I've been getting some rather noisy signals and lots of advice about it. But... First is a 1kHz signal @ 100mV connected by coax from my sig gen to the scope. Nice and clean. Next is the same signal connected by scope probe to mini grabbers on coax. I tried all of my 6 various probes at both 1X and 10X and all 4 of my coax to mini grabber cables and got the exact same thing. So I can only conclude this is an outside EMI causing it. My question is, is this normal and the best I can expect or is there any way to attenuate the noise. I did build the Faraday box because of this but right now it's driving me even nuttier than usual.

 

Does the noise go away if you turn off lights around your home?

That is precisely what happened to me. A LED lamp that had the same format of a 4 foot fluorescent tube was the culprit.
I replaced it with a different brand, and the problem went away.

 

Shift your trigger level to above the waveform and try to trigger on just the spikes, then you should have their frequency and therefore clues to suspect what's causing it.

 

You did something differently. The frequency is not the same.

 

Does the noise go away if you turn off lights around your home?

I waved the probe around with a piece of wire for an antenna and yes, the LED magnifier light on my bench is noisy as hell so I turned it off and it made no difference. Nothing else other then the computer and monitor at close proximity are noisemakers that I have found.

You did something differently. The frequency is not the same.

Nope, all I did was swap cables. The scope is misinterpreting all the noise spikes as frequency I would guess.

Shift your trigger level to above the waveform

Not sure I follow you. Adjust the trigger how? Trigger is set to edge, increasing slope and is the same in either auto or normal. Raising the trigger level all the way to max @ 120mV doesn't change it.

 

Not sure I follow you. Adjust the trigger how?

Set the sweep to 1 uS/div and see if you can sync on the noise spikes.

 

With the trigger level above the fundamental but within the amplitude of the noise waveform use Normal or Single trigger.

 

Set the sweep to 1 uS/div and see if you can sync on the noise spikes.

Sure looks like something switching somewhere.

With the trigger level above the fundamental but within the amplitude of the noise waveform use Normal or Single trigger.

Like this? Nothing changes...

 

I think I can see blips @~50us which puts you in the ~20 KHz range, right where many SMPS operate.
Try turning some OFF even if they're in another room.....suspect any/all of them to be the culprit !

 

I think I can see blips @~50us which puts you in the ~20 KHz range, right where many SMPS operate.
Try turning some OFF even if they're in another room.....suspect any/all of them to be the culprit !

Not sure there are any unless it's a wallwart somewhere. I'll take my little pocket meter and sweep the house and see what pops up.

 

Is this noise bothering a circuit, or is it just an annoyance?

 

Annoying trying to see a signal on the scope while trying to learn about Op Amps. I've even resorted to putting the signal through a voltage follower/buffer to clean it up in order to get a measurable frequency due to the noise. Which doesn't work at the mV level. My question is whether this a normal expectation for mV and lower signals on the scope and what to do about it?

 

I don't see that sort of thing when I use scope probes,
you are certainly getting pick up when you use your probes,
which sort of implies its the probes that are letting in the noise, and the culprit.

I'm using HP probes ( sorry agilent ) , and measuring sub mV,
but I do have very good grounding on the scope probe, not that bit of wire that hangs off the back of them.

 

not that bit of wire that hangs off the back of them.

Actually, I had forgotten about the other grounding option but in the case of what I am doing that is not possible.

 

I don’t think the cause is an external EMF source (the signal is too clean for that).

Think what you have producing a 1kHz sine wave – are you certain that none of your instruments (including the scope) is set to generate/display a 1kHz signal.

A permanent 100mV 1kHz signal makes the scope almost useless.

Personally I think it is a real signal fed to the scope for some reason – is it the same for channel 2?

Can you borrow another scope and connect it to the same circuit?

Measure the signal with a multimeter – it should still respond up to this frequency.

 

The 1kHz 100mV signal was straight from a signal generator. The good waveform was by connecting the signal generator to the scope with coax to eliminate any outside interference. So the noise is not internal, it is from an outside source. Was going to "Wand" around the house with my pocket sized battery powered scope but it is FUBAR so I ordered a replacement that will take 3 months to get here. I had a brainstorm this morning that it could be coming from our wireless dog fence transmitter so I disconnected it but the noise is still there so it's ruled out as the source. Mystery unsolved...

 

Just to be sure.... The signal from the sig gen is fed to the scope AND THE SCOPE GROUND IS CONNECTED TO THE SIG GEN GROUND, correct?

What do you see on the scope if you simply clip the scope probe onto a 12" piece of wire (as an antenna) and nothing else?

 

The sig gen has a 2 wire cordset so no ground and a plastic case. But direct connected to the scope with coax the signal is clean. I can do a 16 "frame" average and get a decent signal but if something changes fast the scope will take time to "catch up".

Probe lying on desk 2-3' minimum away from all sources w/ ~12" wire "antenna".
60hZ hum @ 41.60mV! A few times I have set up an OpAmp voltage follower buffer to clean the signal but then I'm never sure I'm getting an accurate true reading.


Going into that signal closer 400hZ noise @ 17mV


Even closer 100Mhz noise? @ 8.8mV

Doing this there is no ground to connect to and earth grounding the probe ground clip has no effect.

 

Starting at the outer most circumference around the test bench and working your way in is like mapping. A sniffer that is frequency selective can
sort out harmonics. When the major gremlins have been eliminated the more local nearby noise will stand out. Grounding rods usually make a difference. If necessary you can hammer a few rebars in at various locations as temporary grounds when mitigating, This so a screening scheme will be more effective and one ground rod should be as close as possible to the bench. The filters from old computer power supplies (corcom emi/rfi ) are useful on a metal multiple outlet power strip for units that do not have their own EMI filter. and those that do might still need the incoming lines inside a metal conduit within a reasonable degree of shielding not a complete electrical remodel however it has tempted many to resort to drastic measures.
A few feet of conduit might be enough.

 

The two wires coming from the sig gen would consist of one 'hot' wire (with the signal) and a 'ground' wire. This 'ground' wire from the sig gen must end up on the ground of the scope.

The impulse noise seen with the 12" antenna is significantly less than seen when hooked to the sig gen, and that is why I suspect you do not have a good ground to the sig gen.