Why would a scope suddenly give noisy traces when it never used to?

Status
Not open for further replies.
T

treez

Guest
Hi,
What are the reasons for a scope’s traces to suddenly get noisey (fuzz all over them)?
I am using the Tenma 72-8705 scope.

Today some offline LED driver PCBs came up from production having failed. I didn’t get too long to look at them before I got called away…however, as I scoped the nodes of these PCBs, I noticed that there was loads of noise “fuzz” on the waveforms. …I could barely see the signals. I have used this scope on this same board for months and months and never seen it noisy like that before . I wonder if someone has accidentally dropped the scope or connected it to a high voltage with one of its 10:1 scope probes and broke the scope like that?.....making it go “fuzzy”. Perhaps somebody connected it to a non-isolated mains LED driver and forgot to use the diff probe?
I tried using a diff probe instead, but the fuzz was still there (and it never was before when I used that diff probe with this scope). Also, the 10:1 probe I used has its lead wrapped several turns through a torroid. Also, I checked that the scope had “bandwidth limit” selected. Also, the waveforms were all still fuzzy whether the pcb was supplied from our usual AC source, or with the non-isolated mains and diff probe. This is the first time in 2 years I have seen this on this scope.
I will have a look on Monday and scope other things like PSU outputs to see if the “fuzz” is still there.

In the meantime, do you know possible causes of a scope giving noisy traces all of a sudden?...in situations where it never used to be “fuzzy” before.

Tenma 72-8705 oscilloscope:
**broken link removed**

TA 041 DIFF PROBE:
https://uk.farnell.com/pico-technology/ta041/probe-active-differential-powered/dp/1667343
 

Is the noise there with the scope tip grounded or connected to the scope calibration signal?
 
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
Missing from the description, is whether this is confined to
the batch of "bad" boards or whether the "fuzz" now also
appears on boards that formerly tested "good" and "fuzz
free" on the bench.

There has to be the possibility that the "bad" boards are
"bad" because of some internal EMI aggressor, as well as
the possibility of an internal 'scope fault. Which ought to
be testable by finding another 'scope to plug the butt end
of the presently attached probes, to.

Boards might (say) have a different vendor's decoupling
caps with inferior ESR/ESL; might be missing a ground
jumper connecting your test point adjacent ground post
to the real ground plane; might include an unfortunate
design revision. All that kind of thing. You have to figure
out some "cut-planes" across the various dimensions of
failure, to get to a tractable problem-space. For starts
that could be two of everything, not from the same batch
any of it, and proceed with the matrix of observations as
you plug, and play.
 
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
broken ground wire some where in the scope probe - this is quite common ...
 
Reactions: treez and FvM

    FvM

    Points: 2
    Helpful Answer Positive Rating
    T

    Points: 2
    Helpful Answer Positive Rating
broken ground wire some where in the scope probe - this is quite common ...
Thanks, the thing is the noise is still there when i use the TA041 diff probe....i wonder if due to some accidental misconnection to non-isolated mains, the ground inside the scope itself is compromised and therefore causing the noise?
 

I don't see much data on the scope.

I am just giving general guidelines, i see these will help to distinguish noise is actually coming from board or scope.

one of the Common thing to note when we see abnormal behavior with SCopes:
After power ON let it warm up for 15 mins.


Below few things to make sure scope+probe are in good condition
1. Auto calibration or Hardware check of the scope - without connecting any probes to channels.
2. Compensate probes/Check probes using internal generated square waves.
3. Check the probes by sending a known signal from function generator or from any other card
 

Status
Not open for further replies.
Cookies are required to use this site. You must accept them to continue using the site. Learn more…