UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > General Vintage Technology > General Vintage Technology Discussions

Notices

General Vintage Technology Discussions For general discussions about vintage radio and other vintage electronics etc.

Closed Thread
 
Thread Tools
Old 19th Oct 2012, 10:24 am   #1
radiozero
Hexode
 
Join Date: Mar 2008
Location: Wakefield, West Yorkshire, UK.
Posts: 438
Default Valve Testing

Is the "Goodness" of a valve a measure of how close it's mutual conductance is to a published figure?

And, assuming no insulation problems, or gas, is "Goodness" more-or-less a measure of cathode electron emission?

Thanks.
radiozero is offline  
Old 19th Oct 2012, 1:18 pm   #2
Peter.N.
Dekatron
 
Join Date: Jul 2008
Location: Charmouth, Dorset, UK.
Posts: 3,601
Default Re: Valve Testing

Assuming there are no inter electrode faults and the vacuum is good I would say that the amount of current flow is a good indication, so 'yes'

Peter
Peter.N. is offline  
Old 19th Oct 2012, 3:27 pm   #3
GrimJosef
Dekatron
 
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
Default Re: Valve Testing

Valves are often characterised by two numbers - the mutual conductance, mentioned by radiozero, and the amount of (anode) current flow, mentioned by Peter N. The latter is controlled by, among other things, the cathode emission. Some simple valve testers don't measure mutual conductance, so their 'goodness' indicator has to be based on current i.e. emission. More sophisticated testers could, presumably, use mutual conductance as the measure of 'goodness' but whether they do or not I'm not sure.

On p4 of their KT66 data sheet http://www.tubezone.net/pdf/kt66-mov.pdf GEC specify that the valve will have reached end of life when its mutual conductance has fallen below 5.5mA/V under the specified measurement conditions. Of course loss of emission will lead to loss of mutual conductance. In practice the two numbers are somewhat correlated.

Cheers,

GJ
GrimJosef is offline  
Old 19th Oct 2012, 4:10 pm   #4
Herald1360
Dekatron
 
Herald1360's Avatar
 
Join Date: Feb 2007
Location: Leominster, Herefordshire, UK.
Posts: 16,526
Default Re: Valve Testing

I vaguely remember "goodness" having something to do with the mu or amplification factor of early triodes.... though I might be confusing it with Q for inductors.
__________________
....__________
....|____||__|__\_____
.=.| _---\__|__|_---_|.
.........O..Chris....O
Herald1360 is offline  
Old 19th Oct 2012, 4:50 pm   #5
radiozero
Hexode
 
Join Date: Mar 2008
Location: Wakefield, West Yorkshire, UK.
Posts: 438
Default Re: Valve Testing

I'm not sure how you go about measuring mutual conductance. Nor, how my newly acquired AVO MKI works.

The KT66 has a mutual conductance of 7.3mA/V. Meaning that for every 1 volt change in grid voltage, anode current change is 7.3mA.

I'm not sure how the AVO Valve Tester works. I can see that the meter measures anode current. But, how can reading the anode current tell you about the mutual conductance of a valve?

In my ignorant state, to test for mutual conductance I'd be thinking of arranging for a change of grid voltage by 1 volt and looking for a change in anode current of 7.3mA. If the change was below 5.5mA then considered out of spec.

Here is shown the AVO circuit.

I've not a grasp how the circuit works. I do know that the "Good" portion of the scale goes from FSD to 50% of FSD.

The AVO does measure mutual conductance. I believe. By pressing a "Mutual Conductance" button.
Attached Thumbnails
Click image for larger version

Name:	AVO Valve Tester.jpg
Views:	193
Size:	71.9 KB
ID:	71663  

Last edited by radiozero; 19th Oct 2012 at 4:56 pm.
radiozero is offline  
Old 19th Oct 2012, 5:39 pm   #6
radiozero
Hexode
 
Join Date: Mar 2008
Location: Wakefield, West Yorkshire, UK.
Posts: 438
Default Re: Valve Testing

I might be on the trail:

When you put certain voltages onto the grids, (as per AVO Valve Data ) an expected anode current will flow. That current is shown on AVO Valve Data (under heading "Data For Valve Characteristic Meter and Valve Tester Type 160") as 65mA. (No anode current figure given under heading "Data For AVO Valve Tester" - curiously! No idea why!).

So, I suppose that if cathode emission is good you will see 65mA on the scale. Anything less than 65mA tells you that you have something of a cathode emission problem. Here, you are not using the "goodness" scales on the meter. Just noting the actual anode current. However, the anode current is an initial indication of "goodness".

But, we have not measured mA/V yet.

To measure actual mA/V we obtain a balance condition concerning the meter (using the Backing Off Control). That sets the meter to read 0 mA. Now, the mA/V button is pushed and the anode current changes. If you now get 7.3mV on the scale, that means change in current is 7.3mV. If what I say is correct, and the valve is healthy, then pressing the mA/V button 1V was added to whatever potential was on the grid.

So far we have not used the comparative goodness scale. I don't know how that works.

Last edited by radiozero; 19th Oct 2012 at 6:02 pm.
radiozero is offline  
Old 19th Oct 2012, 10:55 pm   #7
radiozero
Hexode
 
Join Date: Mar 2008
Location: Wakefield, West Yorkshire, UK.
Posts: 438
Default Re: Valve Testing

I does help to read the notes.:c)

Okay, this is what I'm getting.

After you balance the meter and push the mA/V button a "small supplemetal change in grid bias in a positive direction" is made to the grid of the valve. So, I was right on that.

The meter is set to the expected mA/V reading. If you set to 100 and you got FSD when pressing the mA/V button, the mA/V figure would be 100. If you set the meter to 2.5, then at FSD, mutual conductance would be 2.5 mA/V.

Of course when simply reading the steady anode current, (without using the Backing Off Control and pressing the mA/v button), the meter does tell you what the actual anode current is (100mA; 25mA; 10mA; 2.5mA.)

P.S. Oh, the divisions 0-100 and 0-25 are both for anode current and mA/V. I thought there was a seperate 0-25mA/V scale.

One thing, if I'm correct about the goodness test, a valve is good when passing only 50% of rated anode current.

Okay, now left with the actual goodness test. This test uses the coloured meter scale and the Set mA/V control. I see that the meter is zeroed as before. But this time, the meter set to mA/V.

In the KT66 case, you would set Set mA/V control to 7.3 (triode connection).

Now, you press the mA/V button. So, how is this working? I don't know. All I can think of is that this is more-or-less the first test, which is based more-or-less on cathode emission. In other words, when the meter reads FSD (right end of green marking), that is a condition when 65mA anode current is flowing.

Okay, I have covered the three tests, but curiously I have not as yet actually made use of the Ma/V scale on the meter!! I've read anode current, read change in anode current with Meter switch set to expected mA/V and read coloured comparison scale.

P.S. Oh, the meter has two scales 0-100 and 0-25, and is both anode current and mutual conductance.

If I am right a valve is good when only 50% of rated anode current flows.

Last edited by radiozero; 19th Oct 2012 at 11:24 pm.
radiozero is offline  
Old 19th Oct 2012, 11:22 pm   #8
Lucien Nunes
Rest in Peace
 
Join Date: Apr 2008
Location: London, UK.
Posts: 2,508
Default Re: Valve Testing

In the real world, 'goodness' could be a measure of either emission (anode current) or mutual conductance, or a function of both according to the particular demands made upon the valve in-circuit. Any given valve tester would show goodness typically as a ratio of one or other of these parameters relative to the figure for a new valve. With a basic tester the assumption is that if the valve is faulty, it will perform poorly whatever test is applied, so the tester only has to test one of these parameters. The need to make such an assumption stems from the high cost of building a comprehensive valve tester.

Measuring anode current at reduced voltage is easy and was the basis for lots of low-cost testers. However they would not generally measure the current under representative conditions, so could not be used for quantitative analysis of the valve performance, only comparison with published figures for a new valve. An arbitrary scale could be devised where a valve was considered OK if its emission was (say) at least 60% of the published nominal, although how this compared to that particular valve's new performance would be an unknown, as would the mutual conductance.

Measuring mutual conductance was a bit harder but not excessively so if one was again prepared to compromise and operate the valve under non-standard conditions. This is the basis of the original AVO valve tester, which uses zero-bias conditions to simplify the circuit. It does not display a useful figure for anode current but concentrates on measuring Gm. After zeroing the meter, you can read off the actual Gm in mA/V while the Set mA/V control remains at its initial 100 position, or by dialling-in the expected value on this control get a comparison with the valve under test using the red / green scale. The measurement is the same in each case, only the method of displaying it differs.

Measuring both mutual conductance and anode current under normal working conditions is the province of the characteristics meter, which can give a numerical answer to both tests. As an example, consider the case of the AVO VCM; when you have tested for sufficient emission, by comparing the meter reading with the published figure and drawing your own conclusion, you null out the standing current and switch to Gm. You can leave the dial set to the expected value and see where the pointer lands on the meter scale's coloured sectors (the goodness scale ranks +/- 33% as good and <50% as bad) or turn the knob until the pointer stands on the calibration mark and read the result off the dial.

So this is the crux of the coloured scale on a tester that measures Gm like the AVO - it is a ratio of the expected Gm that you have dialled in and that of the valve under test, by which the valve is ranked for goodness without specific regard to its emission performance.

Lucien

Last edited by Lucien Nunes; 19th Oct 2012 at 11:28 pm.
Lucien Nunes is offline  
Old 20th Oct 2012, 10:34 am   #9
radiozero
Hexode
 
Join Date: Mar 2008
Location: Wakefield, West Yorkshire, UK.
Posts: 438
Default Re: Valve Testing

Actually, the topic should be "Valve Testing on the AVO MKI & MKII".

That is really it's remit, well, at the back of my mind. I don't know if a moderator can change the subject header to that. I'd be grateful if that could be done.

I say this because of course the general topic of Valve Testing will be quite vast, probably too vast for my limited enquiries concerning a specific model of Valve Tester.

Having said the above, I do think post by "Lucien Nunes" has some useful relevence, as other posts.
radiozero is offline  
Old 20th Oct 2012, 10:57 am   #10
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,059
Default Re: Valve Testing

Well, the ยต of a valve is effectively defined by its geometry - so should be unchanging right up to the moment of death (unless the valve's demise is due to it having been severely shocked).

Similarly, the gm is defined by its cathode - grid geometry. Again it should be constant, unless the emission has fallen to the point that the cathode can't react to the grid and supply more current. Bearing in mind that an oxide-coated cathode has a soggy saturation characteristic anyway, a decrease in gm could indicate that emission is failing. Also, development of cathode interface resistance due to impurities in the cathode material would result in na decrease in gm.

Failing emission would also show up by the valve not passing a certain currenmt when specified electrode voltages are applied.

So, it seems to me that both gm and Ia are useful things to test for. Of course, gm is not applicable to a diode
kalee20 is online now  
Old 20th Oct 2012, 7:24 pm   #11
radiozero
Hexode
 
Join Date: Mar 2008
Location: Wakefield, West Yorkshire, UK.
Posts: 438
Default Re: Valve Testing

Let me first talk about the AVO MKI & II test that tends to come first (after insulation tests) - The Anode Current Test.

That test requires all electrodes to be set to a given voltage. The indicator examined, and which is not set on the Tester is - the anode current. The anode current "happens".

If the anode current is lower than what a table says it should be, then something is amiss. Perhaps usually a problem with the cathode supplying sufficient electrons. But, the current could be higher, due to a soft valve.

Depending on the percentage the current is, higher or lower, with regard to what a table says, then the valve can be accepted as suitable or not. But, what is that percentage? Unfortunately, that can vary depending on use of valve.

Now, I have in possession three tables: The manufacturers table for the specific valve, a table in a radio valve data booklet (which is probably just a collection of manufacturers tables for their valves) and AVO has produced a table for use with it's valve Testers.

Take the KT66. GEC in it's manufacturers table (just for the KT66) tells us that if the anode current reaches 85mA, then the valve is considered at the end of it's life. I'm not sure under what electrode conditions or other conditions that refers to. It would seem to me that perhaps it's 85mA when it's going soft. But, that is just a guess. Is that value of 85mA an in-circuit reading, or is it a reading associated with a valve Tester. It is a value that is NOT associated with a reading to be expected on a Valve Tester. I think.

Then we have the table from my Radio Valve Data book. It gives three anode current values: (BT) 62.5mA; (BT) 80.0mA; (T) 62.5mA.

Then we have AVO's tables. Now, the anode current is stated as 65mA. That is under the heading for a Type 160 Tester. There is no anode current mentioned under the heading for AVO Valve Tester (Presumably MK1 & MKII). But, surely I am suppossed to be looking for a current of 65mA. Interestingly, the electrode voltages are different for a MKI & II, compared to Type 160.

AVO made that table so that you can test valves, with their models of valve Testers.

The two tables then of real potential use are GEC's one for the KT66 and AVO's tables.

Whatever can be said of the usefulness of a particular anode current reading, certainly, GEC laid down something of a marker by saying the valve has reached the end of it's life when Ia equals or is above 85mA. In some conditions or other. And AVO give a value of 65mA for a new valve under a set of electrode conditions in a Valve Tester.

I have a couple of KT88's. Two questions: What value of Ia would be acceptable for audio use (AVO table says 57mA). Why does it say "No data Available" under the heading "Data For AVO Valve Tester"? Thanks.
radiozero is offline  
Old 20th Oct 2012, 7:56 pm   #12
TrevorG3VLF
Rest in Peace
 
TrevorG3VLF's Avatar
 
Join Date: Apr 2011
Location: Matlock, Derbyshire, UK.
Posts: 1,378
Default Re: Valve Testing

I think that the lack of data for the old AVO valve tester is due to the limited power supply. I tested a 6146 and it did its nut, blocking out the TV.
I also had problems with testing a EF80 which flashed over. Is this normal?
TrevorG3VLF is offline  
Old 20th Oct 2012, 9:28 pm   #13
Lucien Nunes
Rest in Peace
 
Join Date: Apr 2008
Location: London, UK.
Posts: 2,508
Default Re: Valve Testing

The original AVO valve tester measures only mutual conductance, not anode current, therefore no data is given in its section of the table. The more complex Valve Characteristic Meter (VCM) operates the valve under simulated* working conditions and is therefore able to measure both parameters, hence the anode current column included in its part of the table.

The manufacturers' figures correspond to a particular set of recommended operating conditions usually stated on the data sheet. The Gm figures in the Valve Tester column sometimes differ from these and from the figures given for the VCM, because the Tester uses non-standard voltages (for simplicity and reduced power consumption) resulting in slightly different performance figures for some types of valve.

An example: KT66
Osram 1944 data book recommends the following operating conditions: Va=250V, Vg2=250V, Vg1=-15V, in which case Ia is typically 85mA and Gm 6.3mA/V. The Valve Tester cannot provide these conditions because its power is too limited and its circuit too basic, hence it tests at 100 / 150 V and gives an expected result of 6mA/V which is a genuine reading under the alternative conditions. On the other hand the VCM can simulate* the recommended conditions correctly and gives an expected result of 6.3mA/V and a valid anode current reading.

What anode current is sufficient on a KT88? That will depend on your amplifier design. If you are coasting along in class A with 10W max. audio output (living room conditions) a pair of valves with only 45mA emission might work OK despite being near the end of their useful life. If you are hammering them in AB1 at 550V expecting 100W output then those valves would not make the grade because you will need a peak emission of up to 150mA

And yes, the original Valve Tester was not designed for later valves with high amplification and high power and can cause them to oscillate if you try to test them.

Lucien


*The VCM works with AC waveforms whereas the manufacturers define the operating point in terms of DC, so it's close to, but not exactly, the manufacturers' recommended conditions.
Lucien Nunes is offline  
Old 21st Oct 2012, 1:03 pm   #14
radiozero
Hexode
 
Join Date: Mar 2008
Location: Wakefield, West Yorkshire, UK.
Posts: 438
Default Re: Valve Testing

Let me just recap what I think I now know:

Manufacturers produced a set of static DC operating conditions for valves, for instance, the KT66. There were various conditions for this valve: As Beam Pentode, as Triode, and class A or AB1, single valve or double valve. These are genuine typical static DC operating conditions to guide the designer.

These manufacturers conditions are included in your general Valve Data books.

These conditions can be used in valve testing. But, they do not strictly have to be adhered to. Because you can alter the DC volts conditions and produce an appropriate value of anode current and mutual conductance associated with a new and healthy valve, which will of course differ from the published operating conditions. Many valve tester manufacturers have taken that approach, depending on the model. Then we obtain a set of tables specifically for valve testing, on a specific valve tester model.

This is what we see when we look at AVO valve testing data. We are looking at model-specific valve testing data, which may or may not agree with a valve's operating conditions. In fact, for the KT66 (not sure about other valves), neither the Type 160 or MKI or MKII testers manage to test this valve at it's manufacturer's operating conditions.

Now, we have the question of what can the MKI and MKII measure. Clearly the MKI and MKII can measure anode current. And, when you do, you can try to infer something about the valve. But, there is no anode current given under the heading "DATA FOR AVO VALVE TESTER". Current value is only given for the Type 160. So, what is the point of noting the anode current, when you have no figure to go by for the MKI and MKII? You are only given Gm.

I bought my AVO MKI hoping to test my KT88s. I hope there is a work-around. When I concocted a homebrew test circuit for current, I got oscillation. I killed it with ferrite beads. So, it readily oscillates. But, even if I kill oscillation, I have no test data for the MKI - "No Data Available" for the MKI and MKII. Apparantly.
radiozero is offline  
Old 21st Oct 2012, 1:15 pm   #15
radiozero
Hexode
 
Join Date: Mar 2008
Location: Wakefield, West Yorkshire, UK.
Posts: 438
Default Re: Valve Testing

One other point here: I'm looking at KT88 test data. Those test conditions under "Type 160" can be met by MKI AVO tester. But whether it has the correct base I know not.

Is data for Type 160 meant to apply to MKI and MKII? But not for the "AVO Valve Tester", the set before MKI and MKII?

Last edited by radiozero; 21st Oct 2012 at 1:22 pm.
radiozero is offline  
Old 21st Oct 2012, 4:58 pm   #16
radiozero
Hexode
 
Join Date: Mar 2008
Location: Wakefield, West Yorkshire, UK.
Posts: 438
Default Re: Valve Testing

Lets look at KT88:

For AVO Type 160:

Vf: 6V
Vg: -20V
Va: 250V
Vs: 250V
Ia: 57mA
mA/V:8mA/V

On my AVO MKI, I can set to the above test conditions. And I believe the MKI will take up to 100mA anode current.

So, I should be able to set to those test conditions and get an expected Ia of 57mA.

But, under heading "DATA FOR AVO VALVE TESTER" I read "No Data Available".

I'm confused over this.

Is there no data because the KT88 valve oscillates in MKI and MLKII? Or what.

And why, very often when I see data under both headings (heading for Type 160, and, seemingly for MKI and MKII) I see different test conditions?

Everything would make sense if the heading "DATA FOR AVO VALVE TESTER" was meant for the "AVO Valve Tester", that AVO set which has the meter part and a seperate sockets part. AND Type 160 meant also AVO MKI and MKII.

Can anyone please clear up my confusion? Thanks.

Last edited by radiozero; 21st Oct 2012 at 5:21 pm.
radiozero is offline  
Old 21st Oct 2012, 8:53 pm   #17
Dekatron
Octode
 
Dekatron's Avatar
 
Join Date: May 2009
Location: Linkoping, Sweden
Posts: 1,463
Default Re: Valve Testing

AVO Valve Characteristics Meter = Mk I, II, III, IV, CT160, CT160A & VCM163
AVO Valve Tester = AVO Two Panel Valve Tester

If you look in the 23rd edition of the Valve Data Manual (VDM) you will see that the VCM163 has two columns for the roller selector and the other VCM types have their own two columns.

Up to the 18th VDM there are values for the AVO Two Panel Valve Tester, but from the 19th VDM those columns are gone and only the VCM types are mentioned.

So any of the VCM types can test any valve in the VDM's as long as there is a socket and you can set the voltages required for the testing. There are however some small differences since the measurement circuit is not identical which results in that some valves test differently due to for instance low Ra.
__________________
Martin, Sweden
Dekatron is offline  
Old 21st Oct 2012, 10:01 pm   #18
radiozero
Hexode
 
Join Date: Mar 2008
Location: Wakefield, West Yorkshire, UK.
Posts: 438
Default Re: Valve Testing

Quote:
Originally Posted by Dekatron View Post
AVO Valve Characteristics Meter = Mk I, II, III, IV, CT160, CT160A & VCM163
AVO Valve Tester = AVO Two Panel Valve Tester

If you look in the 23rd edition of the Valve Data Manual (VDM) you will see that the VCM163 has two columns for the roller selector and the other VCM types have their own two columns.

Up to the 18th VDM there are values for the AVO Two Panel Valve Tester, but from the 19th VDM those columns are gone and only the VCM types are mentioned.

So any of the VCM types can test any valve in the VDM's as long as there is a socket and you can set the voltages required for the testing. There are however some small differences since the measurement circuit is not identical which results in that some valves test differently due to for instance low Ra.
Hi.


Five minutes ago, and before I saw your post, I suddenly realised my blind-spot.

I was looking at the first page of the MKI & MKII manual and it says "Introduction to THE "AVO" VALVE CHARACTERISTIC METER" and the penny dropped. I immediately realised that I have a VCM and the header on my AVO valve data pages, which says "DATA FOR VALVE CHARACTERISTIC METER & VALVE TESTER TYPE 160" is valid for the MKI and MKII & Type 160. I read it as only valid for Type 160! Like I say, I just had a blind-spot for a moment. Silly me. :c)

I am looking at AVO Valve Data that shipped with a Crowthorne Tubes CDROM.

So, eventually, I got there. Thanks for your input, which adds detail and provides comforting confirmation to the conclusion I had about five minutes ago. :c)

P.S. Folks wanting to test KT88's should, then, avoid the two-part AVO Valve Tester. For lack of test condition data. Unless there is a work-around.

Last edited by radiozero; 21st Oct 2012 at 10:16 pm.
radiozero is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 4:45 pm.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


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