Maker Pro
Maker Pro

Heathkit Condenser Checker

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Some of the components have arrived. the 200K resistor (measuring 204.34kΩ) was replaced with another one measuring 199.96kΩ.

The 90kΩ resistor (I found it!) was replaced with a pair of paralleled 180kΩ resistor. The change in value was not significant, but at least you can read the value now.

The 2kΩ resistor will be replaced by several 1KΩ resistors in series/parallel as it dissipates significant power in the worst case. Those resistors are yet to arrive.

The already replaced 2μF precision capacitor (replaced with a 2μF electrolytic! that measured 3.88μF) has been replaced with a pair of 1μF 100V capacitors. These were hand selected, and measure 1.999μF.

The 20nF reference capacitor reads as 20.6nF, and the 200pF reference reads 195pF. They're pretty close, and I probably won't bother replacing them.

A big surprise is that the potentiometer used to balance the bridge is 50kΩ rather than the specified 10kΩ. This probably doesn't make too much difference except in the range where a resistor is placed in series with it. I have a 10kΩ wirewound pot on the way -- this too dissipates some power (it's across 55VAC).

I also replaced the 47KΩ resistor that sets the 150V rail.

edit: I think the main problem (calibration-wise) is that the pot in there now turns about 320 degrees, and the scale is designed for a pot that turns about 290-300 degrees.
 
Last edited:

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
What was a mish-mash of knob styles has been changed to predominantly a single style. In particular this makes the normal/leakage switch easier to hold in position.

I may try to manufacture a replacement for the main dial knob to fit in with these.

IMG_20200507_201427_compress3.jpg
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
The replacement pot hasn't arrived yet, so I got to thinking... Why not use a different (larger) value that won't have a power dissipation issue? Because it's a bridge, there's no load when it's balanced, so a modest increase in values shouldn't be a problem.

So, to do that experiment, I'm going to (later today) install a 50k pot, and replace (temporarily) the 90k series resistor with 450k (430k 1% + 20k 1%).

I've also obtained a replacement knob that I will attempt to fashion a long clear pointer for.

The pot I got doesn't have a round shaft, so I can't calibrate the instrument the normal way, but I can rotate the pot itself, and that should work just as well.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Well, the good news is that I'm now able to get this device remarkably well calibrated.

Here is the bridge balanced with a 100.00kΩ resistor, pointing almost exactly at 1000 on the x100 scale.

The magic eye is very dim, and the following photo was taken in very subdued light.

IMG_20200509_182602_compress31.jpg

Still more work to do, but it's looking pretty good.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
That looks better! It's only temporarily attached at present because I seem to have run out of 2 part epoxy.

IMG_20200509_231755_compress66.jpg
 
Looking good.
Saw a write up where the test voltage (25v) was out.
Don't know if it relates to yours or not.
http://www.wb0smx.net/?p=2550

Heaps of info out there if you ever get stuck so a really good one to work on.
Had several old pieces of test gear from the 40's and 50's I gave away to the local TAFE college.
Always wonder if they kept them or simply saw it as junk back then.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Yeah, the 25V is way out. I posted something previously, however with the 47KΩ resistor across one of the filter capacitors replaced the other rail has increased significantly and now the 450V range is closer to 430V. I think I last measured the 25V test setting at close to 40V.

Today I guess the pointer to the knob. I'm very pleased how it came out.

IMG_20200510_125219_compress67.jpg

The pointers we're cut out of various thicknesses of acrylic. I chose to use the one cut from 1.5mm material because it's pretty stiff, but still looks quite thin.

IMG_20200510_125921_compress22.jpg

I put a very thin smear of 2 part fast curing "crystal clear" epoxy around the outer ring of the knob and on the upper side of the pointer. I placed the etched line downwards (I.e. to end up closest to the panel). The dark ring is where the two parts meet, I couldn't help but get more of the epoxy on the pointer section than I wanted, but nobody will see this.

IMG_20200510_125909_compress30.jpg

Finally I left it sitting for an hour to form a bond before I moved it again.

The knob and the acrylic are clearly made of "most plastic", as the epoxy claimed to bond most plastics, and these appear to be well bonded.

I have also extracted a small autotransformer from a shaver adapter and I will use this to convert the condenser checker to 240V operation.

I'll probably add an IEC socket and a mains fuse later this week before rewiring all the mains connections.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
I've recently received my 2k precision resistor, and I note from the schematic that it is also used as a discharge path for capacitors on the leakage test.

Leakage_Discharge-500x382.jpg

Since there can be up to 450V across the capacitor, the peak instantaneous power dissipation can exceed 100W! I don't think that's the way to treat a precision resistor.

Fortunately, it looks like it will be a simple fix to modify the unit to discharge via a separate resistor. I think a 50W 3k3 resistor should do the trick! I can also add a little indicator that the capacitor is discharging.

Leakage_Discharge-2.jpg

Yeah, it shows a transistor, a diode, and a neon bulb as the indicator. Obviously high voltage transistor. The resistor connected to the base would be about 56 ohms.

The neon indicator could have a couple of hundred volts of zener diodes in series with it and then I can use a cheaper transistor.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
The very nice little neon indicators I have come with absolutely no documentation.

IMG_20200516_122536_compress56.jpg

I don't know if they have an inbuilt series resistor, although given their size it seems unlikely, or what voltage they strike at and operate at.

I connected it across my high voltage power supply, with a random series resistor (around 56k), and measured voltage across the lamp & current through it.

It strikes at around 67V and maintains about 55V across it when operating. At 2mA it has 59V across it (so no series resistor). As soon as the voltage falls below about 55V it turns off.

Oops! I have no access to the -410V rail when the capacitor is being discharged (and from a safety aspect, that's good news). But I do have access to the other end of a 55VAC winding. So an alternative circuit is here:

C3-discharge.png

I don't even have to use anything very special for the transistor. And this also provides protection if you connect a charged capacitor with the polarity reversed.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
In parallel with other mods, I'm also converting this to 240V and earthing the chassis.

I'm replacing the mains cord with an IEC connector, and fitting a small autotransformer.

First step is to nibble out the appropriate size hole

IMG_20200516_163836_compress33.jpg

Then clean it up.

IMG_20200516_171028_compress11.jpg

And it kinda fits

IMG_20200516_171043_compress89.jpg

But I have to rearrange some internal wiring and fit a fuse. I'll also need to nibble out some of the case to allow the connector to protrude.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Hmmm, this looks better

C3-discharge2.png

No need for the transistor to carry the discharge current.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Things still moving slowly. But I have replaced the old terminal posts. I straightened out the panel, but the post to the right still doesn't sit straight. I might have to apply some delicate persuasion to it.

IMG_20200520_163717_compress48.jpg
 

bertus

Moderator
Hello,

Does the post have a rim or slot on the back?
That might be the cause of the slant.

Bertus
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
I think the device was dropped before I got it. The front panel is actually bent a little. I straightened out out a bit, but clearly not quite enough.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
A small step forward. Creating clearance for the power socket.

IMG_20200524_142305_compress89.jpg

The hole is a bit rough

IMG_20200524_142154_compress5.jpg

But it does the job. And the mounting holes for the back panel line up nicely.

One of the problems with the original case was that it was kinda tricky to line up those holes. With the socket there it is much easier.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
It's a cold, windy, and rainy day, so I'm working inside and looking nervously at the weather radar for the forecast 100km/h winds and heavy rain.

So here's the one remaining foot the case had.

IMG_20200524_152301_compress94.jpg

Replaced with some adhesive rubber feet.

IMG_20200524_152645_compress9.jpg

The old remaining foot was hard and ready to fall off, however the case seems twisted a bit, so I'm going to have to check later to see if it straightens out and sits flat when I put it together again.
 
Top