As noted in my previous post, the 25 volt scale for the leakage test function of the Heathkit C3 Condensor Checker provided a voltage that was considerably higher than 25 volts.  It was closer to 50 volts.  This seems to be a design flaw, or a documentation flaw, as I don’t know which side of the team dropped the ball.  I chose to treat it as a design flaw, since the documentation and the front panel screen printing indicated that it should be a 25 volt scale.

25 Volt Mod in place

To remedy this problem, I chose to modify the voltage divider in order to provide the 25 volt output, while keeping the rest of the voltage ranges pretty much the same as originally designed (they were all within 1 – 2% of spec, at modern line voltages).

The 10k-ohm resistor at the top of the voltage divider is the one that is supposed to develop the 25 vdc for the 25 volt range.  The voltage divider is set up to work with about 4.5 ma running through it.

10K Resistor to be replaced

The following table shows the various voltage drops across the resistors in the voltage divider string, as originally designed.  R1 is the original 10k-ohm resistor, and R1a is a place holder for a second resistor that will be included with the modification.  It is set to 0.1 ohm here, to keep the spreadsheet from showing a divide-by-zero error when I set it at 0 ohms for this investigation.

  R1 R1a R2 R3 R4 R5 R6 R_Total Source Volts Current Total
Resistance 10000 0.1 22000 22000 22000 22000 47000 145000.1 655 0.00452
Ind. Voltage Drop 45.2 0.0 99.4 99.4 99.4 99.4
String Voltage Drop 45.2 45.2 144.6 243.9 343.3 442.7
Watts Dissipated 0.20 0.00 0.45 0.45 0.45 0.45

You can see that the 10k-ohm resistor has about a 45 volt drop, consistent with what I measured in use.

25 Volt Tap – original circuit

Now, by substituting a 5.6k and a 5.1k resistor in series for the single 10k, the results show that the voltage drop is a very usable 25 volts, while keeping the other voltage drops close to their original values.

  R1 R1a R2 R3 R4 R5 R6 R_Total Source Volts Current Total
Resistance 5600 5100 22000 22000 22000 22000 47000 145700 655 0.00450
Ind. Voltage Drop 25.2 22.9 98.9 98.9 98.9 98.9
String Voltage Drop 25.2 48.1 147.0 245.9 344.8 443.7
Watts Dissipated 0.11 0.10 0.44 0.44 0.44 0.44

I already had some 5.1k-ohm 1-watt resistors in my stash, but had to order some 5.6k-ohm resistors from China, and wait for the slow boat to arrive.
When I received my order several weeks later, I set about testing the new divider string, using some jumper leads before making anything semi-permanent.

I found that the circuit actually worked best with two 5.1k-ohm resistors in series, rather than one being a 5.6k resistor. That’s probably just from variances in line voltage, and tolerances in resistor values in the string.

25 volt tap – after mod

So here is the final configuration:

  R1 R1a R2 R3 R4 R5 R6 R_Total Source Volts Current Total
Resistance 5100 5100 22000 22000 22000 22000 47000 145200 655 0.00451
Ind. Voltage Drop 23.0 23.0 99.2 99.2 99.2 99.2
String Voltage Drop 23.0 46.0 145.3 244.5 343.7 443.0
Watts Dissipated 0.10 0.10 0.45 0.45 0.45 0.45

So you can see in the new configuration, I simply replaced the 10K-ohm resistor with two 5.1K-ohm resistors in series, and placed such that the tap for the 25 volt range is in the middle of them.  The rest of the resistors and taps remain the same as original.

Old Voltage Divider

New Voltage Divider

25 Volt Mod in place

After the mod was complete, I made measurements at all ranges.  They are still all within 2% of specifications, at 117 volts AC input, as well as out nominal 121 volts AC input.

25 Volt Mod            
  AC Line Input 25 v Range 150 v Range 250 v Range 350 v Range 450 v Range
Original Config 120 vac 121.3 46.5 149.9 254 357.9 460.6
Post Mod 115 vac 116.7 23.4 145.5 244.9 343.3 441.0
Post Mod 120 vac 121.3 24.4 151.5 254.6 357.3 459.2

So, the C3 Condensor Checker now operates within the specifications in the manual, and matching the front panel labels for controls.