Why so much noise on my signals?

[Terry Pinnell]

Well, actually, I just went and checked on my scopes. On the Tek 2465A,
with 20MHz bandwidth limiting switched on and with a 1x probe across a 10k
resistor, I see perhaps 0.2mV pp with the same scope settings you're using.
That is, a blurry line rather than a clean trace, but not big enough to
meaningfully measure. Likewise on my Hitachi 20MHz scope.

With the 20MHz bandwidth limiting turned off on the Tek, so using its full
350MHz (but still with a 1x probe, which has 52pf parallel capacitance and
therefore limits the bw considerably), I see about 10mV pp. In my case that
would actually be *stronger* if I shorted the ground clip to the probe tip;
it's a loop antenna, picking up the local TV transmitter on channel 5.

Tonight's a quiet night, though. Normally I see a bit more than that, but
still not as much as your pictures showed. There may still be some
sleuthing for you to do.

Walter: I just repeated my 'no PC' measurement and got a *much* lower
level of noise to the one I reported above. Similar amplitude to
yours. Darned if I know why!

Here's what it looked like on the ADC/PC:
http://www.terrypin.dial.pipex.com/Images/NoiseTest3.jpg

On the standard Hameg 'scope, it looked like you described, and I'd
guess its amplitude as maybe 300 uV pp.

 

[John Woodgate]

I read in sci.electronics.design that Terry Pinnell <terrypinDELETE@dial

Walter: I just repeated my 'no PC' measurement and got a *much* lower
level of noise to the one I reported above. Similar amplitude to yours.
Darned if I know why!

The noise level measured your way can vary with local climatic
conditions. Imagine your short and not very efficient antenna inside a
thoroughly wet (conducting, if not very well) house and then inside a
dry one.

 

[Terry Pinnell]

The noise level measured your way can vary with local climatic
conditions. Imagine your short and not very efficient antenna inside a
thoroughly wet (conducting, if not very well) house and then inside a
dry one.

Thanks, John. I'll repeat the test at intervals and note conditions.

I also noted one other small contribution to the difference. For the
first test I had the 10k resistor in a breadboard; for the second, it
was just isolated in air just above bench, held only by the probes.
The breadboard introduced a little more noise, particularly if one end
of probe was connected to one of its long strips.