Classic Warranties

[Radio Wrangler]

I've just had one of those odd thoughts, I keep having them, but I've tried to stop.

Way back in the day, all the setmakers offered a 1-year guarantee on their produce, excepting the valves. They got a 90 day warranty.

And nowadays we know it was the capacitors which should have been fitted in sockets.

Ah, hindsight.

David

[Synchrodyne]

So the plug-in capacitors would have had a 90-day warranty from their respective manufacturers. And Hunts and the others would have equipped radio stores with capacitor testers alongside their Mullard valve testers, and would have exhorted customers to have their radio and TV set capacitors tested regularly....

Cheers,

[mark pirate]

Back in the day it was really common to blame valves for a breakdown, but it was more likely to be a failed capacitor.

I remember being given a 405 line Sobell portable when I was around 10 years old, it had frame lock problems that I could not solve by changing valves.
On asking a TV engineer, he explained that it was down to capacitor failure, and gave me a new one to try, as you can guess this fixed the problem and taught me a valuable lesson!

Mark

[turretslug]

It would be interesting to see how much scrutiny was directed towards capacitors by the folk who made the first satellites (and other situations where warranty concerns are, at best, tricky- say undersea repeaters) in (?) the early 'sixties when capacitors were still a little mediochre generally. Or would the absence of water vapour hundreds of miles up make it less of a head-ache? Customers with big wallets, like the 'fifties US military, seemed to be able to call upon capacitors with aluminium casing and epoxy or glass seals, though I suspect the innards might still have been fairly traditional paper-in-oil construction.

Oh dear, a curse upon thee, RW, for setting me thinking...

[Maarten]

Paper in oil was quite reliable as far as reliability of a paper capacitor goes. I'd like to think satellites and under sea repeaters avoided them anyway. By the early 1960's Philips had started using their polyester capacitors in consumer goods, but they might not have passed certification for NASA and the military yet. Would be nice to hear from someone 'in the know'.

[Nickthedentist]

I know this has been discussed here before, but I'm still amazed by the pretty universal 90 day valve warranty, when so many valves are still going strong 50/60/70/80 years later.

[G6Tanuki]

I think the "Capacitor failure" issues we all go on about are often rather unfair: let's face it - your typical consumer-grade TV or radio built in the 1950s/1960s was only expected to have a service-life of a few years. Would purchasers have been prepared to pay even more towards an already-only-just-affordable TV guaranteed to last 20, 30, 40 years ?

I rather doubt it.

Even 'professional' gear isn't immune to capacitor failures [I've heard it said that the R1155 HF receiver wasn't designed with component-level serviceability a consideration because Lancasters didn't survive long enough for the radios to ever need servicing]

So - we got drifty carbon-stick resistors, waxies and Hunts plastic-encapsulated capacitors, of which 99% were probably still working adequately when the TV/radio they were in went to the dump.

And at least people got TVs/radios they could afford to buy.

[G6Tanuki]

Originally Posted by Maarten

Paper in oil was quite reliable as far as reliability of a paper capacitor goes. I'd like to think sattelites and under sea repeaters avoided them anyway. By the early 1960's Philips had started using their polyester capacitors in consumer goods, but they might not have passed certification for NASA and the military yet. Would be nice to hear from someone 'in the know'.

The sealed-metal-can Tantalum capacitors widely used in military/avionics gear [where you'd expect reliability to be a component-selection consideration] have service-life issues - they 'only' last 30 years or so. But how long was a Centurion tank or a Vulcan bomber expected to survive if there'd actually been a nuclear war?

[vidjoman]

Back in the 60's 'domestic' capacitors were built at a competitive price so the set makers could sell finished sets. I have a friend that worked for a capacitor maker on non domestic items and they were testing for a 50 year lifespan for the telecoms industry.
I regularly work on Juke Box amplifiers from the 50's & 60's and some makes of capacitor are still working correctly. Other makers products have already been replaced. Same with resistors - some are still good even after 50 years. Modern resistors do have the advantage of lower noise and closer tolerance.

[HamishBoxer]

Re reses and caps ,iIIIIIII think the life of them depends on what part of the circuit they are in.

[Radio Wrangler]

Hunts did make capacitor testers....

But in all fairness, most of these things proved reasonably reliable for several years, and we're looking backwards at them after several decades. I'm just amused that the warranties turned out to be the wrong way round.

Good point on the metal cased tants. I have a very nice spectrum analyser needing a load to be swapped, and then seeing seventy quid each for something that used to be cents.... Aargh!

David

[turretslug]

True, much of our stuff has now exceeded the expected consumer product cycle by an order of magnitude. Perhaps those famous "Mullard mustards" really were one of the first inexpensive consumer capacitors to last and last.

It's been chewed over endlessly but I think there might be an element of truth in the supposition that whilst the 'fifties/'sixties serviceman was changing the guilty sixpenny waxie, there was always the temptation to change a twenty shilling valve at the same time- it took seconds and bumped up the turnover, leaving the man in the street with the impression that valves really were as consumeable as light-bulbs.

[wireful3]

I wonder whether the unreliability of valves is a result of long memories in the late 30s. I saw a few years ago an article about reproduction bright emitters. The implication was that filament emission degraded fairly quickly. In fact the makers of these reproductions warned they would not last long.

In the early days, valves were very expensive and if they only lasted a few hundred hours it would leave a very big impression on the owners. Presumably the filaments of battery valves in the early 30s could have been somewhat delicate compared with later valves. This would no doubt help maintain this feeling that valves were less reliable than other components. One thing I am not too sure about is whether capacitors in radios that have been kept in operation and in benign surroundings show the problems we notice.

I too have heard the story about the 1155 and serviceability. It seems credible, possibly size and weight were more important than long field service life.

[PJL]

John, you have a good point regarding bright emitter valves and this may well have led to the industry taking advantage of the acceptance that valves were short lived. The next generation of B4/B5/B7 valves from the late 20's to the mid/late 30's used in radio seem to last forever. However, normality was restored with the introduction of the miniature B7/B9 types.

[TrevorG3VLF]

Did early valves have a high failure rate as did TV tubes?
We are left with magnificent cathedrals, the grotty ones all fell down.

[Studio263]

Even B&O (in the UK at least) limited their valve warranty to 90 days. The last thing they made with valves in was the Beovision 3400 but for years after that was dropped they were still issuing the old cards with the exception for valves made clear - they must have printed far too many!

Sony guaranteed their semiconductors for 5 years in the 1970s, this may have been to pacify the dealers over crazy ideas like GCSs and VFETs (both unreliable things as it turns out). What the guarantee on the valve in the KV-1320UB was I don't know, I've never seen a duff one so I suppose in the end it doesn't matter...

[Nickthedentist]

Originally Posted by Studio263

...they were still issuing the old cards with the exception for valves made clear...

Interesting point. I have seen this on other, (non-B&O) solid-state gear from the late 1970s too.

Originally Posted by Studio263

The last thing they made with valves in was the Beovision 3400

Didn't the later ones have a CRT, a huge valve!

[mark pirate]

A CRT certainly could be classed as a big valve, but even when TV was in it's infancy they usually came with at least a one year warranty.
In the case of some Mazda CRT's, a 90 day warranty would have been more appropriate!

Mark

[boxdoctor]

Many '50s TVs had an often under-rated carbon composition resistor for the screen feed to the line output valve. These would go low resistance, causing failure of the line output valve (often a PL81) due to excessive screen current. A new valve would 'cure' the fault for a time. Some of the field engineers of the day had graduated to TV servicing from being plumbers or cycle mechanics, which were the primary business of the TV shop before branching out into TV's. The valve would simply be replaced again, giving the impression that valves were unreliable. A further failure would inevitably follow. At this point the customer would often consult another engineer, and with a bit of luck would get the required resistor replacement, and yet another valve, and the problem solved.
The impression that valves were unreliable would be reinforced by this chain of events. Also this was not good for the reputation of the PL81's reliability! Penny-pinching design was the real problem.