High voltage meter shunts?

[N2DTS]

I fixed the HV meter in the big rig again today.
The meter was reading low, about 1/2 voltage.
This has happened before, last time I used six 1/2 watt metal film resistors in series, and they still measure correctly, but have splits in them, they open up under voltage or something.

This time I used about ten 2 watt carbon resistors in series.

What do they do in industry, what is the voltage rating of a resistor, I cant believe 2000 volts to a meter
(very little current) is exceeding the wattage ratings.

2000 volts across six resistors is 333 volts across each one, thats not a lot...

Brett
N2DTS
   

[Pete, WA2CWA]

Maximum resistor voltage depends a lot on the resistor's composition and size. Check the specs on those 1/2 watt metal film resistors you used.
As an example for some carbon comp. resistors, see below:
Little Demon®
Carbon Composition Molded Resistors
Series Watt   Ohms      Tol.     Dimensions (in. / mm)         Max. Voltage            Dielectric VAC
OD    0.25     2.2-5.6M   ±5%     0.276 / 7.0 0.098 / 2.5               250                          500   
OF    0.50     2.2-20M   ±5%     0.406 / 10.3 0.150 / 3.8              350                           700
OA    1.00     2.2-1M   ±10%    0.591 / 15.0 0.236 / 6.0               500                         1000

[WB2YGF]

I don't know the specifics of the type and value involved here, but film resistors are generally not a good as carbon comp, especially for surge rating.  The old 2W Allen-Bradley carbon comps were the best.  Nobody makes 2W CC's anymore, and specialty HV resistors are MUCH more expensive.  At work, we bought up all the 2W A-B's we could, so we had time to re-design everything that used them.  After about 15 years, we are still drawing on our NOS for some low volume applications.

[WA1GFZ]

Make sure the resistors are not resting on anything. This will limit leakage through the case.

[N2DTS]

The thing is, I tried the metal film resistors last time because the nice carbon comp resistors did the same thing (read low on the voltmeter).

Something is up here though, 5 resistors on 2000 volts is only 400 volts a resistor, I have seen resistors in other circuits running voltages like that forever without problems.

Brett
N2DTS

[WA1GFZ]

What is the dissipation of each resistor. Try to keep it under 1.4 Watts for a 2 watt resistor. Steady state load you want to be well below rating. Resistors derate with heat check the curver for the part you are using.

[Jim, W5JO]

There are ceramic resistors out there for this purpose.  Can't remember where they are, but they are designed for this purpose and other HV circuits.

[WB2YGF]

The thing is, I tried the metal film resistors last time because the nice carbon comp resistors did the same thing (read low on the voltmeter).

Something is up here though, 5 resistors on 2000 volts is only 400 volts a resistor, I have seen resistors in other circuits running voltages like that forever without problems.

Brett
N2DTS

Brett.  Could you PLEASE tell us what the resistor value and wattage rating is (and exact mfg p/n, if possible).

[WB2YGF]

There are ceramic resistors out there for this purpose.  Can't remember where they are, but they are designed for this purpose and other HV circuits.

One of the carbon comp subs we use that is reasonably priced, is the KOA ceramic HPC series.

The 2W has a working voltage of 400V, an overload rating of 800V, and an impulse rating of 25kV

http://www.koaproducts.com/english/catalogue/hpc.htm

http://mouser.com/Search/ProductDetail.aspx?R=HPC2C100Kvirtualkey66000000virtualkey660-HPC2C100K

[Opcom]

I my multipliers for voltage meters, I prefer not to allow as high as 1.4W in a 2W resistor.  Generally I allow 1/4 the rating, so only 1/2W in a 2W jobbie. It cost a little bit more, or makes for more resistors, but then they don't fail or change value. And I observe the voltage ratings. It can be deceiving. I might get scolded but I put a small resistor in series with the wirewound bleeder (which itself is oversized on wattage) and then am looking at about 44V across that, from which any inexpensive resistors will do for a multiplier.

Specifically for an example, used two 17K 225W WW resistors in series from the 3KV HV, and then a 500 Ohm 20W from the bottom end of that to GND. So the bleeder is 34500 Ohms. 270W of dissipation for 450W of resistor is 60%, like 1.2W per 2W in the terms you are using. - but those are large ceramic WW power resistors designed for high voltages instead of the smaller type. A heavy bleed @90mA and others might prefer less.

The reason for the 500 Ohm being 20W is for safety so it will likely never open. I get 44V across the 500 Ohm job so it dissipates only 4W, then use a 50K pot feeding the meter from that tap to set the meter to 0.6mA which reads 3KV on the 5KV scale. So the meter is in parallel with the 500 Ohm resitor. It takes a little more space for the bleeder but they last forever and that's important to me. Just my 2 cents since you are going to replace the multiplier string and have had the problem before.

If your bleeder is robust, putting a small and over-wattaged resistor in its tail and tapping off there at a low voltage might solve the problem without having to buy a string of costly HV-rated resistors.

[N2DTS]

Well, last time I had three 1/2 watt metal film 1 meg, one 2.2 meg metal film 1/2 watt, and one carbon comp 1 meg 1 watt resistor all in series to the meter, measuring 2000 volts.

Thats five 1/2 watt resistors for a meter movement, surely the current cant be very high for a meter.

I have ten 2 watt resistors in there now....

Brett
 

[WB2YGF]

Well, last time I had three 1/2 watt metal film 1 meg, one 2.2 meg metal film 1/2 watt, and one carbon comp 1 meg 1 watt resistor all in series to the meter, measuring 2000 volts.

The 2.2M would have had 700V across it, with 1/4W of dissipation.  My guess is that the 2.2M 1/2W metal film would not take the voltage for very long.

If the resistors are not the same value, the higher value will see more voltage and dissipation than the others.