This is a about a restoration project of an AVO CT160. 

Here is how to use the AVO CT160

I am sorry, it is sold, but you still may find the this helpfull for your own projects.

Just bought another one in very nice condition. Little use on it, not worked on by amateurs, a very clean deck, and perfect meter. It needs a repairs on the grid voltage potentiometer, and the transconductance clockwork, which is sort of normal with CT160 after 40 years. For the rest there is nothing wrong with it. After these repairs it needs full calibration, and then it's ready for a long use period again. I need to do this in between normal work. I expected it to be ready for sales by Oct/Dec 2007. It will cost 1600 Euro, add 19% tax when sold in Europe. It will be 100% checked + full calibration done + guaranteed to have no defect or problem whatsoever.

With it comes a copy of the factory calibration instructions,  and a copy of the advanced 
working manual for engineers.  This explains in great detail how this tester
works, with all circuit diagrams.   Both are very hard to find!! 

Additional is a copy of the CT160 manual, in a binder. This
manual shows a quick test guide line, for beginners, and of course setting data for
all tubes.   So you can start and run a test right away, without having to read much. 


This is one of the finest testers ever made.  It competes with the Mk4 in terms of quality and the period when it was made. I agree the Mk4 looks nicer,  but I think the CT160 is better.  It works by some principles making it more robust than a Mk4, and actually gives a more precise reading at higher current.  Sorry for the nice looks of the Mk4,  but precision comes first with me. So I prefer the CT160.

For instance when you test with a Mk4 on the 100mA scale, you have to estimate the value's  second digit.  Also with the Mk4 you depend too much on the calibration ( is it still good...?) .  With the CT160 you only balance the bridge, and the precision comes from the build in bridge. 

So with the Mk4, when you read 53mA  the 5 is sure, the 3 you must estimate from the scale.  It might as well be 52 or 54mA depending on how much you trust what you see.  More precision is not possible. 

With the CT160 you can READ the first and second digit, each on a separate knob,  and then estimate the third digit!   Because these knobs are part of a bridge, you can't go wrong with what you read.  That is why the CT160 has three digits accuracy, instead of two digits with a Mk2, Mk3 or Mk4.  You have to check the picture above, where you see two knows with "anode current" written under it. You can see the right is set at 50mA in the picture.   The left is between 3 and 4mA.  So you are above 53mA and below 54mA for sure. The third digit is the estimation. The left knob is on this case is half way between 3 and 4. That would be 53.5mA when it is exactly half way. When we assume that the accuracy of this estimation is 0,2mA we get an actual reading of 53,4....53.6mA. You can never do that on a Mk4. 

HOW DOES THE CT160 WORK?

As with most (good!) military testers, it is made fool proof, and works fool proof.  So you can make no damaging mistakes easily, and with little efforts, you can get precise results out of it.  I think this instrument was an overkill for the army.  For engineers,  it can measure plate current,   Transconductance, and draw tube curves with an XY writer by opening the plate current link and put the XY writer in there.   So if the bridge is set to zero (the needle is at zero)  you can read  Gm and at the same time (!) read the DC current from the plate current knobs  (or connect an external mA if you want).  For non-technicians there is a setting that produces a good/bad reading too. It can measure all kind of leakage, with the heater on or off.  Any way you want.  Of course it has a vacuum test too, that can be calibrated. 

Very short explanation:  It has a Wheatstone bridge inside.  The left side is a known tube and a  known resistor.   The known tube is inside, and the only tube it has. The rest is solid state. The right side of the bridge has a known resistor and the tube under test.  So the bridge output is connected to the panel meter, and you have to zero the meter.  You can either zero it with a given grid voltage, and change the plate current until the bridge balances,  OR you can zero it with a given plate current, and read the grid voltage from the bridge potentiometer.   The last is very nice, and very protective for the tube.  I have seen this on no other tester than the AVO. 

Now,  once the bridge is balanced, and you can read plate current and grid voltage from it,  you are going to off-balance it with a know voltage.  The off-balance is read on the meter, and is directly the Transconductance  It is a bit unusual when you do it for the first time,  but it works nice and quick, and amazingly precise.  So you can measure Transconductance at any working point YOU choose.  Lets be honest, which tester can do that?!  All Hickock just say:  this is the Transconductance, and say nothing about the working point.   With the AVO it tells you the working point, and you can change it.  Or just take the manual and take the standard recommendation for the tube under test. So  check at what plate current you should test Transconductance, and what the corresponding grid voltage is supposed to be.  Read Transconductance directly in mA/V, and know what you have.  For the military dummies there is also the option to read Transconductance on the good-bad scale.  Both can be done.  For the last test,  the tester is adjusted for each tube so the "good" scale start exactly where is has to start for a certain tube. 

Rectifier testing is a book for itself, and many testers test "something" which tells you something about the condition of the tubes. But... what is it you want to know? The AVO CT160 is really straight forwarded with that. It simply plugs in the tube under test in a real rectifier circuit, and them when you set it at 120mA,  it expects 120mA rectified DC current from the rectifier. Now this is for me the only real way to test a rectifier. So the way this works is this: You take any rectifier, and rotate the knob through  1,5, 15, 30,  60, 120mA.   When the needle come in the green, the rectifier is actually supplying that current.  So a 60mA tube must test in the "green" area at 1,5, 15, 30 and 60mA, but not necessarily at 120mA. Then the tube is OK. With big tubes at 120mA the load resistor gets very hot, so test it  not too long.   Don't use that setting to burn in tubes. 

With all those excellent possibilities,  for me the CT160 is better than the famous Mk4,  because the CT160  accuracy comes from the resistor decades in the bridge. With the Mk4 the accuracy comes from a potentiometer scale.  We all know  a decade bank gives a better reading than a potentiometer scale.  So the  Mk4 has  a  more convenient way to measure plate current,  but at the cost of less precision.  When you know the Mk4 is an AC tester....  no DC signals are used....   then you quickly start to prefer the CT160.  My conclusion:    The CT160 is the BEST tester AVO made. Only the Mk2/3/4 look nicer, but it these don't work as good as the CT160.      

Here are the repairs done to this tester: The overload relay just actives itself sometimes for no reason, and  when you tap on the case, the needle would jump up and down 1/3 of the scale.  It needed new paint.  The base material was very nice,  and not repaired to death by previous owners.  It was all in original condition, and apparently was serviced. Switches were clean,  etc. Just the previous owner did not notice the problems came from the bad potentiometers.  Now read this:  Underneath the cover plate, I found a stash of pretty rare tubes, carefully packed in plastic wrappings.  What a strange place to store tubes... but OK.,   I gladly accepted those.  It made my day :)   So it was owned by someone who must have loved this instrument.   So because I had a good feeling about that,  I decided to do the extra work and restore it.   What was done was this: 
 

  1. Repaint the case.  It looked terrible. 
  2. Repair the overload relay. It did not reset. 
  3. Repair the grid pot.   It was unstable, due to a loose contact inside
  4. Repair the Transconductance pot.  It was unstable
  5. Check each and every resistor and capacitor. Several had  handmade values.  Apart from the grid calibration resistors, they were all good, which saved me a lot of time :)   So the tester was always stored good
  6. Many small things not mentioned here
  7. After the potentiometer repairs, calibrate it using original instructions.  This actually went quick, after  the potentiometers were repaired. 

Now,  it is a perfect instrument again :)   I sell it because I can't keep them all. 

 


Case needed new paint,  first primer,  then original color


All hardware acid cleaned.


Now come the repairs

The overload relay didn't work good.  When it functioned,  it wouldn't release again.  I had to open the tester to knock on the relay, and then it would release.  I bought this one as a "good working" tester! I am sure the previous owner knew about this defect. This relay looks so simple, but it works works incredible difficult. It has a hold function, which works magnetic, and will hold also with the power off. I never found out in detail how this works. It has three coils, and also it can make a buzz noise, short before it activates. So the noise warns you. There is no information available about how to set it, and I think you were supposed to replace it when there is a problem with it. At Ebay I saw good working relays sell for skyhigh prices. This makes sense, because it is hard to repair and you can not work without it. So I am allways happy when the relay works good, and then don't touch it.

The relay of this had one adjustments screw totally screwed out, which can not be right.  One screw is for the sensitivity,  the other is for the hold function. On the picture you see the broken nut that holds the adjustment screw.  These are the kind of nuts with build-in friction, and old style thread. That part broke off.   Interesting, everywhere inside  the same nuts were used, so I could just pick another good one from there. 
A close up from the relay.  It has three coils, for reacting at three possible problems at the same time.  Before the relay functions it starts to buzz with a bad sounding noise, which sounds quite alarming.  So you hear you overload something, but it's only the coils that make a sound.  At the same time a red lamp starts to backlight the scale.  At severe overload,  the relay comes fully, and then has a hold function.   The tester switches off,  and a red  lamp backlights the meter scale. If this happens you need to switch on and off the tester. 
The lamp is a special 200V, 15 Watt Osram lamp, and the lamp current is in some way used for the relay too.  I think if it fails, you can replace it by a 220Volts 15 Watt lamp, and put a small resistor in series.   Anyway,  the original lamps looks like a new one,  so no problem for this tester.
I have a Mk2,  it uses the same meter. 
Just some picture of the wiring and switches. Untouched, still all OK.
Click to enlarge.   Note the separate filament transformer.  It has a "high" and "low" setting. Very nice principle allowing fine filament voltage adjustments.  They just connect the filament transformer primary to two different voltage adapter taps of the primary mains transformer. 
Each filament voltage has 12% variation possibility, with the toggle switch.  Like this you can make fine steps, and adapt better to the real filament voltage.  When using a KT88 or ECC81, both 6Volts, you will find different voltage.  ECC83 you get a closer to  6.3V by putting the switch on 5.7V.   That is because the tube draws almost no current.
When I see this, I get this feeling that I want to test tubes with it :)
This very special construction I only know from this tester.  Very pleasant to work with!
Click on the picture to enlarge.   This is the pot from the Gm wheel above.  It has a contact that is needed to distinct between "unused Gm wheel"  and Gm Wheel in use.  The contact was noisy,  that caused a lot of trouble.  Took me 4 hours before I located the problem here.  The fix itself was 5 minutes.    This is the same construction pot as the grid pot.   The grid pot had a noisy contact inside.  These pots can be taken apart. Not  easily, but it can be done.   I left the Gm pot closed,  since it worked OK. 

Contact again from the Gm Pot.   The silver needed cleaning, and the contact itself adjust it a little bit for more pressure. 

 

AVO used three calibration  resistors,  and these are in the grid calibration network.  These were hand picked.   In needs to have exactly 1,32 Megohm.  I could get that value nicely with E13 resistors but needed four.  When these resistors are wrong, you can't calibrate the Grid voltage reading and the Gm reading at the same time. I found that out.   So when the calibration resistor (blue resistors in picture) is wrong, you get following problem: First you calibrate the grid voltage, and you do it correct. Then you calibrate the Gm but after that the grid voltage is wrong again. So you readjust the grid voltage, but then Gm is wrong again. You never can solve it. After putting in exactly the value as it says in the calibration instructions,  all I needed to do was calibrate the grid voltage, and automatically the Gm was in calibration too. And that really only works when you put in EXACTLY the right value.  That is why those resistors must be hand picked.

Worth some attention.  This connector probably is made for some tubes,  but.... you can very nicely connect anything else to it,  to get electrical contact to all  functions.    So if you put the rotary witch Nr1 on  position G, as done in this picture, the grid voltage is connected to "1" of this white socket.   So this allows universal connections to anything. Like this you can test even solid state diodes with it, or resistors.  When I have more time, I'll check this with a solid state diode.   See if I am right.

 
NOT SHOWN HERE,   it has a Shielded mains cable :) All in all I spend 10 hours restoring it.  A bit more work than expected, but  fun to do!   The result is really very good, it is a very accurate tester, I compared it with many other (very good) ones I have.