Voltage divider with calibration - for measuring Geiger counter high voltage

Some time ago I threw together (what I thought was) a 1000:1 voltage divider for measuring the HV on Geiger counters. This consisted of a Dale 1% 1G ohm resistor (R1) mounted all nice in a box with big chunky gold banana connectors and another quality 1% 1M ohm resistor (R2) connected to a banana socket and then a bunch of croc leads flying about. It worked, but wasn't right. I hadn't accounted for the multimeter's own internal resistance (known as a burden on a circuit). I'm pretty sure my multimeter is a 20M ohm resistance, or thereabouts, so that made R2 actually 950K ohm when the meter was in circuit. So a 1000:0.95 divider. The voltage shown on the multimeter was not that of the HV on the counter. I did some basic testing with and without the divider to confirm this.

As it is essentially a parallel resistor network, a solution was to add some resistance to R2. Increasing that will offset the meter's resistance. If the meter is 20M, then adding 50K to R2 should compensate for this. I decided to build a better divider on proto board that easily plugs in and unplugs from my meter.

 

I added a 100K trim pot and set it so when the 1M and pot were measured in series it read 1.05M. I calibrated it using 338VDC from inside a switch mode PSU and it worked really well. I first measured the voltage without the divider and made a note, then put the divider in circuit and whilst reading the voltage I turned the pot until the voltage was the same as without the divider, divided by 1000. It seems to be working very well. I have included the circuit below. *** I have just noticed that the R3 (trim pot) value in the schematic is 50K. It should read 100K. ***

 

 

 

That's a good idea nu3e. I have seen it done both ways in circuits but never really understood one way or another. I've never really thought about it in all honesty. It makes perfect sense. I shall make that change when I next have my iron hot.

Something I forgot to mention was that R3 could be swapped for a higher value. Say something along the lines of 500K. this would allow the divider to work with lower resistance meters. Using a value such as 500K should allow for meters with an internal resistance as low as 3M, and up to infinity.