Recently I have begun what might be the start of a nostalgic journey in vacuum tubes. I credit this addiction to Paul of Mr. Carlson’s Lab (You should check out his YouTube channel in case you don’t know him – https://www.youtube.com/watch?v=qqmegXoB7lA – extremely knowledgeable guy with an amazing skill to explain stuff)
The EICO 950B, the resistance capacitance bridge and the capacitor leak tester – I do have other tiny things that can measure capacitance, but nothing compares to this one, or the Heathkit or a high voltage DC capacitor leak tester.
The reason why I got this EICO 950B in the first place is another massive project – the biggest one I’ve undertaken so far – a 1950s TEK 549 (Yes, I am not crazy and I plan to bring an orphaned 549 back to life). It’s been on for a month now, and I’ve spent about 100+ hours on it as of today (weekends, late evenings, day after day). It is a vacuum tube scope and I need a leak tester to check the caps in it. (Watch out for my post on that once I’m done with it)
But I digress. I was looking for a decent heathkit or the EICO 950 to help me restore it when I stumbled upon this little treasure. Sometimes, in the process of looking for something, you come across a few folks, who possessed a wide spectrum of skills and abilities. I don’t want to call it “buying” as I deem it an honor to possess these items left behind by a great soul. I saw a bunch of equipment on sale and when I saw the EICO I asked for a picture of it. I was blown away by how well kept it was, and went there to pick it up and was amazed to see the amount of electronic instruments he had – collections all the way from 1950. It ranged from a tiny 12V DC power supply all the way to RF gear. I became curious to know more about him. His family were great – his daughter and her husband were nice, sweet and patient, they showed me all his toys –
Robert Merville Gale –
I would like to dedicate this project in memory of this wonderful and brilliant soul. Having restored and repaired a great many things in the past, that have been owned and used by others, but I’ve never encountered how beautiful and well maintained this was.
I digress again. So let’s see how I got it back to life.
Important: These instruments contain deadly voltages and exposed leads/components. If you are not familiar with High Voltage circuits, DO NOT attempt to repair/restore or even open this. Even if you are familiar with solid state electronics, do not try this unless you really understand the risks involved with working on Vacuum Tubes and High Voltage DC and know the safety procedures to follow.
It can kill you. Ensure you are skilled enough to do this.
This is how pristine it looked when I picked it up.
But obviously, that’s just the external looks, there are leaky caps on this inside which are waiting to blow up if I ever power it up. So the first thing I did was to CUT THE POWER CORD (to ensure I’m not tempted to ever power it up, that is).
And here you go, this is what it looked like on the inside. The wax/paper caps were all dead, leaking, and waiting to short-circuit if I ever powered it up.
The Magic eye Tube 1629 and Rectifier, the all famous 6X5
The basic circuit of this unit is very simple. It’s an AC bridge, with a differential voltage that is pointing to the grid of the magic eye. In the leak test mode, it’s a simple RC charging circuit with a monitor resistor to reflect the charging current to the grid – this ensures that the eye opens if there is no current, and is closed (fully lit) if there is a constant charging current or a LEAK.
NOTE: DO NOT CONNECT MULTIMETER or any other device IN PARALLEL TO CAPACITOR TO MEASURE VOLTAGE WHILE LEAK TESTING. The current drain by the meter can cause the eye to close (especially paper/mica mode)
ALSO AFTER TESTING, the first step is to rotate VOLTAGE DIAL TO 0, THEN REMOVE CAP. OR ELSE THE CAP WILL BE CHARGED and can surprise you with a nasty shock.
The next step was to do a complete physical inspection, look for broken parts, shorts, missing cables, broken connectors and so on. As expected, I found broken cables, joints, exposed cabling.
Now I needed to formulate a course of action, and this was my plan –
- Clean and inspect (Both inside and the outside)
- Make the solder joints, connections a bit more organized (I might have OCD)
- Verify the schematics – This is not factory made, the person who assembled it may have made mistakes.
- Use heat shrink tubing to cover all exposed joints
- Fix broken joints connections
- Replace all CAPS, no question of reusing ANY
- Check all resistors – replace if required
- Check transformer for short
- Check Tubes – Physical, and Tester
- Power up heater for tubes, ONLY heater and confirm they are good
- Using variance slowly power up to 40V (instead of 120V) and measure all voltages, confirm the voltages at various points to confirm all components are working fine
- Fix issues if any, then slowly ramp up to 120V.
- Test the caps
- Thank Sir Robert Merville Gale for his knowledge and love for all his toys, and for maintaining them so well over all these years.
Here is what all I found,
Some broken joints.
A little dangerous wiring/assembly. Lots of exposed conductors, though I wouldn’t blame anyone as it was common those days to have a 500V line exposed inside the chassis.
So I started fixing it. You can see the pictures and how I’ve worked on it, in stages. I do heat shrink all joints/connections, even to Tube sockets.
I have put in a new power cord, a new y-capacitor and replaced all the electrolytic ones. Have used two in series to keep the capacitance close to the original and the rectifier tube specs call for a maximum filter cap of 4uF. As the tube heats up slowly, it should not cause any trouble even if I have a high value cap, but I wanted to play it safe and keep the surge current as low as possible to protect the rectifier tube. But remember, a very high value cap can load and burn the rectifier tube.
Based on a comment from Michael, I’m adding a bit of explanation here – The 8μF 525V capacitor can be made up using two 16μF or 22μF /350V or 450V in series. Add parallel resistors of 470KΩ/1W across each cap to keep the voltage loading on both capacitors approximately the same. Do not go too high on the resistor as it should have a larger current than the leakage of the caps. Too low value of resistors will load the circuit up. Ensure you have a at least 1mA leakage current through the resistor network.
If you are not adding the resistors, remember the rules for series capacitors. Do add 40% tolerance from its rated voltage. ie, do not replace 2X100V caps for against a 200V Capacitor, as voltages may not be exactly divided across capacitors due to tolerance in capacitance as well as leakage current/resistance.
Another strange value to find today is C8 – 0.25uF/600V. Use any modern film cap, with value of 0.22uF/600V or 1000V. If not available, use 0.33uF/600V or 1000V. You can pick the capacitor with higher voltage value like 1600V if they are available.
The Y cap was taken from an old SMPS – (C7 in the original schematic). The power cord was changed to 3rd pin and grounded earth to chassis.
I disconnected the HV outputs from transformer secondary to keep plates/rectifier from being energized, as well as to not power the filter caps.
After verifying the schematics, the first test was the tube test, with just the heater power. I checked the transformer as well in this process and saw the heater glow from both the tube and the magic eye.
Now, the AC Bridge needed 3 Precision Caps and 4 Precision resistors. Unfortunately I didn’t have the caps to replace it right away, so decided to do the precision caps later as my immediate need was for leak testing, and placed the order for precision caps.
Note: You need not order the exact value as per original spec, as it is rare. You just need to keep the ration (capacitance ratio) the same. You could google it to find more details. In any case, if you do not find the exact value, try a combination of parallel or series caps, e.g. If you need 2uF 1%, get 2 x 1uF 1%. I am ordering 22000pF, 220pF, 2.2uF, 1%. The resistors are good, so I’m not replacing them.
I rechecked the schematic, connections, wiring, and now on to variac on, connected HV secondary to circuit, Slowly raising input voltage via variac to see as the tubes heat up, and watching for magic eye to glow.
You can see the first light up, after so many years, still on 60V power. I used one of its own old caps, to check for a leak and you can see the result for a 400V cap with 88v input 🙂
So it all looks good. While I wait for my precision caps to arrive, I can finish rest of the tasks, replace the last few 0.01uf caps, the main 0.25 (I did 2X0.1uf) and do a final clean up.
And Finally, I am all set :). I am ready to head back to my TEK Project. I will replace the Precision caps once they arrive, but as of now it’s not too urgent as I’m not using this instrument to measure capacitance, but just to test the leak function.
Here is the final picture, all set and ready to help me with my mega TEK project.
Remember : It is not just the caps which needs attention. Always check the resistors too. They can drift and cause errors in leakge measurement or even bridge measurements.
For those of you wanting to attempt something similar, here is the component list –
C1/C9 – Power Supply Filter. Don’t go too high with replacement values as it can load the 6X5 rectifier Tube.
C2-C4 – Precision Caps – used for doing capacitance measurement – get 1% tolerance replacement/ No need for exact value, use exact ratio of values.
C5-C6 -> Coupling
C7 -> Y Cap.
C8 – Not sure, looks like it’s used to reduce the Pot Noise
Pot 1 -P1 – Main Dial – Cap/Resistance Value
Pot 2 – P2 – Power Factor Pot/ On-Off Switch
Pot3 – P3 – Voltage control for Leak Testing
Resistors -> No need of replacement in most cases, still check all values.
If you are using this instrument to test high value electrolytic, eg. 150uf 450V, or similar ones you find in SMPS, ensure you discharge it gracefully after testing, ie, slowly rotate the voltage dial back to zero, else, the Capacitor can burn the voltage control POT by quick discharge, as a charged capacitor can deliver strong current at high voltages on values above 25uF.Or in short there is no discharge resistor in this design.
Here is a quick overview of the schematic.
There are three secondary Windings.
1. for Heater, 12.6v and 6.3v for Magic eye and 6X5 respectively.
2. Second winding is ~ 54V ac, used for the bridge for capacitance/resistance measurement.
3. Third one, this is the High voltage one, which is rectified via 6X5 and doubled/filtered using C1 and C9.
Rest of the circuitry is mostly switches to control the function – ie Leak or R/C Measurement. For Leak testing pot P3 controls the output voltage.
J1/J2 is the test terminal.
J3/J4 is for connecting a known value/reference cap/resistor for comparison.
Important note: In chassis testing of caps
If you check the leakage test schematic, you will notice that the tester sends a -ve test voltage to the capacitor. Or in other words, +ve terminal of the capacitor is close to ground where as -ve is elevated close to test voltage.
Why bother? well, assume you are testing a capacitor mounted on a chassis. Sometimes we can disconnect the only positive terminal of the capacitor from the circuit for testing, as negative of the capacitor typically goes to ground or chassis. Now you attach the tester to test the cap at 300V, even though the positive terminal is disconnected from the circuit, since the -ve is connected to cap and chassis ground, your equipment chassis is sitting at -300V or close to it. So be careful. Either keep the tester off an isolation transformer or do not touch the chassis during testing.
And a snap from the work in progress schematic.