Hickok 539A tube tester

Having 3 tube radios from 1950 .. 1960ies, I though it’s time to also have a tube tester.

You can have the simple ones, but also a lab level device. All these tube testers come from 1950 .. 1960 years. You can have one for european tubes only or more general ones with american and european tubes. So I decided to get an american version. All modern amplifier tubes are american brand, so this is a good approach.

Tube testers are expensive items. Finally I found a Hickok device -The #1 vendor of tube testers back in the days.

My Hickok 539A is a complex and beautiful device.

Hickok 539A overall view

While all/most german tube testers use pile of paper cards to set up the tester for a special tube type, american tube testers usually use rotary switches to individually set up the device a tube to test. In the picture above the selector row contains 7 switches for setting up the device for a particular tube.

Filament voltage can be set up in a wide range between 0.6V and 117V. An expected Micromhos range (in german: Steigung Ia/V) can be set between 3.000 and 30.000.

Filament switch, Micromhos switch, shunt regulator

Gate BIAS can be set with a voltage regulator in range 0..10V or 0..50V. Shorts inside the tube also can be identified (Shorts switch).

Bias regulator

Before using the device for any measurement, the device can be somehow „calibrated“ to AC input (110V US).

Calibration to AC supply voltage

After calibration, the tube can be inserted in one of 11 tube connectors.

11 standard tube connectors

The following image gives the name of the sockets available.

Socket names were collected from Internet sources. Some frequent sockets are missing (e.g. Rimlock and Magnoval). They need to be added via Adapters.

Inside of the device, the following pictures were taken.

Inside the device. 2 transformers at the top left, tube data „display“ at the bottom, and many other switching and regulation devices….
A whole bunch of multi-layer switches
Power supply tubes
The date code, February 1952

For us Europeans: 1000 Micromhos is 1mA/V …

Testing unknown tubes

Knows tubes can be selected by entering a code like ff-abcd-e, example EV-2781-9 with the rotary dials in the middle of the device labeled „Selectors“. These dials allow to connect the tester inputs to the tube socket pins. From the left, the dials are for Filament 1 (f), Filament 2 (f), Grid gate (g1), Plate (a), Screen gate (g2), Cathode (k) and Suppression gate (g3).

For known tubes, there are lists containing the code ready to enter. For example, for EF80 the code is EV-2781-9. For the first triode inside an E80CC, code is EV-7608-0, for the second triode EV-2103-0.

For unknown tubes, testing is still possible. The code to enter then is not contained in any list but must be developed by checking tubes data sheet.

For example, the EF98 has a 7 pin miniature socket. Filament voltage is 6.3 volts. Pin numbering is:

1=g1, 2=cathode, 3=f, 4=f, 5=anode/plate, 6=g2, 7=g3.

EF98, seen from bottom side

From pin numbering we have (in order of code letters):

34-1562-7

Socket numbering for the sockets used in 539A:

Hickok 539A socket numbering

Historically, Hickok has introduced two ideas that complicate things.

Pin numbering in Hickok 539A differs from convention for some sockets. For e.g. 7 pin miniature, the numbers are reordered for pins 1,3,4 to 3,1,8. Don’t ask why. So, the adjusted code is:

18-3562-7

For the filament pin part („18“), numbers are not used but a combination of characters. Check the following table:

Socket pin number Filament 1Filament 2
0AP
1BR
2CS
3DT
4EU
5FV
6GW
7HX
8JY
9KZ
Mapping of filament pin numbers to code characters
Scan of mapping table

So for pins 1,8 we get B,Y. This is the same as J,R. So, the adjusted code is:

JR-3562-7 (I guess BY-3562-7 is also correct)

Required BIAS voltage (gate voltage) can be taken from data sheet. Plate voltage is fixed, on my device about 130 volts. Of course, there is no Micromhos target value because there is no list entry in Hickok lists. But this value also can be taken as a landmark from the data sheet.

This example with EF98 is already some special case, because the EF98 is a battery tube, designed for low plate voltage. It’s plate voltage maximum from datasheet is 50 volts. Because the Hickok 539A delivers about 130 volts if adjusted to 100 volts line, this is already too much. But it is possible to adjust line voltage to lower values than 100 volts and then the tube can be tested – at least if the test runs only for a short time.

This allows to do some basic tests also for unknown tube types. If you have a new and good tube of the same kind, even comparisons of the tube to test can be done.

Radio Loewe Opta Komet 53

I bought this radio as defect, does not play for some euros. It had no power plug anymore, and my initial idea was to tear it apart to salvage the ancient parts used in the 50ies.

Radio after minor repair work

The wooden case was in quite good condition and all switches and knobs were present. So I decided to check for the issues in the radio and maybe repair it.

Look inside the case. At top left, we can see the socket of the magic eye, an EM4. Top right is the speaker, with some fabric around it. At the bottom left, we see the HF input tube, an EF80, with antenna connector in. 2nd tube from the left is ECH81, then 2 times EF41 and the tube at the right is an power pentode EL41 to deliver the 2.5 watts or so. The arrangement right to the EL41 is the power supply transformator. The big knob in the middle is a connector for a remote control (wired of course). The big wheel is driving the variable cap to select radio stations.

For first test, I hooked up the radio to my isolating adjustable transformer where I can start with 0 volts up to 250 volts for power supply of the radio. If there would be a short or so, I could stop at any time before some serious would happen…

Shortwave fine tuning

All tubes were glowing. So their filament at least works.

Chassis pulled out.

This radio has a turntable input, so I used my mobile phone output to check that the NF amplifier works. It did, and I could hear the music quite loud from the tube amplifier part.

When removing the bottom carton, I found there the schematics. What a nice idea from the old engineers…

Komet 53 schematics. It already contains some kind of modernity compared to pre-war receivers, because it uses a semiconductor rectifier – a SSF B 250/85

The schematics told me that the EL41 is working. So far I had no radio sound and no magic eye. It stays 100% black.

I decided to buy an HF generator for further testing. The device, a WF PG1, is described here. It arrived some days later and need also to be repaired 🙂 . After repair, I could continue with the Komet 53…

When creating a 90 Mhz FM modulated signal at antenna, I checked the anode of EF80 with an oscilloscope. Some signal was there. I repeated that with ECH81 (ok) and first EF41 – no signal there. When switching the two EF41, the signal was there. So I derived from that that the first EF41 was broken. I ordered that tube as NOS from a german electronics seller for about 6 euros, plugged it in, and: sound was there!

The sound was not perfect, I guess the radio was working the first time for 30 or 40 years. But I was proud anyway 🙂

HF/IF section. The tube EF41 at the right was found to be broken

Next thing was to check the magic eye. It does emit nearly no light. This is a very early version, the EM4. Magic eyes do age, and loose their brightness with time.

Magic eye

How a magic eye works is described here. The EM4 is well described here. And there is also described how an aged magic eye still can be used.

So I plugged out the magic eye to check if the tube is working at all.

The first image below shows what can be seen if the tube is attached to the radio set and radio gets 200 volts AC.

A very little brightness can be seen. Only if room is dark.

Next I have connected the tube to my NEVA power supply to test it.

The next image shows the tube when connected to 225 volts and 2 resistors with 1 Megaohms are used.

Tube a t 225 volts plate voltage

And the next image shows the tube at 364 volts plate voltage.

Tube at 364 volts plate voltage.

So the tube is not broken, but has aged.

Here, the filament glowing can be seen

Next I tried to vary the gate voltage.

High signal, lower quarter is 100% and upper quarter is already lets say 20% filled. in this and following image, tube connectors 1-4 are on top
Low signal, lower and upper quarter are not 100% filled

This all looks ok, so I think the tube is a) undervoltaged and b) does not get proper gate voltage on the radio.