Questions about a simple HV scope probe (20kV - DC and AC 50Hz) / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Questions about a simple HV scope probe (20kV - DC and AC 50Hz)

1 2 3 next

Move Thread

LAN_403

PhilGood

Mon May 27 2013, 02:25PM

Registered Member #3806 Joined: Sat Apr 02 2011, 09:20PM
Location: France
Posts: 259

Hi all ,

Iâ€™m planning to build a simple HV scope probe for measuring DC or low freq AC (50-60Hz) up to 20kV with a 1:1000 ratio

As I wonâ€™t deal with high frequencies my goal is to keep it as simple as possible, and my plan is to make a simple resistive divider

I found a 200MÎ© 30kV 12W 1% resistor on Ebay that I would use as R1

Resistor datasheet (SSX78 model)

For R2 I would use a resistors arrangement for a few watts and 250k (which is 200k with the 1MÎ© scope resistance in parallel)

Current at 20kV DC would be 0.1mA and dissipation 2W

And then put everything in a 1" sealed PVC pipe filled with olive oil with a tip at one end and the cable to the scope and the alligator ground clip at the other end

Do you think that would work, and what kind of coaxial cable I should use for this (considering the scope will be on 1MÎ© impedance mode)?

Thanks for your advices :)

HV Enthusiast

Mon May 27 2013, 03:53PM

Registered Member #15 Joined: Thu Feb 02 2006, 01:11PM
Location:
Posts: 3068

For low frequency use like that I think you can get scope probes for HV on ebay for very little.

PhilGood

Mon May 27 2013, 04:23PM

Registered Member #3806 Joined: Sat Apr 02 2011, 09:20PM
Location: France
Posts: 259

Thanks EVR for the suggestion

But I can't find any like the one I want to build

The best match I found is a 20kV 20MHz probe, but it is $215

If you find a cheap one for mains frequency please link it...

My goal is to measure MOT's, NST's and other HV transformers eventually with rectification / voltage multipliers / smoothing

EDIT:

I found this intersting article:

In the example of a probe connected to a 1MÎ© scope input, it says "The cable is a high-impedance cable in this case"

I am wondering what exact kind of cable must be used in such case....

Carl Pugh

Mon May 27 2013, 05:07PM

Registered Member #1064 Joined: Tue Oct 16 2007, 05:04PM
Location:
Posts: 42

Get a resistor that you don't have to put in oil, or use resistors in series.
The oil expands and if there's not an air space, the PVC pipe is likely to burst. There may be considerable oil expansion. (From bitter experience)
Calibration at 60 hertz can be checked using a signal generator and oscilloscope and voltmeter.
It is probably advisable to place capacitors across all resistors in the divider. If capacitors are not used and there are high voltage transients, the end resistor will fail then the next resistor and so on.

PhilGood

Mon May 27 2013, 05:47PM

Registered Member #3806 Joined: Sat Apr 02 2011, 09:20PM
Location: France
Posts: 259

Carl Pugh wrote ...
The oil expands and if there's not an air space, the PVC pipe is likely to burst. There may be considerable oil expansion. (From bitter experience)

Good to know that !

I thought I'd use oil for insulation, but maybe I don't need it, or I can use something else ?
It's not gonna heat a lot as I use a 12W resistor for a max dissipation of 2W

And I just thought, maybe the coaxial impedance doesn't really matters for such low frequency as 50Hz, still would like confirmation about this

Steve Conner

Tue May 28 2013, 10:17AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

There is no such thing as a simple HV scope probe.

Coax has a capacitance of about 100pF per meter. The higher the impedance, the lower the capacitance. 75 ohm is the highest easily available. 92 ohm exists, but I've not seen it stocked.

So, if you use 2 meters of coax, your probe will be a low-pass RC filter with a time constant of 250k*200pF = 3.18kHz.

The usual solution is to introduce a small capacitance in parallel with the high voltage resistor. In this case, we need 200pF * (250k/200M) = 0.25pF.

Things now get complicated really fast, because the resistor would typically have at least 0.25pF capacitance between its ends already! But it also has capacitance to ground and to each end distributed along its length. The resulting frequency response can have a practically infinite number of poles, needing a complex compensation network. Tektronix's P6015 probe is a great example of this school of thought.

The brute force solution is to put a large capacitor across the 250k resistor, so you can put a bigger capacitor across the 200M resistor to swamp the stray capacitances. EasternVoltage designed a successful HV probe like this.

Another solution is to use a pure capacitive divider. Resistors have stray capacitance, but capacitors don't necessarily have stray resistance, so you can get a flat response to high frequencies without compensation. Jennings make some beautiful vacuum capacitor HV probes. They can't measure DC, but for many applications that isn't a problem.

PhilGood

Tue May 28 2013, 01:10PM

Registered Member #3806 Joined: Sat Apr 02 2011, 09:20PM
Location: France
Posts: 259

Hi steve :)

Steve Conner wrote ...
There is no such thing as a simple HV scope probe.

Indeed I've seen that reading some threads here dedicated to this subject like Patrick's Project

Thanks for the detailed answer with values and calculations, this is quite instructive.

Steve Conner wrote ...
So, if you use 2 meters of coax, your probe will be a low-pass RC filter with a time constant of 250k*200pF = 3.18kHz.

Isn't that ok for DC and 50Hz AC measurements ?

Would a 200pF of coax capacitance affect a lot a 50Hz signal ?

As for the capacitive divider, I do need to measure DC too

Is the EVR probe you mentionned this one ?

Steve Conner

Tue May 28 2013, 01:52PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Hi Phil

Yes, that is the EVR probe.

An uncompensated probe may indeed be OK for measurements at 50Hz. It would be worth a try.

PhilGood

Tue May 28 2013, 03:14PM

Registered Member #3806 Joined: Sat Apr 02 2011, 09:20PM
Location: France
Posts: 259

Cool, so on an electrical point of view it seems ok

Thanks, I'll try that when I receive the resisitors and let you know how it goes...

My last doubt is about dielectric filling of the probe, I could use olive oil and leave some air to allow oil expansion ? Or use something solid like parafin ? Any better suggestion ?

Tony Matt

Wed May 29 2013, 03:26AM

Registered Member #3700 Joined: Sat Feb 19 2011, 12:59PM
Location:
Posts: 107

Hi Phil !

Your thread make me think about the influence of the cable capacitance, mainly when you are using to test hv at 60Hz or lower frequencies.

It is clear that the cable capacitance presents a capacitive reactance to the 60Hz. And this reactance produce a " load" to the voltage that is going to the meter/oscilocope.

The question is find out HOW MUCH is this load,

The cable capacitance of 200pF running at 60Hz will give a reactance of :

Xc = 1/2*pi*F*C

Xc = 1/(2*pi*60 *200* 10^12)

Xc = 1.33 *10^7

Xc= 13,5megohm.

The capacitive reactance in parallel with the 250Kohm resistor (0.25Mohm)

R = 0.25Mohm >> 1/R = 4 *10^(-6) mho
Xc = 13.5 Mohm >> 1/Xc = 0.074 *10^(-6)mho

1/Z= (4 + 0.074j) *10^(-6) mho

Z= 1/(4+0.074j) *10^(-6)= (0.2499 - 0.00462j) Mohm

|Z| = 0.24996Mohms

ou 249.96Kohms

Conclusion:

The cable capacitive reactance at 60Hz is so high that, in practical way, the mesurement error will be minimum.

And your schematic using only a 200Mohm and 250Kohm resistor will be good for DC and 60Hz with reasonable precision

I think no oil feeling will be better.

1 2 3 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob