Oscilloscope Noise

[Longhair]

I have my oscilloscope plugged into the same outlet line (circuit breaker <--> wall outlets) as my computer, monitor, power supply, etc. and I happen to be getting what I feel like is a lot of noise. Doing the self calibration did not improve (or make worse) anything so now I am wondering what I would be able to do to clean up the noise from the AC outlet. Rewiring the house is not an option.

I took a picture with nothing plugged in to the osilloscope. This is the Rigol DS1052E 50MHz (not hacked) that is very common on here.

[Simon]

is this 200mV/Div the setting you took the measurement on ? if not (you have zoomed in) it is normal, if it is yes that is a bit much, I was a little disappointed with the noise on my scope but this is inherent in digital scopes.

[Longhair]

I just did another one (200mV / 200ns) and it doesn't look like that so I must have zoomed in.

I guess that it is good to hear that it is normal - at least it isn't broken.

Thanks for letting me know.

[Simon]

looks  just like mine with or without the hack, I suspect that SMPS is part of the problem, when I get around to screening the motherboard I'll let the forum know of the effect if any

[cybergibbons]

Is the Rigol normally that bad? 0.8V PtP noise?

You should really ground the probe when look for noise however.

[Simon]

It seems to be a combination of the normal digital sampling issues you get on any digital scope (nature of the beast) and I beleive some power supply noise although that is disputed. I once made up an aluminium screen that helped a bit (I had to recalibrate as the trace was offset negatively) but ideally it should be steel which I'm yet to do.

The noise has roughly a constant VDiv/Vnoise ratio however this does increase on the lowest 2 or 3 V/Div settings which is where i think the PSU noise is coming into play

[Zero999]

The signal appears to be digital so it looks like the noise is within the sampling margin. In other words, suppose at 200mV/division and 8-bit per sample the full scale deflection is +/- 410mV, the quanta would be 410/127 = 3.228mV. If you zoom in 120 times you now have 1.5mV/div so each quanta now takes two squares so 2mV of random noise looks like a squarewave.

[Simon]

on the lower 3 scales the noise is significant although it tend to stay within +/-1 pixel so yea pretty much sampling error, but at the lower scale it does creep up considerably so I think there is some pickup there although not a lot

[chscholz]

Yes, around 0.8 Vpp amplitude noise??? That is HUGE!

What happens if you terminate the inputs to ground?
If the noise is still that high, I'd say the instrument is borderline useful.

Is the Rigol normally that bad? 0.8V PtP noise?

You should really ground the probe when look for noise however.

[alm]

Yes, around 0.8 Vpp amplitude noise??? That is HUGE!

What happens if you terminate the inputs to ground?
If the noise is still that high, I'd say the instrument is borderline useful.

Depends on the settings. Assuming it was set for a 10x probe, the actual voltage is 80mV. If the scope was set to 1V/div (10V/div on display), +/- 40mV is about 1-bit. It certainly looks like +/- 1 bit to me.

[chscholz]

Not familiar with the GUI of this scope but it looked as if he was using 200 mV/div setting. I am also used to scopes auto-detecting probes, so yes, if there is a passive probe attached that might be all good or at least not so bad.
To get a rough estimate of the scope noise one would certainly remove the probes, some would call this "baseline noise".

Depends on the settings. Assuming it was set for a 10x probe, the actual voltage is 80mV. If the scope was set to 1V/div (10V/div on display), +/- 40mV is about 1-bit. It certainly looks like +/- 1 bit to me.

[Simon]

if you take a snap shot of the noise and then zoom in you will get that, the noise will be on the screen about the same size on all ranges as it is +/-1bit, so in effect the bigger the V/Div the higher the noise as it remain a constant percentage of the V/Div

[Longhair]

Just so there is no confusion, I just took a couple of pictures of the current noise. In the very first post, the noise was zoomed in but in the following 2 pictures below, this is how it looked like when I pressed the Run/Stop button.

[Simon]

perfectly normal

[sonicj]

let me guess, either you took over your wife's sewing table or you're into scratch building rc airplanes... 
-sj

[Mechatrommer]

let me guess, either you took over your wife's sewing table or you're into scratch building rc airplanes... 
-sj

both!