Mullard High Speed Valve Tester Schematic

[Electronpusher0]

I am sure I have now found the cause of the problem, I am convinced that the card has a wrong hole punched!

I calculated the expected parameters from the holes in the Card (No 635). From this I determined that the HT should be around 270V, the grid bias -7.42V and the Indicator Anode current resistors selected should give a reading of 86mA at the dotted line in the middle of the green band.
Thank you PLJ for posting your calculator results, it shows my calculations are also good.
I checked these against the curves from the Mullard data sheet and found that at those values of HT and grid bias the Anode current should be 45 to 50mA. Indeed if 86mA was flowing at that HT the valve would be operating well outside its maximum power limit curve!

I started to suspect that the card may be wrong.
I calculated which Anode current sense resistors should be selected to give around 50mA at the dotted line and discovered that a single hole move would achieve this.
The holes on the original card were H5, H7 and H9. If H9 was moved to H8 it would give a value of 54mA at the dotted line.
I created a new card with H8 and no H9 to test my theory and used it to test a selection of EL84s, old and new. All of these have previously been tested on a 2 panel Avo tester and are known to be good.
They now all pass with several showing the dot right on the dotted line.
I can only conclude that the card 635 is indeed wrong.
I attach photos of the cards and a typical result on the tester.

John, is there any chance you could make a new card and confirm my findings?
Peter

[theredhouseinn]

Hi Peter.
I have investigated in more detail.
The heater voltages for the valve bases are derived from a seperate heater transformer so this would not be a good indication of the mains voltage setting.
Also the Avo is not accurate enough for the mains setting.
I have 3 digital multimeters and they all give exact same readings of the mains voltage.
During the day my mains supply is 240 +or- 1v. Overhead street cables. In the evening it can go down to 230v
After much deliberation I decided to measure the voltage across the pilot lamp as this is fed from the same winding as V4 and V7, should be 6.3v.
With the coarse tap set to 240 I had to move the fine adjust to position to 5,i.e. the 10v pos., to get 6.3 volts at the lamp. This would indicate that my mains voltage is 250, not true.
Using the mains balance pot I set the spot to the center line green.
So now, when I use the tester I insert the mains card and if the spot is not on the line I can adjust the fine switch to set it, I will then know that I have 6.3 v on the valves in the tester.
I inserted the el84 card and get 7.4v on the heater pins, I then inserted the card for ef91 and get 6.6v on the heater pins. I don't think that this is important on a good valve but it could give a false emmission reading on a low emmission valve due to the increased cathode temperature.
What do you think?
John.

[theredhouseinn]

Confirmed.
I had a spare 635 card, so I blocked off one hole and drilled a new one in the location you stated and a good el84 now comes to rest on the green dotted line.
This card is now in my collection marked mod. 1.
John.

[Electronpusher0]

Thanks for the confirmation John! Makes you wonder how no-one has questioned this before.

Re the heater voltages, the transformer tappings are selected to give the correct voltage on load. Did you check the heater volts with a valve inserted. You can measure the voltages on the octal base as all the valve bases are connected in parallel.

Peter

[theredhouseinn]

You are right Peter. With the valve plugged in the heater voltage reduces to 6.5v. That is pretty poor regulation on the heater transformer. Perhaps that is why the instructions say make the test as short as pos.. Don't hold the buttons down, etc..

John.

[Electronpusher0]

I am not surprised as the heater transformer has tappings on the primary as well as the secondary to select the voltage.
Peter

[GMB]

Perhaps people can check their 635 cards to see if Mullard themselves had issued a revised card. I have not seen one but lots of other cards got revised, sometimes more than once!

[PJL]

Any result in the green section is good >43mA
In the red is bad <29mA

As far as I can make out the datasheet would come out at about 52mA. So these thresholds seem reasonable.

The dotted line is for the MAINS and HT checks only.
If you don't have a copy of the operating manual it is online somewhere.

[PJL]

Originally Posted by theredhouseinn

You are right Peter. With the valve plugged in the heater voltage reduces to 6.5v. That is pretty poor regulation on the heater transformer. Perhaps that is why the instructions say make the test as short as pos.. Don't hold the buttons down, etc..

John.

Nope, it is because it is a true static DC test at typical maximum working conditions and if your valve has better than average gain it will run above the maximum anode dissipation. Look back at my post with the calculator results and I have based the 'good' results on the dotted line (assuming that a NOS valve would run well into the green band) and dissipation is very high.

[Electronpusher0]

Originally Posted by PJL

Any result in the green section is good >43mA
In the red is bad <29mA

As far as I can make out the datasheet would come out at about 52mA. So these thresholds seem reasonable.

The dotted line is for the MAINS and HT checks only.
If you don't have a copy of the operating manual it is online somewhere.

I still could not get ANY EL84s, including brand new ones to get out of the Amber. My tester is now fully restored (I have all the manuals) and passes the Reject and HT cards.
I agree that around 50mA is the value from the data sheet but the settings on the original card would mean that good valves would be only just on the green amber threshold. Great for selling customers new valves but not realistic.

Peter

[PJL]

Originally Posted by Electronpusher0

Great for selling customers new valves but not realistic.
Peter

And you must buy a Mullard of course!

I am surprised NOS ones fail to make it into the green. If I get a chance I will try to find a few and test on mine.

[Electronpusher0]

That would be great, it would be good to have another machine check the card.
Peter