AVO VCM MKII Screen voltage question

[stevefuzz]

Hello

I've just finished the restoration of my VCM MkII and I'm testing it for correct voltages.
Following the instructions on the EMER manual I've checked the anode voltages (in AC) between cathode and anode on the octal socket properly set up and everything look correct.
Then I've checked the screen voltages between the anode of V1 and cathode on the octal socket; everything is good till here.

If I measure the tension between cathode and screen on the valve socket what I have is a steady -64v DC c.a. or 0v AC in all position with "Circuit Selector" on "TEST".

Maybe I didn't fully understood the working principle of this unit, but this sounds strange to me.

Looking for faulty connections I've noticed that the rectifier tube (V1) is wired up in a different manner respect to the schematic.
One of the filament tap, in facts, is connected to the cathode (see picture).
I'm asking to myself if this is correct or maybe someone has put his hands on this unit before me.

Here's the picture

[John_BS]

I'm not familiar with the Mk11, but I've had a vaguely similar issue with a CT160.

The screen voltage should be (+ve) half-wave rectified, but somewhat modified (according to this simplified drawing?) by the presence of a resistor (R35?) across the diode. So, you'd expect net positive but with reduced-amplitude negative half-cycles. On the CT160 I couldn't observe the correct screen waveform until I loaded the screen circuit with c 200k to ground; the problem was merely that with a 10M scope probe, the diode was not loaded sufficiently to show rectification. The Mk II schematic looks correct: Diode heaters to a 6.3V floating winding, cathode to S, Anode to switch matrix.

John

PS Admin: this is my 100th post. Do I get a gold watch or something?

[Dekatron]

John, you are absolutely correct that you need to load these circuits with a resistor to get some current flowing which results in the diode working in its forward direction conducting current. With too high resistance while measuring no, or almost no, current will flow giving rise to erronous measurements. In the later AVO Mk IV, CT160A and the VCM163 AVO loaded all silicon diodes with a 100 kOhm - 1 Watt resistor to the common ground.

At this time I'd just like to point out that there is a big error in the AVO Mk I / Mk II circuit diagram and also in the simplified diagram, the one in the picture you included in the backing-off circuit. There are atleast three serious errors known to me in the AVO Mk I / II diagrams.

Perhaps the most important error is that the diode MR1 is shown the wrong way around, it is the diode in the backing off circuit, the topmost diode in the simplified diagram that you included, it should be connected the other way around. In the simplified diagram the round symbol with the negative half-wave rectified signal should in fact show a positive half-wave rectified signal. The backing-off circuit will not work with the diode MR1 shown the way it is in the diagrams and neither with a negative half-wave rectified voltage.

The backing-off circuit works by putting a variable (positive) half-wave rectified voltage on one side of the meter, this voltage being in-phase with the measured voltage present on the other side of the meter - when these two voltages are equal the meter reads zero as there is no current flowing through the meter.

The explanation for this is that the measured voltage is a positive going half-wave rectified voltage, the valve under test acting as the rectifier in this circuit. If you would put a positive going half-wave rectified signal on one side of a meter and a negative going half-wave rectified voltage on the other side of that meter you would never get the meter to reach its zero position except when these two voltages both are zero, so that rules out the negative voltage in the diagram.

My friend Euan MacKenzie is writing a restoration document where all AVO Mk I / II errors will be discussed and a corrected circuit diagram will also be included.

[stevefuzz]

Thank you very much for your help.
I'm aware of that error on the schematic about MR1, but on my unit that diode is wired up correctly.
However it's seems that something go wrong with the screen voltage.
I'll give you some note about my unit strange behavior, I hope you can suggest me where to trace down the fault.

With a good tube under test (6V6) I set up the unit in this way:

Anode 250V
Screen 250V
Grid - 12V
Circuit Selector on "C"

I go through the "COLD" ELECTRODE LEAKAGE TEST and everything is ok, but when I switch on "S" (screen), the needle fall beyond zero.

With the Circuit selector on "TEST", there's no current flowing on the meter or in the external milliammeter connected to the LINK.

I've double checked every connections and everything looks correct.
I begin to be seriously puzzled about this fault.

Stefano

[Dekatron]

Stefano, have you checked the Cut-Out relay and the Reset button so they work correctly as that circuit switches off the Anode & Screen voltage to the Valve Panel on the top?

[stevefuzz]

Thank You Martin

The Cut Out circuit is the first thing I've checked and everything is ok.
In facts when I measure the voltage between cathode and anode, I have the expected results.

I've made another experiment.
I put a WW 2K2 variable resistor between anode and cathode at maximum resistance, then I set up the controls in this way:

Anode 20V
Screen 0
Circuit Selector on TEST

I begin lowering down the WW pot, but the needle on the VCM doesn't move.
The result is the same even on different "Meter Selector" position.
I have to assume that there's no current flowing, but I can't figure why.

I have rechecked the connections for the tenth time and everything seems correct.

[John_BS]

Do you have an oscilloscope?

John

[stevefuzz]

Yes!

[Dekatron]

Stefano,

First of all you will have to do all tests with a silicon diode in series with the resistor as the valve works as a rectifier during the tests, otherwise you will have an AC signal present on one side of the meter and a DC signal on the other side and the meter will not work then. You can read a little about how the tests are perfomed in the AVO Mk III or AVO Mk IV Service Manual where they use a silicon diode in series with a resistor to simulate a valve.

If that does not change your measurements I would check two things with a DMM on the resistance range (with the meter in the AVO Mk II shorted with a wire so you do not accidentally destroy it while measuring resistance!) first set the Circuit Selector in the TEST position and then test:

*. Connect one lead of the DMM to switch SGab in the EMER diagram I sent you. Then test if you have connecteion to Anode-1 or Anode-2.

If you do not have any connection at all above there is an error in the Anode Circuit you just tested that you will have to find, otherwise continue below:

*. Leave one lead on SGab. Now check if you have connection to both sides of the meter in the AVO Mk II, this test will have to be with one side of the meter removed, you can't do this with the meter shorted out as before and not with the meter in the circuit as you will get measurements that won't let you see if any of the sides has a broken connection.

If any of the sides do not give you aresistance value then you need to check that side.

When you have checked the above I can continue to try to guide you to the error.

[stevefuzz]

Martin,

thank you very much for the EMER service manual, now I can go deeper in the working principle of this unit.

I've just finish to perform the test you suggested me.
I've put a silicon diode in series with the resistor (the cathode of the diode connected to cathode of the valve holder) and the current began to flow.

When I tried with a known good tube I got no results.

[Dekatron]

Stefano,

That is strange, what valve are you using for your tests? have you tried a rectifier/diode valve like an EAA91 or similar to see if the Diode / Rectifier tests works? Do not start with a high current rectifier as that might burn things up if there is an error in the wiring!

Could it be that the heater voltage is not getting to the valve in full power? Try to measure the heater voltage directly on the pins of the valve when you test it. Do not test the heater voltage in the contacts near the roller switch but directly on the valve pins, that will show if the voltage is correct or not.

If you do not get the correct voltage there and it is too low you should check the heater voltage circuit if it works correctly or if there is some contact that is bad - do this without a valve in the tester or you might destroy the heater.

Then check the heater voltage in the H+ & H- contacts and see if it gets lower when you insert the valve, that will show if the valve is drawing some current. if you can you can try to measure the valve heater current by disconnecting the H+ wire on the valve panel contact and putting an ampere meter in series, that way you can see if the valve draws the specified current.

If the heater is to low the valve will not conduct as much as it should, and possibly not at all.

If all above is correct I should check the circuit around switches SHbb & SHbc and R2 to see if they are connected correctly. I'd also check the big roller switch on top to see if I have correct contact to the valve pins and when testing the valve I would leave SJ in the SET AC position so it is not connected in the cirucit possibly shorting any of the valve pins if it is broken.

I'll try to think of anything more to check while you test these things and get back with the answers.

[stevefuzz]

Martin,

I think you're right suggesting me to check the filaments voltages.
While the coarse switch give results within specifications, the fine switch appears too low even without tube under test.

I've tried two different tube with two different filaments current drawing (a 6V6 450mA at 6,3V and a 5U4GB that draws 3A at 5V); when the tube are inserted the result is the same: the measured filament voltage is 0V at the pins!!

Today I'm gonna check for bad contacts or cold solder joints in the filament switches.

[Dekatron]

Stefano,

Yes, checking the heater voltage is just what I intended you to do.

I think that you are onto something now, possibly a bad contact or a switch with high contact resistance which means that the voltage drops when the load is put on the transformer.

Now you have to check the roller switch on the valve panel, the switches SIfa & SIfb and also SC. I'd start by measuring the voltage at the SC switch to chekc if it si ok there. You can always measure the resistance of the heater of a valve, say the 6V6, and then use a high power resistor to connect it across the wires on SC to simulate a load and then see what happens. Don't forget that this resistor can become hot so you do not burn any wires inside the tester, so use long wires and put it outside and isolate it as you can get 117V across it! Preferrably use a 50W or 100W resistor so it doesn't burn up if you accidentally switch to 117V heater.

I hope that you find some easy problem and fix it quickly!

[stevefuzz]

Hello Martin

I've made some tests that helped me to restrict the field for fault finding.

First of all, as you suggested me, after having completely cleaned the sc switch, I've checked the filaments voltage directly on it.
They resulted all below the specs, referring to the EMER manual that you sent me.

Following the calibration procedure, with anode voltage on 200V, when you set the unit for AC till you have 220V AC between anode and cathode (about half way on the SET switch), the heaters voltages are far away beyond the suggested results.

I've tested a brand new rectifer tube (5U4GB), though heaters were at 4v instead of 5V, and the unit worked perfectly. The needle moved and stopped at about half of the GREEN-GOOD scale.

Then I tested a double triode tube (12AX7), here too the unit worked perfectly on the emission side at various meter scales.
The GM test failed; when I pushed the button (carefully and slowly) the needle swinged at full scale.
The only electronic parts that I radically changed are MR4 and MR3.
I put two 1N4001 silicon diodes instead of the originals that were shorted.

The VCM doesn't work with pentodes.
I mean that there is no current flowing.

I have to suppose that the fault is in the SCREEN supply.
My guess is that the faulty supply for the screen pull down other voltages, just like as the transformers work under load even with nothing connected to them.

Having carefully checked the connections for the screen on SI and SH I'm beginning to think that V1 is failing.
Unfortunately I have no spare for this at the moment.

[Dekatron]

Stefano,

If you have a pentode that it does not matter if it is damaged when you test it, you can easily use a silicon diode instead of the EA50 valve, just connect the diode the correct way when you test the pentode and everything should be fine.

[bud allen]

Best greetings to everyone.

I am rarely stumped, but I am now.

Just rebuilt the AVO MkII, however the mains meter advanced only one division when circuit and leakage are on the proper settings for the tester to advance to the black calibration mark. Changing the leakage to (H) and the needle returns to zero.

Reset is set. All diodes (1n4007) are in proper polarity as per the original ones ( backing off diode positive on pot side). Continuity from the meter to R4 and beyond is right all the way to junction r10 and r9 and diode. All bobbins are good or replaced, same with resistors. Pots are also good.

The valve for the screen only puts out about 3vdc, and I guess this is the problem, but Anode ac follows dial but dc is also about 3vdc. I think I know how to simulate it with a 1N4007 and resistor, however a comment is welcome.

Once, after setting for a few minutes and plugging in, the meter jumped up and then fell again. Solder joints are refreshed.

When I follow the calibration by Hein Ros, the first instruction to position the circuit selector to TEST, pins the meter into the negative. There is no mention if this is expected.

If any of you have proper schematic or suggestions, I would appreciate.

bud

[bud allen]

Success!
Right off if you don't have a clear and correct copy of the schematic, you have a problem. That was my biggest problem right off, that and AVO documentation is very poor, has many mistakes, and uses 50 words when 3 would do nicely.

Anyone having issues with any AVO tube/valve tester and has a good schematic should suspect poor solder joints as a possibility. Having found 13 cold and 1 that completely slipped off the terminal and left a hole in the solder behind, I now have voltages. One which messed up the bias, line mark, and V/ma was......the 'backing off' pot directly behind the front cover had one wire off the extreme terminal clockwise terminal, another poor solder joint, and that had an affect on the other circuits. To add to that the pot itself had no contact with the resistance on the last 2 or 3 degrees of rotation. Another pot had the last inch of the element flattened straight in order to stretch and make the tab connection, instead of hugging the roundness of the shell as expected ( in other words, the pot was resistance was no longer linear for the last part of the arc because the element was flat).

How on earth AVO let such out the door is a mystery

My advice to all is: Clean every pot and contact, and check all solder joints before you begin to trouble shoot the rats nest of wires and components. Assume nothing is good. Expect to readjust all voltages which have new diodes. The feedback from one component to the next means the component must be unsoldered to get a proper reading.

Many thanks to Gary.