3KV Meter Multiplier Resistor

[W9ZSL]

Just double checking.  I scored a Simpson 3KV meter.  Full scale reading is 8 MA.  I will need a 375K resistor, right?  Probably 50 watts to be safe?  I may include a divide ohm in the string so I can calibrate it against my known-value meter.

[ka1tdq]

When I measure high voltage on my linear, I just measure the voltage on the electrolytic closest to ground - then multiply by 10 since there are 10 capacitors.  

My meter only goes to 600 volts, so that's how I hafta do it.

Jon

[KA2DZT]

Are you sure that the meter movement is 8ma full scale??  Your calculation of 375K is correct.  You'll need to use a string of resistors to get to 375K.  The resistors will dissipate about 24 watts.  A lot of wattage for meter multipliers.  8 ma for this type of meter doesn't seem right.

The meter may prove to be more trouble to use than it's worth.

Fred

[W9ZSL]

Nope Fred, it's a brand new Simpson made for Motorola and 8 MA for full-scale.  I already have a 2KV Simpson with the matching multiplier but thought this would give me more "F" factor with a power supply that will probably yield more than 2KV.  Just wanted to make sure the 375K was right.  As I said, I'll probably add a slider resistor in the chain to calibrate it.  I'm going with all the old 3" meters like Collins used back in the day.  Here is the array I had before scoring a new HV meter.  They are grid current, screen current, modulator current, plate voltage and PA plate current.  I might need a 500 MA for the modulator because a pair of 811A's might draw up to 350 MA.

[KA2DZT]

300 ma for the mod current is OK, that's all I use and it works FB.  Try measuring the DC resistance of that 8 ma meter.

[W9ZSL]

Eventually.  I have a bigazz volt meter and the multiplier salvaged from a BC-1T KW transmitter I can use to calibrate the meter starting with 375 K ohms.  Like I said, I was just double-checking my computation.  This meter is flawless.  It will replace the 2KV shown.  All are part of a long-term effort over the past 15 years to get the best parts possible and have everything match.  Almost everything for my project is either new or NOS.  I know I keep changing my mind and when I nail this down (813 modulated by p/p 811A's) it will be exactly right.

[WZ5Q]

Snip...
All are part of a long-term effort over the past 15 years to get the best parts possible and have everything match.  Almost everything for my project is either new or NOS.  I know I keep changing my mind and when I nail this down (813 modulated by p/p 811A's) it will be exactly right.

The collecting, substituting, redesigning, & upgrading of the parts as the project morphs over time is half the fun of the build!

For me, the most enjoyable part is the troubleshooting & head scratching to get all the pieces working correctly after it's built. Although fun as well, the actual construction takes a back seat to the collecting and troubleshooting.

73,

[KA2DZT]

That meter is 8ma like it reads.  My guess is that the meter along with the multipliers was also used as a bleeder load.  The meter is reading the plate voltage and also indicating the bleeder is working.  You should reduce your other bleeder current to take into account the added 8 ma.

Fred

[w4bfs]

That meter is 8ma like it reads.  My guess is that the meter along with the multipliers was also used as a bleeder load.  The meter is reading the plate voltage and also indicating the bleeder is working.  You should reduce your other bleeder current to take into account the added 8 ma.

Fred

It sounds like there is a parallel resistor inside the meter case .... this is actually a safety feature because in the simplest HV metering there is only the big R acting as the multiplier.   If the meter movement should open then the current goes to zero and the full value of HV is now in the meter and can become deadly to the operator .... if you are steady with your hands you can remove the 3 screws (usually flatheads ) that hold it inside ... if you find one in there you can change it for a higher shunt value, if you need to .... do yourself a favor and label 'shunt internal' and info on the back side for future ref.

[KA2DZT]

That meter is 8ma like it reads.  My guess is that the meter along with the multipliers was also used as a bleeder load.  The meter is reading the plate voltage and also indicating the bleeder is working.  You should reduce your other bleeder current to take into account the added 8 ma.

Fred

It sounds like there is a parallel resistor inside the meter case .... this is actually a safety feature because in the simplest HV metering there is only the big R acting as the multiplier.   If the meter movement should open then the current goes to zero and the full value of HV is now in the meter and can become deadly to the operator .... if you are steady with your hands you can remove the 3 screws (usually flatheads ) that hold it inside ... if you find one in there you can change it for a higher shunt value, if you need to .... do yourself a favor and label 'shunt internal' and info on the back side for future ref.

Very good point, I didn't even think of that.  Most likely the reason the meter is made that way (8ma). 

[W9ZSL]

The meter's resistance is 6230 ohms and there is a note on the back "calibrated for steel panel" so I'll just have to play with it a bit.  I have a 750 volt supply with a variac and a calibrated volt meter.  I'm planning on using a 100K bleeder and if this is put in parallel with it, the total bleeder would be about 79K provided that the multiplier is 375K.

          381K X 100K
        _____________
              481K

[N8ETQ]

    Yikes!



   Over 100 watts of bleeder current.

   That 60's stuff was cool but....

   Sure wish you would consider using a string of
300 Uf @ 450 V caps for the filter bank. With a
100K 3 W resister across each cap you could pick
300V off the cap closest to ground and just use
a 37.5K series resistor with it. As Jon had
suggested earlier.

/Dan

[w4bfs]

not so fast .......

I = E/R = 300 / 100000 = 3 mA in each of the resistors across each cap, Right ?  where are you getting the extra 5 mA just to swing the meter ?

maybe I misblunderstand .... by the way .... that additional piece of info is changing this scenario .... the meters resistance is stated to be 6230 Ohms and requires 8 mA for full scale activation so this is basically a 50 Vdc meter with a 3 kV scale .... this needs to be verified with a power supply ....it will change the value of the external multiplier resistor ... when a meter is marked 'shunt internal' , the internal resistance is low (a few ohms typically)

consider this:                                     +
                                  |--------res1------O--------|-----------res3--------- HV             the voltage across res2 will be 50V at full scale and if we want
                                mtr                                res2                                                an open movement safety factor let the current thru res2 be say
                                   |                     _             |                                                  2 mA so res2 = 50/.002 = 25kOhm and now the value for res3
                                   -------------------O--------|                                                  would be   (3000 -50)/.01 = 295k Ohm  ... if the meter movement                                                  
                                         I=8mA -->             gnd                                                were to open then the voltage inside the meter case would be 235V

        res1 is the meter internal multiplier           the meter may still have an internal shunt resistor     I see 30 W of heat in this ckt

[N2DTS]

I make all my own shunts and dropping resistors and its best to use multi bleeder resistors in series and take the voltmeter off of a lower voltage tap.

I have some meters in RF decks that the entire voltage needs to be dropped and its a problem, even with many series resistors.
I try and include a pot so I can adjust/calibrate the meters, and using a dmm, run things up as high as the dmm will take and calibrate at say 1000 volts.

All current meters go in the cathodes, no high voltage on plate current meters.

[W9ZSL]

I'm going back to two separate power supplies.  The 2KV supply will be 12 Hy choke input and 12 MFD; 4KV with just one 100K 250 watt bleeder.  I know that's over-kill but I just happen to have one on hand.  Had I used a pair of 4-65A's in the modulator, then I would have gone with one supply even though I'd be stretching the current rating.  The plate supply for the 813 will do 2250 VDC no load except the bleeder.  I already bread boarded a supply and measured that voltage.  I should end up with at least 1900 VDC under load.  The main question is what will it take to get a correct reading on the 3KV meter?  I considered full scale reading of 3KV requires a current draw of 8 MA in the meter itself through the 6230 ohms plus a 375K multiplier.  The 100K bleeder alone draws 22.5 MA at 2250 volts.

Considering full-scale is with the meter drawing 8 MA from a 3KV supply, a multiplier would be a must since alone the 6230 ohms meter resistance at 3KV would result in a current draw of 481+ MA!  Perhaps the multiplier should be 375K, the total resistance at 3KV to draw 8 MA, but the multiplier probably should be 368,770 ohms or 375K MINUS the internal resistance of the meter.

If in parallel with the 100K main bleeder, the total current draw would be right around 30 MA.

For the 811A's I have 1575 VDC no load.

[KA2DZT]

When you put a meter across another resistor and in this case a string of equal bleeder resistors, the meter loads down the resistor and your reading will be in error.  The meter circuit resistance in parallel with the bleeder resistor has to equal the value of the other bleeders in the string.

My HP DVM reads to 1200 volts.  I made a voltage divider so it can read to 2400 volts  The voltage divider was made with two equal resistors, but the top resistor has a shunt resistor equal to the internal resistance of the DVM.  It works perfect as I can check it against itself by measuring voltage slightly less than 1200 volts.

Fred

[W9ZSL]

The current meters are a whole 'nuther subject!  I have a pair of 300 MA, one for the RF and the other for the AF.  I'll cross that bridge when I get that far.

[KA2DZT]

The meter's resistance is 6230 ohms and there is a note on the back "calibrated for steel panel" so I'll just have to play with it a bit.  I have a 750 volt supply with a variac and a calibrated volt meter.  I'm planning on using a 100K bleeder and if this is put in parallel with it, the total bleeder would be about 79K provided that the multiplier is 375K.

          381K X 100K
        _____________
              481K

6230 ohms makes no sense for a 8ma meter.  The resistance should be more in the range of 10-20 ohms.  Don't worry about the steel panel notation.  It will work fine in a Al panel.

[W9ZSL]

Which is exactly my problem Fred.  My "bench-mark" meter is is a Simpson 3KV model 29 that requires 1 MA to read full scale which measures 88 ohms internal resistance.  The multiplier is 2.993 Meg Ohms and came from a Gates BC-1T...the same source as my 100K bleeder and the meter / multiplier was IN PARALLEL with it.  Since the multiplier was of such high value, it didn't make much difference when the 100K was in parallel.  Three KV divided by 3 megs equals 1 MA for full-scale reading so given age and tolerance, together they read 2.993088 Megs.  Close enough for government work!

Given this other meter needs 8 MA not 1 to reach full scale for the same voltage, if you were to divide 3 megs by 8, you come up with 375 K.  How about that.  

I'm sure when in parallel with the 100K bleeder, I'll have to do some calibration with a pot or slider to compensate for the shunt effect of the bleeder.

[W9ZSL]

Nope, I checked it with two different meters and it's 6230 ohms!  Keep in mind that this was custom-made for Motorola by Simpson, so how and where it was used is a mystery.  It may well have been part of a bleeder chain or maybe a "safety" bleeder in case of a breakdown of the primary bleeder.  Either way it would read the actual voltage.  However, if in the primary bleeder chain, I can't see a value of that in the 375K range.  I'm betting this was in parallel with something else.  In my case, if in parallel with 100K, I'd end up with close to 80K which would draw a comfortable 30 MA...right in the ballpark and should the 100K open, the meter combo would act as a safety.  My plate voltage to the 813 would drop a bit closer to 1900 volts.

Of course, I could always use the big model 29 meter, but I like it more as a test instrument.

[Jim, W5JO]

 Keep in mind that this was custom-made for Motorola by Simpson, so how and where it was used is a mystery. Of course, I could always use the

Those meters, as best I can tell from the pictures, were used in the old tube type, high power base stations, in VHF service  You might query Glen Zook, K9STH about how they were used.  My memory is not that good.

[N8ETQ]

  Glad I went drinkin' with theol...


/Dan

[KA2DZT]

Which is exactly my problem Fred.  My "bench-mark" meter is is a Simpson 3KV model 29 that requires 1 MA to read full scale which measures 88 ohms internal resistance.  The multiplier is 2.993 Meg Ohms and came from a Gates BC-1T...the same source as my 100K bleeder and the meter / multiplier was IN PARALLEL with it.  Since the multiplier was of such high value, it didn't make much difference when the 100K was in parallel.  Three KV divided by 3 megs equals 1 MA for full-scale reading so given age and tolerance, together they read 2.993088 Megs.  Close enough for government work!

Given this other meter needs 8 MA not 1 to reach full scale for the same voltage, if you were to divide 3 megs by 8, you come up with 375 K.  How about that.  

I'm sure when in parallel with the 100K bleeder, I'll have to do some calibration with a pot or slider to compensate for the shunt effect of the bleeder.

Don't confuse using the meter in parallel with one resistor of a string of bleeders and using it with a full multiplier resistor in parallel with the full main bleeder.  Two different things, used with the latter method you don't need to compensate for anything.

As for meter multipliers, a 1ma movement requires 1000ohms/volt.  3KV requires  3megohms counting the 88 ohms of the meter.  Like you say 2.993megs is close enough.

[KA2DZT]

Nope, I checked it with two different meters and it's 6230 ohms!  Keep in mind that this was custom-made for Motorola by Simpson, so how and where it was used is a mystery.  It may well have been part of a bleeder chain or maybe a "safety" bleeder in case of a breakdown of the primary bleeder.  Either way it would read the actual voltage.  However, if in the primary bleeder chain, I can't see a value of that in the 375K range.  I'm betting this was in parallel with something else.  In my case, if in parallel with 100K, I'd end up with close to 80K which would draw a comfortable 30 MA...right in the ballpark and should the 100K open, the meter combo would act as a safety.  My plate voltage to the 813 would drop a bit closer to 1900 volts.

Of course, I could always use the big model 29 meter, but I like it more as a test instrument.

Seems that the meter requires 50 volts to read full scale (6230x.008).  So, subtract 6230ohms from the 375K multiplier and the meter will read correctly.

I think you're right, it was set-up to be a safety bleeder in addition to some main bleeder.

[W9ZSL]

I came to the same conclusion.  Subtract the 6230 Ohms from the apparent 375K and blunder on from there!  I included pix of all my other meters just to illustrate how intent I am to match components to not only do the job but be as impressive as well (hell).  Gorsh, lookit all them cool meters!  I always liked the look of the post WW2 Collins rigs.  As long as I have a meter, use it.