A fascinating view of Mullard valve manufacturing - UK Vintage Radio Repair and Restoration Discussion Forum

steviewonder · 18th May 2014, 10:51 pm

just found this on youtube.
The Mullard Blackburn factory...Fascinating viewing on valve production.
Not sure if the posting on this site is allowed,but here goes:
http://youtu.be/GDvF89Bh27Y.
I'm somewhat amazed at the process.
Happy viewing,
Steve

Bazz4CQJ · 19th May 2014, 12:47 am

Thanks for that; I think I may have seen it, or a similar film, back when I was at school in the 60's.

I noticed that the commentator only referred to the use a backing pump for evacuation, though that was followed by firing the getter. Does anyone know what sort of pressure was typical in a valve? IIRC, a backing pump alone will typically give you between 10E-2 and 10E-3 Torr.

B

Radio Wrangler · 19th May 2014, 8:24 am

It becomes a matter of holding the valve at backing-pump pressure while promoting outgassing with some heat before sealing. The aim being to fire the getter and for the getter to have enough capacity to absorb the remaining gas and any future outgassing/leakage over a reasonable life.

High voltage transmitting devices couldn't afford getter metal fired around inside, so they had no getter and were taken the rest of the way down on a diffusion pump before sealing. Outgassing time and heating was even more important.

David

woodchips · 20th May 2014, 11:50 am

If of any interest I have gleaned the following from Eastman's Fundamentals of Vacuum Tubes.

Vacuum level is determined by the chances of an electron hitting a gas atom. So the electron has to travel 1cm without encountering a gas atom. An electron is 4e13cm diameter, a gas atom is about 1E-8cm. If the valve is evacuated to 7E-5mm, or Torr, leaving about 1E12 atoms per cubic cm. The spacing between atoms is about 1E-4cm, or nearly a billion times the diameter of an electron. A collision is very remote.

Seems that a backing pump is good enough.

kalee20 · 20th May 2014, 12:44 pm

That's a good analysis, though it needs to factor in the number of electrons in transit, and the number of possible collisions that can be tolerated before ionisation troubles become a nuisance.

Quote:

Originally Posted by woodchips

An electron is 4e13cm diameter

Hopefully 4e-13cm diameter, or I'll have to use MUCH bigger wires on my next project to let the little blighters through...

woodchips · 21st May 2014, 9:31 am

Ha, kalee, you are too fussy about the sign of the exponent! It is just a mere fraction of a bankers' bonus so must be irrelevant.

One other thing I turned up was the Paschen voltage curve. Not really sure but this seems to say that you could have too good a vacuum in a valve. The arc over voltage between electrodes varies as the vacuum pressure varies, and will reduce as the vacuum improves. Is this one reason why modern valves can't cope with as high a working voltage as their 50 year old forbears? Modern manufactures use diffusion pumps, not backing pumps. Any ideas?

It has always seemed strange to me that modern power valves fail at high voltages, I had assumed the cathode coating just wasn't as good and shed bits and pieces causing the arcing. The Paschen curve might answer?

kalee20 · 21st May 2014, 12:51 pm

Well, Paschen's law says that the lower the pressure inside, the lower the breakdown voltage. It was well demonstrated with conventional petrol ignition systems - when the points got dirty etc, it was reasonably common to be able to remove a sparkplug and get a spark at its electrodes outside the cylinder, but replace it and under compression pressure it would not spark due to insufficient voltage.

If there are any free ions in a gas, under an applied electric field they will accelerate towards an electode. Sooner or later the ion will encounter another gas molecule - if on average it's going fast enough to knock an electron out of that, then there will now be more ions to start moving. And you get breakdown.

If the gas pressure is high, there are lots of gas molecules so the chances are that a free ion will encounter one before it is going fast enough to ionise it. So you don't get breakdown unless you raise the voltage (giving more acceleration in the distance available). But if the pressure is low, gas molecules are few and far between, so you needless voltage for breakdown.

It falls apart for really low pressures as if the free ion or electron is trapped by an electrode before it encounters a gas molecule, you won't get breakdown then. So if the pressure is such that the mean free path is greater than the anode-cathode distance, Paschen's law becomes increasingly inapplicable. Basically,there's a pressure such that the breakdown voltage is a minimum - more gas or less gas results in a higher breakdown.

I rather suspect modern power valves might be suffering from not enough vacuum, rather than too much!

lesmw0sec · 21st May 2014, 6:16 pm

Having enjoyed this, I then turned to watch a film by a chap making homebrew copies of the type R triode. I noticed that his test set had a meter calibrated in gm. I was under the impression that gm should be measured as delta IA/VG, but this chap seemed to be measuring it purely on the basis of a steady anode current. Am I right, or have I forgotten my student days?

18th May 2014, 10:51 pm	#1
steviewonder Tetrode Join Date: Apr 2012 Location: Manchester, UK Posts: 92	A fascinating view of Mullard valve manufacturing just found this on youtube. The Mullard Blackburn factory...Fascinating viewing on valve production. Not sure if the posting on this site is allowed,but here goes: http://youtu.be/GDvF89Bh27Y. I'm somewhat amazed at the process. Happy viewing, Steve __________________ Due to recession, the light at the end of the tunnel has been switched off...

19th May 2014, 12:47 am	#2
Bazz4CQJ Dekatron Join Date: Apr 2009 Location: Oxfordshire, UK. Posts: 4,935	Re: A fascinating view of mullard valve manufacturing Thanks for that; I think I may have seen it, or a similar film, back when I was at school in the 60's. I noticed that the commentator only referred to the use a backing pump for evacuation, though that was followed by firing the getter. Does anyone know what sort of pressure was typical in a valve? IIRC, a backing pump alone will typically give you between 10E-2 and 10E-3 Torr. B

19th May 2014, 8:24 am	#3
Radio Wrangler Moderator Join Date: Mar 2012 Location: Fife, Scotland, UK. Posts: 22,901	Re: A fascinating view of mullard valve manufacturing It becomes a matter of holding the valve at backing-pump pressure while promoting outgassing with some heat before sealing. The aim being to fire the getter and for the getter to have enough capacity to absorb the remaining gas and any future outgassing/leakage over a reasonable life. High voltage transmitting devices couldn't afford getter metal fired around inside, so they had no getter and were taken the rest of the way down on a diffusion pump before sealing. Outgassing time and heating was even more important. David __________________ Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done

20th May 2014, 11:50 am	#4
woodchips Octode Join Date: May 2010 Location: Grantham, Lincolnshire, UK. Posts: 1,177	Re: A fascinating view of Mullard valve manufacturing If of any interest I have gleaned the following from Eastman's Fundamentals of Vacuum Tubes. Vacuum level is determined by the chances of an electron hitting a gas atom. So the electron has to travel 1cm without encountering a gas atom. An electron is 4e13cm diameter, a gas atom is about 1E-8cm. If the valve is evacuated to 7E-5mm, or Torr, leaving about 1E12 atoms per cubic cm. The spacing between atoms is about 1E-4cm, or nearly a billion times the diameter of an electron. A collision is very remote. Seems that a backing pump is good enough.

21st May 2014, 9:31 am	#6
woodchips Octode Join Date: May 2010 Location: Grantham, Lincolnshire, UK. Posts: 1,177	Re: A fascinating view of Mullard valve manufacturing Ha, kalee, you are too fussy about the sign of the exponent! It is just a mere fraction of a bankers' bonus so must be irrelevant. One other thing I turned up was the Paschen voltage curve. Not really sure but this seems to say that you could have too good a vacuum in a valve. The arc over voltage between electrodes varies as the vacuum pressure varies, and will reduce as the vacuum improves. Is this one reason why modern valves can't cope with as high a working voltage as their 50 year old forbears? Modern manufactures use diffusion pumps, not backing pumps. Any ideas? It has always seemed strange to me that modern power valves fail at high voltages, I had assumed the cathode coating just wasn't as good and shed bits and pieces causing the arcing. The Paschen curve might answer?

21st May 2014, 12:51 pm	#7
kalee20 Dekatron Join Date: Feb 2007 Location: Lynton, N. Devon, UK. Posts: 7,087	Re: A fascinating view of Mullard valve manufacturing Well, Paschen's law says that the lower the pressure inside, the lower the breakdown voltage. It was well demonstrated with conventional petrol ignition systems - when the points got dirty etc, it was reasonably common to be able to remove a sparkplug and get a spark at its electrodes outside the cylinder, but replace it and under compression pressure it would not spark due to insufficient voltage. If there are any free ions in a gas, under an applied electric field they will accelerate towards an electode. Sooner or later the ion will encounter another gas molecule - if on average it's going fast enough to knock an electron out of that, then there will now be more ions to start moving. And you get breakdown. If the gas pressure is high, there are lots of gas molecules so the chances are that a free ion will encounter one before it is going fast enough to ionise it. So you don't get breakdown unless you raise the voltage (giving more acceleration in the distance available). But if the pressure is low, gas molecules are few and far between, so you needless voltage for breakdown. It falls apart for really low pressures as if the free ion or electron is trapped by an electrode before it encounters a gas molecule, you won't get breakdown then. So if the pressure is such that the mean free path is greater than the anode-cathode distance, Paschen's law becomes increasingly inapplicable. Basically,there's a pressure such that the breakdown voltage is a minimum - more gas or less gas results in a higher breakdown. I rather suspect modern power valves might be suffering from not enough vacuum, rather than too much!

21st May 2014, 6:16 pm	#8
lesmw0sec Octode Join Date: Jul 2009 Location: Carmel, Llannerchymedd, Anglesey, UK. Posts: 1,509	Re: A fascinating view of Mullard valve manufacturing Having enjoyed this, I then turned to watch a film by a chap making homebrew copies of the type R triode. I noticed that his test set had a meter calibrated in gm. I was under the impression that gm should be measured as delta IA/VG, but this chap seemed to be measuring it purely on the basis of a steady anode current. Am I right, or have I forgotten my student days?