UK Vintage Radio Repair and Restoration Discussion Forum

UK Vintage Radio Repair and Restoration Discussion Forum (https://www.vintage-radio.net/forum/index.php)
-   Vintage Test Gear and Workshop Equipment (https://www.vintage-radio.net/forum/forumdisplay.php?f=36)
-   -   VCM163 Calibration Question (https://www.vintage-radio.net/forum/showthread.php?t=148077)

cathoderay57 12th Jul 2018 5:50 pm

VCM163 Calibration Question
 
Hi folks,

I have been attempting the calibration procedure in the Avo Service Manual for my VCM163. In the Setting-up Procedure, Page 7 Para (s) it says "Connect the electronic multimeter across the red and brown (earthy) sockets on the amplifier board. Set the electronic multimeter to the 30mV range (or as appropriate) and adjust RV1 on the oscillator board until the multimeter reads 15mV".

There appears to be an error in the above script as there are no such sockets on the amplifier board and so I assume it should read oscillator board. Connecting to the red and brown sockets on the oscillator board, my multimeter (on 200mV AC range) displayed a voltage briefly then the voltage decayed away so I assume it was loading the circuit. Therefore, I used the scope instead. Since a meter would display RMS voltage, whereas I am using a scope, should I adjust RV1 for a 15mV peak voltage or 1.414 x 15mV=21.2mV peak?


Cheers, Jerry

cathoderay57 12th Jul 2018 8:16 pm

Re: VCM163 Calibration Question
 
I should maybe add that the oscillator board outputs to which the multimeter is intended to be plugged carries a 14.7kHz sinewave signal. The measurement was intended to be made using an Avo EA113 electronic multimeter which is of a very high impedance. I don't have one but my assumption in the above post is that whatever the EA113 did, it had an analogue meter display, and so on the AC range with a sinewave input it would probably display an RMS as opposed to a peak value. Hence my question whether using a scope I should set the value at 21mV peak (=15mV RMS). I have in fact already done this and there is sufficient range on RV1 and RV2 to align the gm meter at the cal mark so I will leave it like that unless somebody convinces me differently. Cheers, Jerry.

Radio Wrangler 12th Jul 2018 10:00 pm

Re: VCM163 Calibration Question
 
The EA113 uses FET-input amplification and an input attenuator of very high resistance. It drives a diode rectifier and so responds to peak voltage, not RMS, but it is scaled to show what the RMS voltage would be on the assumption that the waveform is a sinusoid.

Without the careful language, it'll read RMS on sinewaves, and it'll be wrong on anything else.

With some of the waveforms floating around in valve testers (halfwave rectified and things like that) meter readings are very wrong, though the sinusoid applied to the grid for gm measurements ought to be sensible. Other voltages in the machine may have values where you do need the cited meter to get the same amount of wrongness :-)

I checked my EA113 some time ago and found the accuracy excellent on DC and sines. On AC, it seemed OK up to 200kHz. Checked against the Datron calibrators in HP's standards lab. With 10mV FSD on its most sensitive range, it's a useful audio millivoltmeter, and the 10mV DC range is most unusual on an analogue meter.

David

Bazz4CQJ 12th Jul 2018 10:23 pm

Re: VCM163 Calibration Question
 
That service manual never went past a technical editor! I hadn't spotted that mistake but the whole thing is jumbled up.

I've been looking at my old notes to try and see how I've set the oscillator up previously but cannot find any specific mention. However, I don't have a DMM which would measure 15mV AC so I guess I must have used the scope to do it. I did make a note that I reckoned the frequency was 15.8kHz (not the 14.7 usually referred to) so that also implies I had the scope on it.

My 163 has a meter on the gm side which is faulty (shorted turns on the coil) so I've had to fit an op-amp in there to achieve a correct reading. It always seems to me that owning an AVO VCM is a hobby in its own right.

B

Radio Wrangler 12th Jul 2018 10:40 pm

Re: VCM163 Calibration Question
 
It's always struck me that for a firm based on d'Arsonval meter movements, they seem to have been much worse than other makers. I've got loads of other makes of movements of around the age of the VCMs and all have intact coils. Maybe the total lack of effective protection?

David

Phil G4SPZ 12th Jul 2018 11:32 pm

Re: VCM163 Calibration Question
 
Jerry, I've replied to your PM and it seems that I have said much the same as David. I'll add here for the benefit of others that the EA113 has an input impedance on all AC ranges of 1 Megohm shunted by ~25pF, according to the instruction manual. Hence if you can find a sensitive valve voltmeter or high impedance AC millivoltmeter (even a DVM) you may be able to shunt it to simulate an EA113.

I haven't got a VCM163 myself, but I reckon you have done pretty well using your own calibration method. Perhaps testing a couple of valves with known characteristics would give you more confidence.

Phil

Bazz4CQJ 13th Jul 2018 12:51 am

Re: VCM163 Calibration Question
 
1 Attachment(s)
I've just been looking at the circuit diagram of the oscillator board. The test points (red and brown sockets) used to set the level actually measure the voltage across R19, a 10ohm resistor (+-1%) see attached.

Perhaps I'm missing something, but does that really need such a high impedance test meter or is it just a case that any meter that will accurately measure 15mV at 15kHz will probably be a 'quality' instrument anyway?

B

Radio Wrangler 13th Jul 2018 5:10 am

Re: VCM163 Calibration Question
 
I wouldn't get too precious over that calibration, that resistor is sensing the current through the feedback loop thermistor of an audio oscillator, which isn't quite the same thing as the output level. AND the output goes through a different type of thermistor and a twiddle pot before it goes to the transformer and the valve under test. This is not high precision stuff

An audio millivoltmeter with a few k ohms or greater input impedance would be fine.

You have to look at this circuit to fully appreciate the benefits in the Sussex design :-)

David

Craig Sawyers 13th Jul 2018 7:21 am

Re: VCM163 Calibration Question
 
In which case, the voltage across that 10 ohm resistor via the red/brown test points will only reveal that the oscillator is oscillating. Those test points have nothing to do with the output setting resistor.

So the OP's dilemma is still unresolved.

cathoderay57 13th Jul 2018 8:41 am

Re: VCM163 Calibration Question
 
Thanks for all of your replies. The only things I can add are that:
1. I initially used an Iso-tech IDM91 Digital Multimeter on the 200mV AC range to measure the said voltage. As I said in the post, it showed a few millivolts initially (I think around 10) then the voltage decayed to zero.

2. The scope used was a Tektronix T912 that has an input impedance marked as 1M 30pF so looks to be a fairly close match to that of the EA113.

3. I don't have a standardised valve to complete the VCM163 cal proc but used an ECC81 that appears to be new and the test results were as expected so hopefully all is well.

I'll keep a sceptical eye on the test results for valves for a while and if I get any unusual readings I'll try testing a similar valve for comparison before I do any readjustment. Cheers, Jerry

Radio Wrangler 13th Jul 2018 12:58 pm

Re: VCM163 Calibration Question
 
Some people have DC-based valve testers whose accuracy can be verified against calibrated voltmeters, and with them, they can create standardised valves.

One very generous regular on this site made a group of valves with accurately measured individual characteristic curves and then shared them round interested parties.

But if you're testing valves in order to see whether they are the cause of a problem in a radio, then getting a reasonable result for a trusted ECC81 says you are as close as you need to be.

If you want to pick a matched pair of some type for a push-pull amplifier stage, then you are doing a comparison and you don't need absolute accuracy for that.

Swap a few valves with other people and you can build confidence that all is well.

David

Bazz4CQJ 13th Jul 2018 3:33 pm

Re: VCM163 Calibration Question
 
Although the 163 needs a fair bit of TLC at its current age, my experience over the last 10 years is that they are quite accurate and that the unique (?) mode of operation does indeed work well.

Referring to the recent exercise on this Forum initiated by David Simpson, that seemed to me to demonstrate that the well maintained 163's which participated in the exercise gave excellent results. Ultimately, the 163's service procedure does need to checked by using a reliably standardised valve. Using valves which I've DC standardised myself, and one from David, I've always been happy with the results for the 163.

An alternative "self check" method for the 163 is to draw the Vg-Ia curve of the valve just by carefully setting the Vg on the front panel control over a range of voltages around the nominal setting, noting Ia. You then have go and plot the results in Excel or similar, but what you should then get a value for gm which you can compare with what the 163 tells you on its gm meter. In my experience, those two values tend to be very close.

Over the years, I've used various valves as standards, but on the 163, a particularly good choice is an ECC81. A carefully selected one will have Ia and gm values which will be almost FSD on both meters on their lowest ranges, and will just about give adequate deflections for checking the meters out on the higher ranges.

My view of the 163 is that the build quality could have been improved, but it does deliver the goods!

B

chriswood1900 13th Jul 2018 5:58 pm

Re: VCM163 Calibration Question
 
Jerry
If I can be any help send me a PM I have some standardised valves and an EA113 and may live quite close!
Chris

cathoderay57 14th Jul 2018 9:10 am

Re: VCM163 Calibration Question
 
Thanks again to all. Chris you have a PM. Cheers, Jerry

cathoderay57 15th Jul 2018 8:55 am

Re: VCM163 Calibration Question
 
Very grateful to Chris. Checked out the VCM163 with 2 standardized valves yesterday and performance was as expected. Mods please close thread. Cheers, Jerry


All times are GMT +1. The time now is 7:09 am.

Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.