Avo Valve Tester

[Sparks]

I just got the AVO MK1 valve tester out again, (1937 vintage, purchased on car boot sale), and found to my surprise, that it appears to be working normally now! When I first got it several weeks ago, it hardly worked at all. After cleaning the switch contacts, it was a lot better, but still way out, showing unused valves as poor/low. Now, it is showing unused valves as up in the 90s! I can only assume that it has been lying around for years in the freezing cold, and the last few weeks inside a warm house has rejuvenated it. I recall that at sea, we had a fixed lifeboat radio station in a small cabin in one of the lifeboats. This consisted of a set of 24 Volt batteries, and an MF/HF transmitter and receiver (valves). The small cabin had a low power heater in it that had to be plugged into a socket aboard the ship to keep the equipment warm. Sometimes, someone would pull the heater plug out, and when the radio equipment was checked, it would not work at all, having gone cold. After plugging the heater in again, it would start working after about half an hour. This was the case even in the tropics when it was normally boiling hot anyway. Maybe something to do with condensation more than heat!. Anyway, my ancient AVO Mk 1 is now OK, and will be kept warm and dry from now on. I find that most of my EF80s (ex TV sets), are round the 50% mark, but my larger EF39s, are all between about 60 and 90% with the odd one at 100%. Very pleased about this!
Bob

[Mike Watterson]

The meter can get "sticky".

[Sparks]

The meter wasn't sticky though, and worked quite smoothly, but the readings didn't make sense.     What still puzzles me is reading the scale (See below).     The Replace and Good sections with the short bit in between seem to indicate that the higher the reading, the better the valve is, i.e 0% to 100%.  (I often see valves advertised as "tests at .....%"  But it says mA/V above it.     Also, if I test a valve that has a gm of 3 mA/V, should the needle only go up to that figure, or go right up into the green section that starts at 5.6.    If someone could explain it to me in clear language, I would be very grateful, but I have never had any experience of valve testers before!       To put it even more simply, I don't know what I am reading on the scale, the goodness of the valve  in percentage, or the actual mA/V.
Bob

[Mike Watterson]

If it was mA/V, then good or bad might depend on the valve.
Isn't there a switch and an adjustment knob for mA/V mode?

mA/V needs a change in input to measure gain.

Replace / Good is usually just actual absolute current, emission. Again though without settings per valve type, it doesn't mean anything.

So in one state it's a very primitive go/fail, but only probably for particular valves as low current is bad on an 6V6 or EL34 or PX4, but fine on an HL1 or EF39?
The other state is ma/V and you need to check the valve data and also change the input to read it.
The needle isn't at zero in the photo.

Tests at say 70% means that the Ia is 70% of what the quoted figure is for that valve at makers setting of Vg1, Vg2 & Va. I think for many types 66% might still count as "OK". The reading can be 120%, i.e. 1.2 times more that quoted in data sheet. Unless you know that the correct Vg1, Vg2, Va and Vh was uses, the % test figure is meanigless and it's only one parameter, though an indication of wear. A valve might still be usable even at 30%.

[Sparks]

To use the meter, you turn the mA/V control to the gm from the valve data book. The Set Zero control on the other side is then turned until the needle goes onto zero. To test the valve, the central "joystick" is pushed to the left, and the needle goes up to indicate the condition of the valve. Push it the other way and it indicates the cathode insulation.
So why do they have mA/V write on the scale?
When I got it, the needle was off zero, as you state, and could not be corrected by the manual control on the front. I took the movement out and found the needle was bent. I straightened it out (in fear and trembling) and put it back. It now rests exactly on zero!
Bob

[Mike Watterson]

Note that mA/V of a triode or pentode does vary with operating voltages. You need a minimum Anode voltage, correct heater (10% can make a big change) and be well away from Vg1 cut-off voltage.

Valves with variable spacing on the winding of grid are "remote cut off" or variable mu, the mA/V changes far more with Vg. E.g, EF80 and EF86 are sharp cut off (more linear) and EF89 is variable. The less linear variable spacing on winding grids is better for AGC.

The Vg2 (screen volts) dramatically changes mA/V. Hence Vg2 must be at correct setting. The DF92 or EF80 work better with amplified AGC on G2, the DF91 and EF89 work well with control grid AGC.

[Sparks]

I am just following the instructions from the book. It tells where to put the anode & screen switches for the correct voltages for the valve to be tested, it tells where to put the mA/V control. There is nothing else to do. Why do they put mA/V across the top of the scale? I still don't know. I really need someone who is familiar with this type of tester to explain it simply!
Bob

[pwdrive]

The meter wasn't sticky though, and worked quite smoothly, but the readings didn't make sense.     What still puzzles me is reading the scale (See below).     The Replace and Good sections with the short bit in between seem to indicate that the higher the reading, the better the valve is, i.e 0% to 100%.  (I often see valves advertised as "tests at .....%"  But it says mA/V above it.     Also, if I test a valve that has a gm of 3 mA/V, should the needle only go up to that figure, or go right up into the green section that starts at 5.6.    If someone could explain it to me in clear language, I would be very grateful, but I have never had any experience of valve testers before!       To put it even more simply, I don't know what I am reading on the scale, the goodness of the valve  in percentage, or the actual mA/V.
Bob

For a mutual conductance reading in mA/V consult the Avo valve data for the valve you want to test and set the anode/screen/heater switches etc accordingly, set the mA/V control to 100, set the anode selector to normal then plug the valve in, once  the valve has heated up back off the meter current to zero then set the mA/V switch to mA/V, press the key switch to mA/V and read off the top scale (0 to 10) that's the valves mutual conductance figure in mA/V at that moment in time.

The Replace/Good test does not give a mutual conductance read out, to do this test follow the same procedure given above so far as and including backing off the meter current to zero then set the mA/V control to the mutual conductance figure given in the Avo valve data for the valve you want to test, re adjust the backing off control to ensure the meter current still reads zero, press the key switch to mA/V  and make a judgement from the Replace/Good scale.

Hope this helps.

Lawrence.

[pwdrive]

The above was based on this manual:

http://www.jacmusic.com/Tube-testers/AVO...VO-Mk1.pdf

Lawrence.

[Mike Watterson]

I thought it had to be some adjustment like that!