Proper HV probe design for DMM?

Thread Starter

smilem

Joined Jul 23, 2008
141
Hello, I found these two videos describing HV proble design for a DMM.

1.
2.

1. Uses 200M and 200K resistors
2. Uses 2x 20M and 40K resistors (the 40K is under rated should be like 2W)

Questions:

What is the better probe design? The 200M resistors are expensive and hard to find. So the second design is way better to me. Also it uses resistors on both wires meaning you can't mix wires.
 

ronsimpson

Joined Oct 7, 2019
344
Did watch the video1. In most applications the RrR design leaves the meter at 1/2 voltage. If you are looking at 0V and 40kv the meter is at 20kv. Not a good idea. Now if you are looking at a project with -20kv to +20kv on the two wires then it will work fine. For measuring my high voltage it is a bad idea.

Using 200mohm or 20mohm depends on power lost in the resistor and how much current you can pull off the supply.
 

Hymie

Joined Mar 30, 2018
721
Commercial high voltage probes will use the design principle shown in the first video. The important thing to be aware of when using such a probe is that the common input to your DMM will be at whatever potential it is connected to in the HV circuit. For this reason you should only make circuit connections with the HV circuit de-energised with any residual capacitance discharged.

As ronsimpson says, the design principle shown in the second video will result in the meter being at ½ the HV supply. Interestingly in the second video the measurement appears to be being made by a class I scope (earthed 0V input). Therefore the measured result would be in error if the HV circuit was referenced to earth.

The advantage of using the two resistor method is that there is a high impedance between both DMM inputs and the high voltage circuit; the disadvantage is that the internal meter circuit will be a ½ HV supply, which could cause measurement error due to stray capacitance etc.

There will be no shock hazard as a result of the meter being at ½ HV supply, since the maximum current through 20MΩ, connected to a 20kV supply will be 1mA.

If you search this ebay seller’s shop for ‘glaze’ (for glazed resistor), you will find a large range of HV resistors of varying values and power ratings.
https://www.ebay.co.uk/sch/m.html?_ssn=gadgetskingdom&LH_PrefLoc=&_from=R40&_trksid=p2499338.m570.l1313.TR12.TRC2.A0.H0.Xglazed.TRS0&_nkw=glazed&_sacat=0
 

ronsimpson

Joined Oct 7, 2019
344
This is about what I have. 40kv. You connect the clip to ground. The meter stays near ground. The handle is near ground. They have been used for longer than I have been alive.
1576368128160.png
I like the ones like this: The HV does not clime up the probe as bad.
1576368264427.png
 

ronsimpson

Joined Oct 7, 2019
344
There will be no shock hazard as a result of the meter being at ½ HV supply, since the maximum current through 20MΩ, connected to a 20kV supply will be 1mA.
Wikipedia said you can feel 1mA. I have a real aversion to having my meter at 10 to 20kv. Just like I hate to cut the ground wire on my scope and connect the case to 220vac. Yes people do that all the time.
 

SamR

Joined Mar 19, 2019
1,506
There are, to me, some things it doesn't make sense to home-brew. An HV probe is one of them. I would much prefer to buy a commercial product that has gone through compliance testing by a well-established organization than place my life in the hands of some guys whizbang untested idea of what an HV probe should be. Caveat Emptor when it comes to this one.
 

Hymie

Joined Mar 30, 2018
721
Wikipedia said you can feel 1mA. I have a real aversion to having my meter at 10 to 20kv. Just like I hate to cut the ground wire on my scope and connect the case to 220vac. Yes people do that all the time.
Because you can feel an electric current does not mean the current represents an electric shock hazard – otherwise TENS machines would not work.
 

Thread Starter

smilem

Joined Jul 23, 2008
141
Thank you for replies.
HV Probes cost too mutch to buy them just for few times to use. The prices of 60USD is outrageous for 6USD resistor.
 
My red plastic KV probe got brittle and cracked, but I didn't replace it because I have only repaired one CRT TV in the last 20 years, and that one needed a video driver chip.

If you're going to make a KV probe for yourself, have something soft behind you in case you need a landing pad and have another person nearby in case you need to be restarted. (He can also double as the landing pad.) I finished my serious mistakes in my early 20's so my heart didn't stop. Now I'm 69 and I would not bet my life on a home made KV probe. Just keep your safeties all in a row!
 

Janis59

Joined Aug 21, 2017
906
Second video in the mail No.1 is dreadly dangerous. Must condemn it, because no resistor over the planet except seme very very specialized allows the voltage over 750 Volt per piece. Most of resistors even 350 or 150 Volts. Thus, the only safe combination is to serialize them some 10...50...500 pieces in series according the voltage You’ve select. For example, one of my projects operated with a 36 kV 5A source what must be safe against 1...10 full short-comes every each minute of it`s life, just such is that specific plasma process standing outside of my will. The solution was larger coil in resonance but damped to criticality let the frequency response is flat, thus it gives only the proper time delay let the automatics may stop the H-bridge gates cycling. The damping resistors, logically has enormous voltage and power stress when the short is coming in, thus I used two coils in series, and 6 plates of resistors mounted on each col 3 in one side and other 3 in other side. Each plate was consisted by 3 lines of resistors parallelly and 150 SMD of size 2520 resistors serially. Thus the later monitoring with thermal camera shown it s trustable enough, but count now a number of resistors - it several thousands of.
 
Top