High voltage meter woes!

[N2DTS]

Since building my amp, I have noticed that (once again) my meters all read different.
I have a volt meter on the power supply, one on the 4cx250b mod deck, one on the ppp 100th mod deck, and the power supply feeds all that plus the new amplifier, but each meter reads different, the higher I go in voltage, the more difference.
They are all close at 1000 or 1500 volts, within 100 volts maybe, but up at 3000 one reads 2500, one reads 3000, and one might read 3500!!!

I use an old DMM to measure up to 1500 volts (the meter says 1000 volts MAX).
I use that to set the meters in everything, but the resistors seem to change in value through the high voltage, over time, even if I use 5 or 6 in series to give the correct voltage reading.

What do they use for this application, special resistors?

What does everyone use to measure 3000 volts?

I have a hv adaptor/probe but it was designed for a type of vom that I dont have, so its uncalibrated...


Brett

[kg8lb]

Are these panel meters ? What is the panel material ?

[WD5JKO]

Brett,

   This is an area where we often fall down, and have an opportunity for improvement. Several folks make stable high voltage resistors where you can use just a single resistor in series with say a 50 ua to 1 ma meter. Here is a link from Ohmite:

http://www.ohmite.com/cgi-bin/products.cgi

scroll down to the High Voltage resistor listings, and look for resistors rated for 5 kv or more across a single resistor.

I would also get one of those hand held 6kv probes that work with a 10 meg-ohm input DVM.

http://www.frys.com/product/1440364

Those probes are a little costly, but vary handy to have around.

Jim
WD5JKO

[N2DTS]

Alluminum panels.
I have used radio shack meters in all my projects because they are cheap (in more ways than one) take a small hole, and I like matching meters.
They used to make 0to 1 ma, 0 to 15 volts, 0 to 500 microamps, and 0 to 500 ma meters, and I just read them as whatever I want, 0 to 1500 volts, 0 to 15 amps, 0 to 5000 volts, etc.

I should get off my butt and make real scales for them, there are computer programs that will do it, and I already remove the scales and use copies on paper for some scales.

In order to make them work, I make shunts or dropping resistors, which works fine but the high voltage does something to the resistors over time.
I used about 4 half watt metal film ones in series, corbon comp ones in series, they seem to all change over time.

I could try about 10 2 watt carbons in series and see if that lasts any longer...

Brett

[N2DTS]

The high voltage series resistors look the bomb, but where to get them, and I would need a large assortment to adjust all the meters to read correctly.
These are not precision meters and need to be calibrated to scale, so I usualy adjust the resistance till I get the correct reading at 1000 volts, or maybe 1500.

I suppose I could go through them all, find the resistance, and order something as close as possible, sounds expensive....

I might try twenty 2 watt resistors in series first....

Brett

[WA1GFZ]

metal film resistors are usually rated for 300 to 500 volts so just build a string of them in series to drive a 1 ma meter. 4 KV you need 4 meg worth of resistors in series. Leave some extra space arounf the top few since they are running clost to plate voltage. 2 watt resistors are good so they run cool.
8 500k 1 watters would be good. Carbon comps drift around but usable.

[WQ9E]

Brett,

I am not sure how you have your power supply bleeder set up but you can use that as part of your multiplier string for a HV meter.  At the formerly grounded end of the bleeder string insert a small value but high wattage resistor.  The value of this added resistor should be sufficient to get a 3/4 or so of full scale reading on your meter.  For safety, I would suggest using two identical resistors in parallel between the bottom of the bleeder stack and ground (wire wound is preferable) in case one of the two opens.

For your remote mounted meters, the previous posters are correct in that you must either find HV rated units or use a number of regular units in series to stay well within the voltage rating of the individual resistors.  I would prefer using a decent meter movement so calibration is assured by choosing the proper resistor values but you could make one of the resistors in the string an adjustable wirewound for setting the meter.  I really don't like using a regular pot for this in a high voltage multiplier string although it would work.

Rodger WQ9E

[N2DTS]

Great idea Rodger!
In one power supply, I used a bunch of dale 50K resistors in series/paralell, a tap off the low end would reduce the voltage going to the meter dropping string.
The other power supply might use 2 or 4 resistors in the bleed, I will have to check it out.

I dont know why I did not think of that....

Brett

[WD8BIL]

Be sure to put a 100K 2W or so at the bottom of the 4 meg string to complete the voltage divider. Don't use the meter as the bottom of the string.

[KC2IFR]

When Jamie and I were finishing my 4x1 I decided to buy all new Simpson meters. If the meters needed a shunt or multiplier I used the ones that came with the meters ( extra cost). I wanted the old look and all meters to match.
As u can see in the pic I achieved my goal. The 2 empty holes were for the cathode current meters for the 833 modulators but I decided to leave them on the modulator chassis front panel.
Checking the calibration of all the meters proved they are right on!

BTW......the 2 meters not labeled are grid current and cathode current for the 4x1 and they were NOT connected when the picture was taken.

[KC2IFR]

Forgot I had another pic.....

[W2PFY]

What do they use for this application, special resistors

Most commercial gear that I have seen use 1 percent precision resistors that are wire wound style. I have one string at my camp that uses 5 resistors that resemble thread spools that are wire wound and are 1 meg ohm each.

[N2DTS]

I have had a resistor for 30 years that sounds like that, dont know where it came from, but it looks like a ceramic tube with 2 little tabs off the ends, and I remember it was 33K ohms, 1%.
I wonder if it is still in the resistor stash....
I always wanted to do something with it, but its no good at RF...or audio...

I need a lot more than that for the hv meters.

Brett

[W1RKW]

Brett,
What Frank suggests below is what I did with the 813 rig here using a 30v meter.  I mounted 6 metal film resistors on porcelain  insulators and the insulators on to a piece of lexan.  The meter has been accurate since putting the rig online almost a year now.  Also, mounted them in an air flow stream to keep them cool.

metal film resistors are usually rated for 300 to 500 volts so just build a string of them in series to drive a 1 ma meter. 4 KV you need 4 meg worth of resistors in series. Leave some extra space arounf the top few since they are running clost to plate voltage. 2 watt resistors are good so they run cool.
8 500k 1 watters would be good. Carbon comps drift around but usable.

[ashart]

Note that meter accuracy specs for a D'Arsonaval movement in a panel meter are commonly in the 3% range, and are specified at the FULL SCALE of the movement.  Thus, if you're using a 5000 volt meter, the reading may be off as much as 150 volts.  The error, whatever it is, exists even if you're reading a lower voltage on the meter.  Thus, if that meter is reading 1000 volts, the actual voltage may be anywhere between 850 volts and 1150 volts, a much larger percentage of error!  This assumes the 3% figure, assumes no mounting error, and assumes no human reading error.  Further, if you're using more than one instrument, and if the errors should be of opposite sign, the errors are cumulative.  Add to all that, any errors in the resistances of a divider network.

73 de w8vr

[N2DTS]

Are you going to leave the holes there?
Maybe put a filiment voltage meter in one hole, what for the other?

The meters must have cost.
Are they still in business?  made in USA?

Brett

[WA1GFZ]

I bet 1% ww power resistors cost a lot. They war a better choice if you can find them.  1,2,3 watters don't go that high.

[KC2IFR]

Nope.....not going to leave the  holes there. Will be using filament voltage meters for the 4-1000 and the 833's.
The meters cost about $70 each.....depending.

Bill

[w3jn]

Be sure to put a 100K 2W or so at the bottom of the 4 meg string to complete the voltage divider. Don't use the meter as the bottom of the string.

A very, very important point Buddly.  If the meter coil opens and you don't put the 100K across the meter you now have full B+ exposed at the meter.  Note that rigs like the T-368 and GPT-750 have their meters behind a piece of plexiglas so you can't touch the zero adjust etc and get blasted.

[KB3DKS]

Most commercial gear that I have seen use 1 percent precision resistors that are wire wound style. I have one string at my camp that uses 5 resistors that resemble thread spools that are wire wound and are 1 meg ohm each.

Brett,
I have a pile of those old bobbin types including epoxy dipped mil spec around somewhere.
Will take a look and get back to you.

Bill, KB3DKS in 1 Land

[The Slab Bacon]

What does everyone use to measure 3000 volts?

For testing purposes, I just use a Simpson 260 on the 5000v scale.

Used but useable / repairable ones have become cheap and plentiful at swapmeets and hamfests. At one fest I bought 4 of them in a box for $5 ! ! All 4 were either working or easily repairable.

[WQ9E]

For testing I use a Simpson 467 with the 5,000 volt multiplier.  The meter along with leads, HV multiplier, and current shunt show up in the matching case at most hamfests.  The military dumped a ton of these about 15 years ago so there are a lot of them out there.   

When getting around these sorts of voltages you really should set the probe in place and set up meter.  Now, without touching the meter or the probe turn on the power and make the measurement.  For the few occasions where I felt the need to make multiple measurements with the power on I used my Tektronix P-6015A 1000X probe and scope which puts a lot of insulation between you and the voltage being measured.

[Bill, KD0HG]

Wirewound resistors will take a lot of voltage.
The BC-610 uses them in series with its plate voltage meter.

[W2PFY]

The BC-610 uses them in series with its plate voltage meter.

I didn't know that the BC-610 had a plate voltage meter. I owned a C model for twenty years and sold it. I now have the use of a D model and neither one has a plate voltage meter. Some people install  them on the lower right hand side where the card holder is mounted on the older 610's.

What model has the plate voltage meter?

[Bill, KD0HG]

The BC-610 uses them in series with its plate voltage meter.

I didn't know that the BC-610 had a plate voltage meter. I owned a C model for twenty years and sold it. I now have the use of a D model and neither one has a plate voltage meter. Some people install  them on the lower right hand side where the card holder is mounted on the older 610's.

What model has the plate voltage meter?

I don't remember the model..It was a beater parts rig and I gave it away to someone that had one for parts. I did keep the 0-3000 plate voltage meter and the wirewound dropping resistors to use in my homebrew rig.

It definitely was a BC-610 and it had a plate voltage meter.