Heathkit Condenser Checker

[(*steve*)]

240V mod completed.



I've moved the filter caps to keep them away from the incoming mains, and the chassis is now earthed.

Rather than having tag strips with mains voltage on them, all the primary side wiring is taken directly to the fuse, switch, etc and the exposed connections are covered with heatshrink.



The wiring to the autotransformer goes through one of the existing grommets and passes just behind the fuse.

I have carefully checked the secondary voltages, and in about to fire it up again!

 

[(*steve*)]

Finally I have had some time to do stuff...

I have found that the autotransformer seems to put out a little lower voltage than I expected. This causes all sorts of problems related to the various unregulated voltage rails.

This causes the voltage for the magic eye tube to fall to as low as 120V. They're designed to operate at 250V, the designed voltage in this unit is around 180V.

From my reading I have noted that these tubes can be operated at higher voltages (but low currents) to extend their life.

The power supply is a simple half wave rectified 500V AC that feeds a potential divider. A trap in the potential divider is grounded, and the negative voltages are used for leakage tests. The positive voltage is used only for the magic eye tube.

My first thought was to replace the tube rectifier with a silicon diode. I thought that removing the voltage drop across the tube might be sufficient. It wasn't. I tried a few variations like the one below, but I only got an extra 10V or so.



The second option was a voltage doubler. A voltage doubler will provide a much higher positive rail. Theoretically it could reach almost 900V DC with respect to ground. Practically, it won't get anywhere near that.



I used some 2kV 500mA diodes I have, and a pair of 400V 6.8uF capacitors in series to handle the peak voltages that are possible. Each capacitor has a 1.2M resistor across it to ensure approximately equal voltages, and a reliable discharge path.

This arrangement gives me approximately 550V DC for the magic eye tube.

This is what it looks like now:



Nice and bright! Maybe a bit too bright, I still have to adjust the biasing. I also need to adjust the potential divider to increase the -ve rail (it's about 30V low).

Incidentally, the image above shows the bridge in balance using a precision 100K resistor. It still needs a bit of tweaking, but it's pretty close.

I've realized I don't need to add an indicator to warm that the capacitor is still charged, the eye tube does that too.

Next time I have time to do some more work on this I'm going to make the modification to discharge capacitors through something other than the precision 2k resistor! At that point I'll also fit the new (and somewhat more delicate) precision 2k resistor alongside a big dumb discharge resistor.

 

[(*steve*)]

I did a bit of fiddling after reading up on these tubes. The idea of using a higher target voltage is also to reduce the current in the tube.

The only practical way of doing this is to reduce the heater voltage to reduce the emissivity of the cathode.

After some fiddling and various measurements I decided to add a 2.2Ω 1W resistor in series with the heater.

This results in a target current of about 0.65mA with a heater voltage of around 7.8V. Oh, and with this load, the target voltage is 680V!!!

It looks plenty bright:

 

[(*steve*)]

Here's an update schematic. The changes I've made are indicated in red.



In the voltage doubler the diodes are 2kV 0.5A, the capacitors are 400V 6.8uF, and the bleed resistors are 1.2M.

The 2.2k resistor in series with the high voltage to the 1629 is to allow current measurements.

The main control is now a 50k pot, and the series resistor for the extended capacitance range is increased proportionately.

The next step is to use some zener diodes to create a stable -450V rail, and to change the resistors in the potential divider to get voltages closer to that indicated (2 x 200V 6W and a 51V 1W). The zeners will allow approx 20mA, but I intend to keep the total load to about 2 - 3mA.

With the series resistor for the 1629's heater, it takes about a minute before the voltages around that tube stabilize. I have a diode across the 1626, so the HT comes up immediately. I think I can remove the diode across it without causing strange things to happen.

I think this unit is coming along quite nicely

 

[(*steve*)]

I think I'm almost done!

After changing the voltage divider resistors, I'm getting almost exactly 100% of the marked voltages. They rise to about 107% when switched to the electrolytic setting.



If there's anything else to do, it's probably regulating the high voltage supply.

 

[bertus]

Hello,

@(*steve*) How is the C-3 behaving?
I found a complete manual here:
https://www.vintage-radio.info/download.php?id=45

More heathkit manuals and schematics can be found here:
https://www.vintage-radio.info/heathkit/

Bertus

 

[73's de Edd]

Well . . . . . . since this thread got brought up again.

And since you mentioned your HIGH voltage level that you ended up having to use on that tuning eye tube.
(That's being almost a klystron tube beam voltage activating levels . . . . . in my estimation.)
Will you check out something . . .on that unit ?
Is it using a round metal shell that is holding / surrounding that tuning eye tube socket ?
If so, will you look inside that shell and see if you don't find that there is being a " hidden" 1/2 watt CARBON COMPOSITION
resistor of an initial ~ 1 meg resistance . . . . . .that with time-age-use has NOW transitioned / shifted on up into the tens of megohms of resistance.
Now, wouldn't that explain the need of that present cosmic supply voltage ?

 

[(*steve*)]

Hello,

@(*steve*) How is the C-3 behaving?

It's behaving quite well

Another project was adding a method of reading the actual leakage current in a safe way. This is a two edged sword. On the one hand, I can tell you to a fraction of a uA what the leakage current is, but on the other hand it's not a go/no-go test any more :-D

Thanks for the links, I'll check them out.

Well . . . . . . since this thread got brought up again.

And since you mentioned your HIGH voltage level that you ended up having to use on that tuning eye tube.
(That's being almost a klystron tube beam voltage activating levels . . . . . in my estimation.)
Will you check out something . . .on that unit ?

I sure will

Is it using a round metal shell that is holding / surrounding that tuning eye tube socket ?
If so, will you look inside that shell and see if you don't find that there is being a " hidden" 1/2 watt CARBON COMPOSITION
resistor of an initial ~ 1 meg resistance . . . . . .that with time-age-use has NOW transitioned / shifted on up into the tens of megohms of resistance.
Now, wouldn't that explain the need of that present cosmic supply voltage ?

I'll take a look, however I replaced the existing resistors if they were out of spec.

edit: take a look at the photo in post #3 in this thread. No metal shield.

There was an initial fault that made the tube much brighter than it should have been. I think this damaged the tube. When adjusted to operate at the correct current, the display was almost unreadably dim.

Some research indicated that a higher voltage and a lower current could restore operation at the cost of reduced tube life. The higher voltage worked, but at the expense of (again) hugely increased current. Because this tube has no method of changing the gain, the solution was to reduce the heater voltage. This reduced the current to significantly below the normal value, but with higher brightness.