The Radio Constructor Dynamic Mutual Conductance Valve Analyser
 

I do rather like the look of the 1956 Radio Constructor 'Dynamic Mutual Conductance Valve Analyser' (see attached pdf). Has anybody constructed one?

Peter
G3PIJ

Re: What has happened to all the AVO type valve holder panels
 

Hi Peter, I think I remember seeing that the first time round and it got a few favorable comments on the letters page.
I looks as if it could operate with the Sussex quite successfully.
My thoughts on this unit are that it could also be used with a computer sound card for distortion measurements at various points on the valve curve and under different loads as well as a means to match levels of distortion across a range of valve types/ batches.
By using 2 frequencies intermodulation distortion could also be analysed.

It would be possible to build your own "ultimate" amplifier and have actual figures and measurements to prove it.

Ed

Re: What has happened to all the AVO type valve holder panels
 

The Radio Constructor article warns against stray capacitance in the grid and anode circuits. I wonder if the sort of selector switching shown in the Sussex photos in post 6 at https://www.vintage-radio.net/forum/...ad.php?t=48853 would introduce problematic strays - or are the Radio Constructor grid and anode circuits at sufficiently low impedance (25 ohms?) for this extended wiring not to matter?

Peter G3PIJ

Re: What has happened to all the AVO type valve holder panels
 

Hi Peter, further thoughts on this one to improve performance etc.

There is an excellent book "Alternating Current Bridge Methods" by B Hague, very much the standard work years ago. This gives all the tips and tricks needed to construct a stable and repeatable bridge.
I'm a little surprised at the possibility of stray capacitance problems at the frequencies involved but these can be over come with screening/ neutralisation.

The accuracy can also be improved by using better accuracy resistors, but film types may be inductive due to the trimming spiral used to manufacture them. For the low values required they can be bifilar wound using nichrome or similar wire nad set accurately, or more important, balanced.
A modern electrolytic should have a low ESR for the C4 position, but suitable plastic film caps could also be used in parallel. If mounted close together, and "ends about" it should be possible to cancel out the inductance of the film windings.

Use of a mutli turn pot and /or a small ac voltmeter (as Sussex) for both signal and GB voltage will allow better setting.
C3,5 and 6 should be low inductance types and preferably matched.

There are also techniques to use a coarse/ fine balance control to give a more accurate balance.

Ed

Re: The Radio Constructor Dynamic Mutual Conductance Valve Analyser
 

I have been working on this for sometime!! Another very similar design is available from March 1952 Practical Wireless magazine. I am thinking to combine the two instruments into one with variable supplies and multiple digital meters.

Joe

Re: What has happened to all the AVO type valve holder panels
 

Using a single 50 k linear balance pot gives a very cramped scale for gm values greater than 4. Values between gm=5 and gm=14 are found between 83% and 93% of full rotation. What about a log-taper pot wired in backwards?

I am also thinking about a 5k pot wired as a variable resistor in series with the lower arm of the balance pot - i.e. in the grid connection. So gm=0 to gm=4 will be covered by the main 50k pot with the auxiliary set to zero; and then, setting the main pot to full clockwise (50k), the auxiliary variable resistor covers gm=4 to gm=15 but spread over a greater angle of rotation.

However . . . I think that I should now build the thing as published and then go on to experiment with the bridge balance. Short of shielding all grid and anode leads, I am still not sure how to limit grid-anode stray capacitances in the bundles of wire that will loop around between seven selector switches and half a dozen valve holders.

Thanks for the PW suggestion - I shall check and add to my thinking.

Peter G3PIJ

Re: The Radio Constructor Dynamic Mutual Conductance Valve Analyser
 

Update: I have completed the wiring of the ‘Dynamic Mutual Conductance Valve Analyser’ - see http://www.r-type.org/articles/art-118.htm - but have yet to connect it to multiple PSUs, an audio signal generator and an oscilloscope to find out if it actually works. With an audio generator and an oscilloscope available, I reckon that V1 and V3 are not needed. The balance pots must be calibrated first of all but that exercise is being put off until the solder fumes from 120+ solder joints have died down. If this device does work correctly as advertised in 1956, then I intend to work my way through the myriad valves in the collection here and carry out a rigorous triage. Just how many EF91s do I really need / want ??!

Peter (G3PIJ)

Re: The Radio Constructor Dynamic Mutual Conductance Valve Analyser
 

The workings of this design were discussed on the Forum in some detail about a year ago https://www.vintage-radio.net/forum/....php?p=1122885.

You (Peter) might be the first person to complete the construction of it.

B

Re: The Radio Constructor Dynamic Mutual Conductance Valve Analyser
 

The advantage of bridge circuits is that they use a cheap, simple, uncalibrated null detector such as a pair of headphones.

If you have calibrated test gear such as AF generators and oscilloscopes, isn’t it easier to just feed a known AC voltage to the grid and measure the AC voltage across a low value anode resistor to give you the corresponding anode current? With an AC millivoltmeter, you could even have a direct reading mA/V scale.

But this thread is in the home brew section, not the test gear section:)

Stuart