Scope probe for low amplitude signals with lots of noise.

Status
Not open for further replies.
C

cupoftea

Advanced Member level 6
Joined
Jun 13, 2021
Messages
3,261
Helped
65
Reputation
132
Reaction score
150
Trophy points
63
Activity points
16,945
Hi,
Does anyone have a scope probe with a 50R resistor on the end?..do you have a part number?......so that when in an offline smps, you are scoping the current sense input say, you dont suffer so much noise in the scope trace.
 

Sort by date Sort by votes
If I'm trying to kill BW I use a lot higher resistor than that.
But I just solder tack one end, clip the witch's hat on the
other and find a place to solder a ground horseshoe for
the ground clip.

You might do better with a 'scope jack and as short as
practical wiring from that to the measurement point-pair.
.
 
Upvote 0 Downvote
I presume, common mode noise is the most severe problem when scoping small signals in a noisy environment, e.g. SMPS. It's not reduced by resistors.

You need differential probes or at least common mode chokes to handle it.
 
Upvote 0 Downvote
Modern scopes have math functions, like digital filtering, that can pull signal
out of noise, if its uncorrelated. Problem is in a typical embedded system noise
tends to be correlated as its driven off common clock(s), in that case you get
diminishing returns depending on level of correlation present in the noise.

Some ref material :




A simple but good design tool is put scope on infinite persistence, channel on AC coupled, set
trigger to edge, a few mV, and let it run. This will capture over time pk-pk noise. Can be surprising
to look at power rails on processors, when you think you have adequate filtering and you find
maybe not so much.

Not all bypass caps equally effective (ESR) for same C value :



OS-CON is polymer tant


Regards, Dana.
 
Last edited:
Upvote 0 Downvote
cupoftea said:
Hi,
Does anyone have a scope probe with a 50R resistor on the end?..do you have a part number?......so that when in an offline smps, you are scoping the current sense input say, you dont suffer so much noise in the scope trace.
Click to expand...
What you're describing is basically just a coaxial cable with a terminator at one end (provided by the scope, if it has a 50ohm input mode) and the "probe" end connecting directly to the DUT. This is something that's straightforward and worthwhile to craft yourself. But I suppose you could find companies willing to charge thousands of dollars for one.

Whether such a thing actually solves your "noise" problem is hard to say.

Anecdotally, in the past, I've found that in such situations a 1x probe often works better than a 10x probe. Both to overcome common mode effects and the input noise of the scope.
 
Upvote 0 Downvote
We often use a 22k SFR25 with 2-3 mm of lead to probe sensitive pins etc, the other end goes to the scope probe,

great for seeing lower freq control signals and similar without injecting noise into that point !
 
Upvote 0 Downvote
great for seeing lower freq control signals and similar without injecting noise into that point !
Click to expand...

Thanks, that sounds like a great ploy...for scoping eg the current sense input of an offline SMPS though, i presume you would say it was not suitable , as it would filter the signal too much?

I'm literally considering making a scope probe, with eg a bit of RF coax with a BNC on one end...then add a small common mode choke in the cable...then bring out the end for the probing? Has anyone done this?
 

Upvote 0 Downvote
" though, i presume you would say it was not suitable , as it would filter the signal too much? "

you can presume what you like - notice I didn't say what you are inferring
 
Upvote 0 Downvote
Thaanks, yes, i mean, it is known there is a series 9 MEG resistor in a 1:10 probe, and there is capacitance in the probe connection, and that RC doesnt filter the signals significantly , so i believe we will be ok.
 

Upvote 0 Downvote

Upvote 0 Downvote
Thanks, yes, i am seeking low noise method of scoping relatively high bandwidth signals like eg the current sense of an SMPS.
The choice between x10 scope probe, with a scoping "precipe" thing, or just using a piece of coax with a resistor on the end.....
...The use of a scope probe in x1, with the end cut off, and soldered to the circuit, seems a poor method compared to using standard coax, since the scope probe on x1 has low bandwidth.
 

Upvote 0 Downvote
Status
Not open for further replies.

Similar threads

Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more…