

You're not an idiot. But I think you have a ground loop that's acting weird when you connect your probe ground. In a perfect setup where the sig gen has a good earth ground, your scope has a good earth ground and your amp has a good earth ground, you don't need the ground clip in the probe, except in some hi freq or narrow pulse circuits. Your probe is connected across the dummy load in that case even without the ground clip.

I still think this booger will go away when you get your 209A. And that reminds me, there's probably a ground strap on that 209A that will allow you to float the sig gen output, or connect the black banana jack to chassis ground. It may be on the back.

[snoof]

So i guess i'll still get correct results w/ out using the grnd clip on the probe huh?

Yes mine will have a ground strap deal on the back, how should I run that jumper?

Thanks for all the help thus far Sluckey 

[sluckey]

So i guess i'll still get correct results w/ out using the grnd clip on the probe huh?

Yes, most of the time. But if you're looking at a signal on a speaker and that signal has no reference to earth ground, you may have to use the ground clip. Or if you're looking at low level hi-freq signals or narrow pulses with fast rise/fall times, you may need the clip.

Yes mine will have a ground strap deal on the back, how should I run that jumper?

Most of the time you can just run it open and rely on the ground connection at the amp.

Be careful about probing a transformerless amp with any line operated test eq.

[snoof]

Be careful about probing a transformerless amp with any line operated test eq.

Indeed.  I've read about the dangers of the TFless amps...