[jimmc101]

Sorry I hit send too soon, this is the full post

It is with some trepidation that I have to disagree with Bill, but I am sure that your results indicate that your meter is working correctly.

Referring to the circuit of the early meter...

The circuit is arranged such that the sum of the current through R10 and the meter is just sufficient to start to turn off TR1 and hence reduce the drive to the transformer via oscillator formed by TR2/3.

When calibrated this current is 10uA assuming that the meter resistance is much less than 2.5M (R10) (so that negligible current flows through the resistor).

When the load across the test terminals is open circuit the meter current is of course zero and all of the 10uA must flow through R10.
On the 25v range the -ve test terminal is connected to the 'top' end of R1 giving a test voltage of 25v (10uA x 2.5M). As a load is applied and current flows through the meter the current through R10 to keep the total constant.

On the higher ranges more resistors (R 9, 8, 7, 4, 5, 6) are added in series with R10 to give a total of 100M on the 1000v range.

If the feedback via TR1 was perfect so that the current was always exactly 10uA (after calibration) then the open circuit voltage across the test terminals would be that given by a 10uA current source shunted by a resistor (RS) equal to the range (V) x 100k.
eg for the 1000v range 10uA x 1000 x 100k = 1000v

Adding a load across the terminals drives current through the meter and the current through R10 must decrease to keep the sum constant

This load (RL) changes the voltage to 10uA x (RS//RL).

Calculating the result 0f a 10M load on each range gives
25v______20.0v
50v______33.3v
100v_____50.0v
250v_____74.3V
500v_____83.3v
1000v____90.9v

Remember these figures assume perfect feedback (infinite loop gain) in real life the actual figures will be slightly different and I'm sure your results show that everything is working as it should.

As a final confirmation my meter gives very similar voltages to yours.

Jim