EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: Per Hansson on March 08, 2021, 07:02:42 pm
-
I'm interested in the safety of the following scenario:
I'm measuring a 400VAC line to line motor brake in a CAT-III environment.
As the motor brake contactor disengages we get a high voltage pulse.
I privately own a Fluke 225C scopemeter that I really don't want to damage.
And via my employer I have bought a Fluke 28-II
I have seen the DMM register a peak of 1099VAC at the time the contactor disengages.
From the Fluke 28-II manual the peak detect mode works at 250µS.
But the DMM has also just as many times just shown overload (OL) reading
Now obviously these meters are CAT-III rated.
But what kind of probe would I have to use to actually measure the peak properly?
As it is now I guess the meters built in safety devices limit the peak of the pulse...
From Fluke's website their 80K-40 high voltage probe is not recommended for this type of work.
But I am having problems finding what to search for to find properly rated equipment.
What I am trying to do is create a snubber filter to reduce the peak.
I have so far tested Littlefuse TMOV's and they really help.
But even so the machines bus system (Profibus-DP) still gets interrupted by the HV pulse intermittently.
And so I need something to better help me measure the pulse properly.
Both without affecting the pulse or destroying my equipment...
-
My first thought is to use a step down transformer. Reduce the signal by, say, a factor of 10 and then it gets easy. The problem lies in faithfully transforming the waveform. If you know the pulse risetime you can perhaps design a transformer to do this.
Failing that, maybe a resistive divider. The details become important. Using two high voltage probes with an oscilloscope in differential mode could work.
-
I'm interested in the safety of the following scenario:
I'm measuring a 400VAC line to line motor brake in a CAT-III environment.
As the motor brake contactor disengages we get a high voltage pulse.
I privately own a Fluke 225C scopemeter that I really don't want to damage.
And via my employer I have bought a Fluke 28-II
I have seen the DMM register a peak of 1099VAC at the time the contactor disengages.
From the Fluke 28-II manual the peak detect mode works at 250µS.
But the DMM has also just as many times just shown overload (OL) reading
Now obviously these meters are CAT-III rated.
But what kind of probe would I have to use to actually measure the peak properly?
As it is now I guess the meters built in safety devices limit the peak of the pulse...
From Fluke's website their 80K-40 high voltage probe is not recommended for this type of work.
But I am having problems finding what to search for to find properly rated equipment.
What I am trying to do is create a snubber filter to reduce the peak.
I have so far tested Littlefuse TMOV's and they really help.
But even so the machines bus system (Profibus-DP) still gets interrupted by the HV pulse intermittently.
And so I need something to better help me measure the pulse properly.
Both without affecting the pulse or destroying my equipment...
You need to get a high-voltage scope probe which has a 100:1 or 1000:1 division ratio to reduce the peak voltage to a safe level for your Scopemeter. A probe designed for scope use will have better bandwidth than a probe, such as the Fluke 80K-40, designed for use with a multimeter. You will probably need at least several MHz of usable probe bandwidth to be able to see the fast transients of the waveform and to determine how well your snubber is suppressing those transients.
Another option is a high-voltage differential probe as this will allow you to make floating measurements and not require the scope common be connected directly to the circuit common. These probes are also useful for measuring signals in motor controllers etc.
There are plenty of these types of probe on the market but it may be challenging to find one that is Cat III rated. Expect to spend a quite bit of money though.
-
Okay thanks, after another Googling session yesterday evening I found the 100:1 Fluke VPS220 scope probes.
I'm not 100% sure yet if they where designed for the old 190 series Scopemeters (like mine is) or the new 190 series II Scopemeters.
I guess they should be compatible, probes for the newer scope do seem cheaper.
What about the DMM, I understand the problem is where to clip the ground lead.
Is there nothing cheap at all available for multimeters to handle high voltage transients?
And what I mean about cheap is not exactly that: but all I found is those Fluke probes for working on CRT monitors.
Or the super high voltage probes rated from 10 to 400kV, not exactly useful to me, or my wallet :D
-
100X probes sound like what you need, you don't know how high that spike is. Is it causing some other problem?
Is this 400VAC 3-phase and ground-referenced or is it isolated? If it is ground-referenced, you can start by just scoping from ground to each phase and see what that looks like.
-
100X probes sound like what you need, you don't know how high that spike is. Is it causing some other problem?
Yes, the machines bus system (Profibus-DP) gets interrupted by the HV pulse intermittently.
I thought I had found the culprit: the shield was not connected in the cable ends, after doing so it improved allot.
But still there are some intermittent faults just much less common so I need some way to measure this properly...
Is this 400VAC 3-phase and ground-referenced or is it isolated? If it is ground-referenced, you can start by just scoping from ground to each phase and see what that looks like.
Yes it is ground referenced, and actually in the photos attached the brake is not 400VAC but 230VAC.
However it is connected to a virtual neutral formed by the three phase motor setup in WYE configuration:
So there is a common neutral point and that is where the brake gets his neutral from, this is only ca 16v above PE.
But there is also machines where the brake is true three-phase and it then has no snubber.
So I based my question off that, as that is the worse system for me to troubleshoot.
I attached a photo where I previously measured this virtual neutral ground brake:
This was actually not to catch the transient:
I was just making sure it was 230VAC and was too lazy to go to my car and get my multimeter :D
-
I looked at the Micsig DP20003 differential probe because at that price point it is much cheaper than the passive Fluke probes!
But then on an external seller website (Batronix) there is this warning in bold text:
Only connect to a grounded oscilloscope!
What is the reason for that warning, obviously a Scopmeter is not grounded in itself...
Late edit: found this quote by David that seems to explain it really well!
If a differential probe is used with an isolated input or an oscilloscope with a missing ground connection, then the input or the oscilloscope if its ground connection is missing will charge to the common mode input voltage which can be considerable resulting in potential damage to the oscilloscope. Tektronix includes a specific warning with their high voltage differential probe not to use it with an isolated oscilloscope input like on their TPS series. Note that the attenuation of the probe make no difference since the attenuators go to the now isolated ground.
I also searched eevblog and found the video where Joe tears down a P4100 probe, quite horrible Chinese crap!
There is also the P2301C / P2300C probes being sold on eBay, they look just like the Flukes but exterior looks can be deceiving of course!
Then there is the German Testec TT-HV 250 probes, they do look quite nice.
But I noticed they also have a range of probes specifically for Scopemeters called TT-SP 1000
But I found no reviews of them or reputable places that sell them, maybe they are quite new?