Cossor 1044K Valve Voltmeter.

[Joe]

This was part of my latest auction haul.

I saw at the viewing that the glass was loose and fouling the needle so my first job was to sort that out. I refixed it using PoundLand roofing sealant. This may seem an odd choice but it looked like what had originally been used and I've used it before for such small-scale glazing jobs.

I then put it back together and it seemed to work but I then found that the needle was still catching the glass as I'd pushed it in again while re-assembling it. I've now done a better job on that, knowing that it's worthwhile, and will let it fully harden overnight.

There doesn't seem to be any information about this model on t'inter-web except for a few e-Bay listings. It does seem though that it was originally supplied as a kit which explains why everything inside is well labelled.

The valves are a Cossor (not surprisingly) 12AU7 and a 6AL5. There are no wax or Hunt's capacitors but a couple of electrolytics which may warrant closer examination.

As well as AC and DC voltage it measures resistance so could well prove it's worth, especially as I've yet again left my DMM on all night and flattened its battery.

I'll take some photos when it's back together.

- Joe

[Skywave]

Hi Joe.
I don't know how much that VTVM cost you - and I don't need to know - but in my opinion, these days a VTVM is a rather under-rated item of test kit. Agreed, modern technology has given us DMMs that do a similar job - primarily measurement of voltages with minimal loading on the cct. under test, but a VTVM is always useful as a device to confirm that reading and unlike a DMM, will usually enable measurements up to several MHz - with a diode detector. (And if it's mains-powered, you don't have to worry about flat batteries! ) Plus, of course, VTVMs are a lot easier to repair and aren't as prone to giving misleading indications as modern DMMs can do. Of course, if a 'scope is to hand, then that will tell you everything that a VTVM will, but not everyone owns a 'scope, and even then a VTVM is usually a good deal smaller that a 'scope (thinking of the ever-shrinking bench-top space.) And then again, there are some maintenance procedures that specifically call for a VTVM to be used: the handbook for the Racal RA-17 springs to mind.

I look forward to the pictures that you say you'll upload.

Al.

[Joe]

I quite agree, Al. It seems to have the high impedance of a DMM but a 'proper' dial. It also looks like it should be good for measuring high resistances; my DMM only goes up to two meg.

I'll photograph it when I put it back together tomorrow and give it a better test too.

- Joe

[Joe]

Pictures:-

       

The glass has now stuck properly in place but it doesn't seem to work very well. Resistance is a dead loss and the voltage readings are way out. I am at least pretty sure that the movement is OK so will investigate the electrolytics first and then hope somebody can turn a diagram up rather than fiddle about blind; I'm not in a rush to use it but think it will be very handy when sorted.

As for what it cost, Al, I truthfully don't know as it was part of the job-lot.

I've been wondering if there were ever transistor-based versions of these, ie a mechanical movement driven by transistors, or did they go straight from valve to digital?

- Joe

[Joe]

I will now try and explain how this is acting.

Firstly resistance. It has both 'Set Zero' and 'Set Inf' knobs so I assume that the leads are shorted together and 'Set Zero' adjusted so the meter reads zero and then with them disconnected 'Set Inf is used to make it read infinity. The zeroing bit works but with the leads disconnected the needle is hardly above zero and the 'Set Inf' knob moves it very little - maybe the thickness of the needle.

Now for voltage. I've tried sticking various potentials up up it with the following results:-

   

I'll try and rule out anything obvious and post an appeal for a diagram o'er yonder.

- Joe

[Skywave]

If you can discover a cct. diag., obviously that will help. But looking at your results, there are some clues as to where to look for faults.

Your assumption about setting up the 'Set Zero' & 'Set Inf' is the same as my understanding.
As it stands, the 50 v. F.S.D. has the most linear scale, albeit reading very low. As the gain of the amplifier within the instrument is increased, indicated by progressing to more and more lower F.S.D. settings, the linearity slowly degrades. The fact that on the 50v F.S.D. range, the reading are all too low implies that high-value resistors in anode circuits have probably gone high in value. The increasing non-linearity as the gain is increased could be due to defective bias on the amplifier valve, leaky capacitors or possibly a low-emission valve. Finally there is the question of the HT: if that is seriously too low, then that could easily explain all those symptoms!

So treat it like any other valve-based item of kit that's in for a repair: check the HT; check the value of all the high ohmic value resistors; check for leaky capacitors; clean switch contacts and finally replace a valve or two. The instrument will certainly need re-calibrating via its internal adjustable pre-set resistors: even these might be defective, although unlikely.

Al.

[Yorkie]

The only thing I'd add to what Al has said is that often in older test gear, the HT is rectified either by a stick rectifier or contact cooled rectifier. (EG, Hunts CRB3 etc). Over time, if those rectifiers don't fail, they often give low HT. Easy enough to rectify ('scuse the pun!) by replacing with a modern silicon rectifier diode.

Some test gear did use high stability wirewound resistors in the divider chain, (usually recognizable by their black painted appearance) which often retain their values, but any carbon composition resistors tend to drift high over time. Maybe not so important in a radio, but if they're in the measuring circuit of test gear, obviously they throw the calibration right out. The good news Joe is that it does actually work, if only after a fashion!

[pwdrive]

I've been wondering if there were ever transistor-based versions of these, ie a mechanical movement driven by transistors, or did they go straight from valve to digital?

Yep, transistor ones were made, typically 1 M ohm per volt, a good valve VTM with a decent diode probe would typically be 2 to 3 M ohm input impedance to the probe and would be good from approx 20 hz to 100 mhz.
If you can't get the diagram then try to reverse engineer the circuit, shouldn't be too difficult on a quiet day. The 12AU7 might well be a balanced amp, if it is then I doubt if it is the simple wheatsone bridge type Valve VTM.

Lawrence.

[Joe]

I did a bit of digging 'round last night and found this model is suspiciously like the HeathKit V-7A, the only obvious difference being the sockets.

This also revealed that it should have a battery in it so after fitting one the resistance measurements 'work' but are not accurate; a 1KΩ resistor reads 1.2KΩ and a 10KΩ reads 18KΩ.

I've checked the HT which is only 50V across what I take to be the smoothing capacitor. This itself is fine in terms of both leakage and ESR so it looks like the rectifier is to blame. I take it a silicon diode would want a 100Ω resistor in series with it?

I've spot-checked a few of the precision resistors which all seem to be bang on - that is one job where a digital meter scores over an AVO.

More later.

- Joe

[Joe]

I've had a better look at the HT now. According to the Heathkit circuit - and I can't see this bit being much different as they are the same valves - there should be 50-80V positive relative to chassis at one end of the smoother and 60-85 negative on the other. Mine reads 24V and 29V. As there's 165V AC going into the rectifier it seems like that is past its best.

I attach the Heathkit circuit below and have put the whole manual in The Archive.
   
- Joe