AVO CT160 set mA/V issues and calibration

[waran2005]

Hi all

I have recently purchased an AVO CT160 and I'm having problems with the set mA/V control. When I move it to the set zero position and adjust the fine anode current the needle never really settles at zero, it is either falling or rising - is this normal for this tester? So far, I have removed and cleaned out the mA/V potentiometer, cleaned the switch contacts but this has made little difference. Another issue with the mA/V pot is that I cannot calibrate as per the manual, it is not within specification when tested. I can get the 0 to 8 range to calibration perfectly, but the 8-20 is way out. I've checked the mA/V potentiometer, and it is was high at 2,78K, so I've bridged it with a resistor and brought the value back down to 2.5K and it is still not correct. I have also checked the negative grid voltage with the link open and confirmed that it is 20.8v with the mains set at 235v (and confirmed with a meter the mains voltage is 235v) and that the set AC is within the calibration segment on the meter. I have also checked R3 & R4 and they are within specification. Is it likely that the issue is with the fine anode current control used to balance the meter, maybe people have had experience with this in the past? To be honest I'm tempted to replace both anode and mA/V potentiometers, are there any recommendations for a suitable replacement?

Thus far this has been an interesting project, cosmetically it is pretty good condition, but it has had numerous faults, luckily the meter is working.

Here is a list of some of the problems I've had:

Negative grid pot did not work.
Meter housing was virtually broken in 2.
Meter front and rear glass were on the point of falling out.
Compensation coil rattling around inside meter - nut missing!
Anode rotary selector switch bottom support was broken.
Sticky operation of the electrode selector wheels - stripped and cleaned.
Various nuts and bolts loose.

So far, I have repaired the neg grid pot, plus I've made up and fitted a replacement as I don't trust the original - I think it is probably worn out and will fail again. Repaired and tested the meter for FSD at 30 uA, I'm lucky it was working considering the damage to it. Fitted meter protection diodes/capacitor as recommended. Made a new support bracket for anode current rotary switch. Checked all resistance values, replaced capacitors and tightened any loose nuts and bolts and I'm currently working through the calibration procedure - hence mA/V issues.

The negative grid potentiometer did not show any signs of stress/heat damage, the resistance wire was broken where the riveted connection washer is fitted. I drilled out the rivet and fitted a fractionally larger washer to remake the connection. I think the damage is a result of wear and tear, I don't think the original neg grid pot would have lasted long and so I've made a new one up from a NOS 25K potentiometer which works really well and keeps the original look of the front panel with the log dial etc.

I think this is a good tester, it could be great when fully calibrated and working as it should. I would think twice though before buying another one, there was a lot wrong with it (still something wrong) and parts are non-existent. The biggest problem is that it's clearly had a lot of use and has had quite a hard life - as expected with use in the military I suppose.

Another issue, mains voltage fluctuations significantly affect its operation, I have connected a meter to the mains while in operation and I've seen the mains voltage fall from 241 volts down to 230 volts over a few minutes, as such it requires constant voltage adjustment to keep it within the set AC range on the meter. I have bought a servo variac/auto regulator, but it is not that precise at controlling the voltage setpoint and I will need to see if it can be improved.

Sorry to have gone on and on, just my experience with this antique tester. As expected with AVO, build quality is excellent.

Any help appreciated.
Waran

[Station X]

I would just point out that the mA/V pot in these testers is a special with three different sections of track. Look at the non-linear scale and you'll see what I mean. Connecting a resistor in parallel with it to achieve the correct value may not be a good idea.

[waran2005]

Thanks for the reply

Is the mA/V also a log potentiometer, I thought it was only the negative grid volts potentiometer that's a 3 section log type pot. It the mA/V one is also a log type, then that may explain some of the mA/V issues I'm having.

[Dekatron]

You are correct waran2005, the grid volts potentiometer has three linear sections to make up for the three grid volts areas on that dial.

The mA/V (gm-wheel) potentiometer is a normal 2.5k linear potentiometer and has to be for the dial of the mA/V wheel to work properly - the circuit around the mA/V potentiometer is what makes the 0.52 calibration factor work as that is how the resistors around the mA/V potentiometer makes sure that the dial has a factor of 0.52. I wrote about it here: https://www.vintage-radio.net/forum/...ad.php?t=86262

You can't modify the 2.5k mA/V potentiometer in any way if you want it to be correct against the dial on the gm-wheel - unless you modify all of the components in the circuit to compensate the circuit to the dial. So try to find a new potentiometer on eBay, like the Clarostat 2.5k 4W types from the RA30 series, or from any place where you can buy one.

If you have access to the service manual you can read that AVO recommends you to split the error for the gm-wheel by either adjusting the potentiometer or the dial to get the best possible tolerance on all three points. If you know how big the error is you can then calculate the difference to the dial but usually it is small enough that it really doesn't matter.

[Station X]

Sorry I screwed up there. Martin is correct.

[waran2005]

Hi all

No problem Graham, what you were saying made sense and would have also explained what was going on - thanks anyway.

Thanks for your input Martin, not sure why had not seen those documents that you've created before, they are fascinating and I am going to print them out and study them in detail.

I have a conductive plastic Honeywell potentiometer that looks identical to the one you recommend, so will replace that later today. I also have a 100 Ohm Cermet potentiometer ordered which I am going to use to replace the anode potentiometer, hopefully that will be okay when bridged with a 1K resistor - I think others have done this before. Getting the roll pin out was a real pain, I tried a G clamp without success, I then tried Mole grips with a nut over one side and that did the trick. I just don't like the anode pot; it’s probably worn out plus it has poor resolution, maybe due to the thickness of the resistance wire.

Set up my new Negative grid volts pot in situ, but have now removed it to seal it up, when tested it reads exactly 10K across it, that surprised me.

Had to make a jig with holes to be able to precisely drill them in the side of the new potentiometer, see photos. If anyone else is interested I can post more details.

[Dekatron]

Seems a lot of people miss the AVO article I wrote, maybe I should change the title to "The magical AVO 0.52 factor demystified!" or something like that to spread it more and get people to read it?

Now, I haven't checked the 100 Ohm Anode potentiometer much but before you do anything you might check to see if a new wiper can be fitted, usually you can remove the wiper from a similar potentiometer and remove just the old piece touching the wire and then solder the new wiper in place. Just snip off enough of the metal arm that carries the new wiper so that you can use it to solder it to the old arm.

Adding a 100 Ohm resistor across the outer connections might change the scale of the Anode potentiometer - but I haven't really thought a lot about this nor tested it.

[waran2005]

Hi all

Thanks for your advice Martin, I will leave the Ia potentiometer as it is for the time being as I now have a new problem. I have printed out and read all the information you have provided and it is fascinating, I now have a much better understanding of how it all works thanks to your hard work.

The new problem I have that maybe you can help with, or anyone that may have had a similar issue with a CT 160 is that when carrying out the negative grid calibration, I set the voltage to 20.8 as specified, but after closing the link I'm not getting 47 volts across RV2, only 46 volts? This lower voltage means that the meter needle never quite reaches the calibration zone.

I have checked all the resistors and all are within tolerance, checked through the wiring, checked the earth connections, checked the 66 and 55 volt winding on the transformer, checked the mA/V calibration and it is still reading high, as it always has. The only thing that I've found is that RV3 has suffered some sort of heat damage at one end of the wire wound track, as a result the resistance of the potentiometer was much higher than 500 Ohms. Even after correcting this issue the voltage is still low. If I increase voltage across RV2 to 47 volts by increasing the mains voltage with my Variac, the meter needle falls nicely within the calibration zone.

The diode valves are new, and they are the only things I've not replaced since fitting them several weeks ago, I have swapped them over but it has made no difference. The only other thing I have noticed is that R1 gets too hot to touch - is this normal for a CT 160?

Something is causing this voltage drop, and I'm now out of ideas and so any help would be greatly appreciated. One other thing, I maybe clutching at straws, but how important is it to use an Avometer for calibration of the valve tester? I'm currently using a vintage mains powered digital Fluke meter, I can get hold of an Avometer if required, plus I have an old oscilloscope and a VTVM.

Waran

[Dekatron]

What mains voltage do you have in the wall, and what setting have you used for the mains voltage settings on the CT160? Can you use a lower setting on the CT160, which will raise the secondary voltages a small amount but still leave them within their tolerances according to the service manual, and get high enough voltage across RV2?

Some multimeters don't measure these half wave rectified sinusoidal voltages well, especially when there are two different sized voltages.

The peak power through R1 is close to 2.7W and the mean power is close to 1.1W, I don't know how hot that makes the resistor though.

If you unplug your CT160 from the mains, set it to SET AC and then unplug V2, the valve supplying the grid-volts potentiometer, and then measure the resistance across the 10 kOhm grid-volts potentiometer as it is in the circuit, what resistance do you get then? It should be close to 1938 Ohm if your components have the correct value and all switches are working properly.

[waran2005]

Hi Martin

I have approximately 235v at the socket, it does drift around a lot but spends most of its time at that voltage. I think you are right about the old Fluke not giving me an accurate average reading when calibrating, Avo specifies an Avo Electronic Test Meter or a DC mean equivalent VTVM - I'm thinking they had a reason for using this high impedance meter, and maybe to achieve the same voltage specified then I need to use something similar. I have not tried my PACO VTVM, but I have connected RV2 with the link open to my old scope and set it to DC/MEAN. Then when RV3 is adjusted to achieve 20.8 volts on my scope, with the link closed I now have the 47 volts and the meter needle sits perfectly within the set AC/mains zone. I don't know how accurate my scope is, I only paid £10 for it at a local radio fair, so it is probably questionable at best.

I have also done what you have asked, and total resistance in my tester across RV2 measures 1929 Ohms – double checked with several meters and all readings are within 1 Ohm - Is this an acceptable value for the total resistance, even though it is 9 Ohms lower than you specified? This is with my new DIY negative grid pot.

I have tested some new valves with it, and it appears to work well, giving correct values from the valve data manual for the few new valves that I have, and so I’m happy, I imagine it is working well enough for most people but it has turned into an obsession - I really want to make sure it is perfect!

I'm going to make a DC rig and standardise a valve to confirm it is as it should be. I also have a box of valves that I bought a while back, they have all been tested on another Avo valve tester, and I have all the results/data which were supplied with then - so I can also use those to roughly check it.

I fitted a new Ia potentiometer for the time being and it works very well, resolution is better and it is so much easier to zero the meter before the GM test, and as a bonus the current measurement is much more accurate - beforehand it was consistently 5 to 10 milliamps low!

Still not happy about R1, it gets very hot, too hot to touch - why have they used a 3/4-watt resistor when the nominal is 1.1 Watts? R3 also gets warm, but not hot. Whilst carrying out a GM test, they must be overloaded even if only for a brief period, maybe a higher wattage RV3 and R1 will improve reliability/accuracy considering RV3 had gone into meltdown – what do you think?

Thanks for all your help
Waran

[Dekatron]

The resistance you measured is ok, well below 1% off.

What value did you use for the Ia potentiometer? Is the scale for the Ia potentiometer still "calibrated" with the new one in place?

Well, if I remember correctly R1 should be a Welwyn C23 resistor rated at 500V working voltage and 0.75W at 70 degrees centigrade, maybe it has a somewhat higher rating at lower temperatures. I've never felt one that have gotten uncomfortably hot though. I know from speaking to a gentleman who worked at the factories when these and other types used in the AVO series of valve testers where manufactured, that they weren't always as good as being specified. You can replace it with modern types if you want.

[waran2005]

Hi Martin

I have replaced the Ia pot with a 2W conductive plastic one purchased from RS, it appears to work well - much better than the original. I will clean up and try re-fitting the original Ia pot in the future, just out of curiosity and for originality. I have checked the mA setting with an Avo 8 in circuit, and it is pretty much spot on from 0 to 10, with a few extra past the ten for the range selector overlap.

I have put it all back together now happy in the knowledge that it may not be 100% perfect, but it is fundamentally OK and as far as I can tell it's accurate and more than adequate for what I need it for. The new negative grid volts potentiometer works well, I have enough parts to make a few spares when I have time. I will revisit this unusual tester at some point, I plan on replacing RV3 and R1 in the future.

Thanks for all your help and for the information explaining how this tester works.

Waran