UK Vintage Radio Repair and Restoration Discussion Forum

UK Vintage Radio Repair and Restoration Discussion Forum (https://www.vintage-radio.net/forum/index.php)
-   Components and Circuits (https://www.vintage-radio.net/forum/forumdisplay.php?f=40)
-   -   Valve Life (https://www.vintage-radio.net/forum/showthread.php?t=91341)

alanworland 11th Dec 2012 9:57 pm

Valve Life
 
As a novice at this radio stuff I have been reading what I can to help the situation. I have been plodding through the book 'Introduction To Valves', a Wireless World publication printed in 1953. This suggests that a valve could have an expected life of 1000 hours!

Now I know they won't just stop working but gradually reduce in efficiency to a point where they are no longer able to do the job for which they were designed, but wow, don't they do well.

I think the valve is a wonderful piece of (over) engineering.

Alan

paulsherwin 11th Dec 2012 10:36 pm

Re: Valve Life
 
Valves can have an extremely long life, depending on their construction and the conditions under which they operate.

The popular idea that valves wear out fairly quickly originates from service practices in the past. Many repairmen would routinely change a valve when some fault was present, as this increased the amount that could be charged for the work.

Sideband 11th Dec 2012 10:49 pm

Re: Valve Life
 
It's been said in the forums before that valves are not the first thing to go for when fault finding.

I (and many other collectors/restorers) have very old sets with some if not all of their original valves. Example 1: Philips 274 Superinductance set from 1934, three of the five valves are original. The two that had been changed many years before I restored it were the first RF amp and the rectifier.
Example two, Philips 470A from 1938. Only the output valve which also contains the detector and AGC diodes has been changed...one of the diodes were faulty.

Other sets in the collection from the 50's and 60's mostly have all their original valves.

One thing you have to remember now is that valves are a finite source. Once they've gone...they've gone. It pays to be economical although there are stocks to last many years yet.


Rich.

Guest 11th Dec 2012 11:07 pm

Re: Valve Life
 
The valves used in the original transatlatic telephone cable repeaters where designed for 100,000 hours life (I think, could be more). I have many valves in radios that still work after 60 to 80 years, some even older. Once they get over a few years of use and are still going they plod on and on and on.

steve_gus 12th Dec 2012 2:22 am

Re: Valve Life
 
In the context that transistors are supposed to last tens or hundreds of thousands of hours (there are a few million in the CPU of this PC im using) then a valve lasting 1,000 hours is quite low.

However, I suspect its a lot longer than 1000 hrs.

there are over 8,000 hours in a year. People likely have their tv on 6 hrs a day, and on that basis a valve tv would have gone faulty in 6 months.

It wouldnt be unusual in a working environment to have a valve radio on 8 hrs a day, so it would likely have a valve fail in 4 months.

They didnt go wrong that often :)

steve

Glowing Bits! 12th Dec 2012 5:43 am

Re: Valve Life
 
I have a GEC set from 1938, there is one valve in this set branded 'Osram', this valve is likely to be original and it still works.
There is also a Pioneer stereo tuner unit from 1967 still using it's original valves, the output valves have very low emission but are still performing, the valves are a mixture of Hitachi & Toshiba.
The stereo has been on for an average of 4 - 5 hours a day for about a year to the best of my knowledge, that is over 1000 hours.
The time I'd expect an EL84 (output valve) to last no more than 1000 hours is when it's used in something like a VOX AC30 gigged rather heavily!

GrimJosef 12th Dec 2012 9:13 am

Re: Valve Life
 
I've seen plenty of Quad II amplifiers with their full complement of original valves from the 1950's and 60's. Sometimes these amps have been left in storage for a decade or two and then switched on and run 'without due care and attention'. Assuming the smoothing block doesn't immediately blow the next fault which can rear its head is gas in the KT66's. This can build up while the amp's in storage to the point where one or other of the 66's goes into thermal runaway, destroying itself (bad) or one of the amp's transformers (worse) or both (tears before bedtime ...). However, had the amp been run from time to time - a couple of times a year might well have been enough - the getter, once hot, would have dealt with the small amount of gas then present without a bad accident occurring. So in this case at least it seems that the valve's life can be longer if it is used once in a while.

Cheers,

GJ

Sideband 12th Dec 2012 9:23 am

Re: Valve Life
 
According to Wikepedia:

'The longest recorded valve life was earned by a Mazda AC/P pentode valve (serial No. 4418) in operation at the BBC's main Northern Ireland transmitter at Lisnagarvey. The valve was in service from 1935 until 1961 and had a recorded life of 232,592 hours. The BBC maintained meticulous records of their valves' lives with periodic returns to their central valve stores.



Rich

mark pirate 12th Dec 2012 10:20 am

Re: Valve Life
 
I have a 1933 Cossor set that still has all it's original Cossor valves, and still working well, so 79 years old!

I was given a 1947 GEC that was used all day, every day in a bakery for over 25 years before being replaced with a transistor set, that also retains all original valves, The internals were covered with a carpet of fluff & flour half an inch thick.

Most of the radio's I have restored still retain some, if not all original valves, however, valve life in TV's is much shorter, as they have to work much harder for a living....
:beer:
Mark

Alan Stepney 12th Dec 2012 10:20 am

Re: Valve Life
 
If I was a betting man, I would give far better odds on a 1930's or 1940's radio with its original valves, still working, than I would with a 1960's transistor radio.

n_r_muir 12th Dec 2012 11:37 am

Re: Valve Life
 
Hi,

Valves can be very reliable provided that they are run well within their ratings and kept on continuously, this was basically one of the secrets of Colossus, the WW2 computer used to crack the German high level codes. Colossus had about 2500 valves. The conventional thinking of the time was that valve failures would prevent the machine from doing any useful work, but the designer, Tommy Flowers a Post Office engineer, new that most failures happen during the uncontrolled conditions that occur briefly as circuits are switched on and off. The answer was to leave the equipment running continuously. Time proved him right.

Ross

alanworland 12th Dec 2012 11:39 am

Re: Valve Life
 
I must admit all the sets I have played with (only back to mid 50's) all have the original valves fitted and working.
So, do I understand the 'getter' is an ongoing working part of the valve?
I thought it was only used initially to create the working environment?

Alan

GrimJosef 12th Dec 2012 11:55 am

Re: Valve Life
 
The getter does both these jobs. It locks up any gas remaining at the end of the manufacturer's pump-down sequence. But it will also sequester gas which seeps out of the metalwork after that. It works much better when it's hot though.

Cheers,

GJ

paulsherwin 12th Dec 2012 12:29 pm

Re: Valve Life
 
Quote:

Originally Posted by steve_gus (Post 575054)
However, I suspect its a lot longer than 1000 hrs.

1950s special quality valves (intended for industrial uses such as computers and lab equipment) generally had a specified life of 10,000 hours. They would often last much longer than this of course, but this allowed operators to change all the valves at specified intervals as preventative maintenance if reliability was essential.

http://tdsl.duncanamps.com/link.php?target=A025977B

Tractorfan 12th Dec 2012 1:24 pm

Re: Valve Life
 
Hi,
I learn something new every day. I didn't realise that the getter carried on mopping up gas throughout the life of the valve. I'd always thought that its job was done & dusted once the valve had been evacuated and the getter was "fired".
Just shows to go!:)
Cheers, Pete:wave:

Boom 12th Dec 2012 9:04 pm

Re: Valve Life
 
I have PM2s here coming up to 100 years old now. They still work well.

Fidelity Fan 3rd Mar 2013 9:50 pm

Re: Valve Life
 
It appears that the higher heater voltage 'U' series valves were more prone to heater filament failure than the low voltage 'E's series.
I can recall three different Fidelity record players using UY85 - UL84 amplifiers with open circuit UL84 heaters. This may be due to the fact that the heater supply was taken from a 90v tapping on the BSR motor and was 7v above what it should have been for the valves.

ricard 4th Mar 2013 9:37 am

Re: Valve Life
 
Quote:

Originally Posted by paulsherwin (Post 575038)
The popular idea that valves wear out fairly quickly originates from service practices in the past. Many repairmen would routinely change a valve when some fault was present, as this increased the amount that could be charged for the work.

Wouldn't the fact that valves are mounted in sockets and hence were intended to be replaced also serve to cement this view? One would tend to assume that the components intending to replaced fairly often would be the ones most easy to replace. I.e. having valves in sockets mean they are intended to be replaced fairly often compared to other components.

Perhaps this was a case of the wearing-out mechanisms of valves being well researched and understood, whereas the longevity of, say, paper capacitors was not as clear at the time?

G6Tanuki 4th Mar 2013 3:59 pm

Re: Valve Life
 
Quote:

Originally Posted by ricard (Post 593301)
Wouldn't the fact that valves are mounted in sockets and hence were intended to be replaced also serve to cement this view? One would tend to assume that the components intending to replaced fairly often would be the ones most easy to replace. I.e. having valves in sockets mean they are intended to be replaced fairly often compared to other components.

Quite a few transistors were also socket-mounted until the early-1960s, or even later in the case of some Soviet gear - I recall a little Russian-made "Vega" portable TV/radio I bought in the early-1980s to use as a DX-TV receiver (it being one of the few ways to get video-coverage of bands I and III as well as audio FM-broadcast reception on the 'eastern european' 68-85MHz band: being able to hear bursts of Radio Gdansk on 70.31MHz by Sp.E was a good indication it was time to get the 2-metre gear fired up).

To my amazement its IF/video/audio stages were almost universally Germanium transistors in 'top-hat' cans reminiscent of the old 1950s Ediswan XA/XB-style, and most of these were fitted in PCB-mounted sockets rather than being hard-soldered to the PCBs.

I wonder if this was perhaps a tacit admission of the poor reliability of Soviet-era consumer semiconductors?

wireful3 4th Mar 2013 4:19 pm

Re: Valve Life
 
I think there was a general caution about that time. They were somewhat an unknown quantity like valves were in the early thirties. I have a Zenith Transoceanic that has plug-in transistors. The Bush portables had the transistors mounted so that they seemed to be almost plug-in.

The tradition that valves were unreliable and might need changing seemed deeply ingrained. I wonder whether valve manufacturers thought the same.


All times are GMT +1. The time now is 12:00 pm.

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