Test your Valves Regularly

[emeritus]

A 1000 hour soak test used to be part of the MOD-specified production test procedure for military equipment when I was at Plessey. This simply involved leaving the equipment powered-up for 1000 hours and then testing it to its full specification. The object was to flush out any "infant mortality" failures.

[julie_m]

Originally Posted by G8HQP Dave

These days everybody assumes it is a capacitor, so change the lot.

To be fair, though, it often is a capacitor. Underneath the faults they introduced with wiring errors, while replacing many capacitors in one go .....

[MurphyNut]

I wonder if later valves don't last as well. I have many pre-war sets still going strong with what looks like their original valves.

[Phil G4SPZ]

I have worked on dozens of pre-war radios over many years, and only ever had to replace a handful of valves. I've also tested nearly two hundred NOS pre-war valves and they were virtually all good.

Valves (other than rectifiers) in pre-war radio sets were operated somewhat more 'gently', electrically speaking, than valves in later TV and industrial applications, which may account for some of the apparent difference in longevity.

I agree that a dealer recommending regular valve testing was something of a sales ploy. Many sets will often continue to operate prefectly well even with valves that are below par!

[Craig Sawyers]

Tektronix used to burn in valves in massive racks for the purpose, both the weed out infant mortalities and to stabilise characteristics. Characteristics were then measured, and selected parts from the batch used for specific applications (such as low microphony for sensitive plugins like the Type D).

[Heatercathodeshort]

Originally Posted by HamishBoxer

As Gerry Wells used to state, you took your radio in for repair to the cycle shop or where ever and you said "I think it is a valve" and it always was!

The customer is always right! J.

[alanmoore]

I so don't think so!

[Hartley118]

AFAIK, the best predictor of the future lifespan of a valve under 'normal' usage is the length of time it's already been in use. For example, the valves in the original 1955 TAT1 transatlantic telephone cable undersea repeaters had already been running for 18 years before being installed. Here's a quote from an interesting 2006 IEEE article by Jeremiah Hayes, which also emphasizes the perceived importance of benign operating conditions:

"The electrical challenge was reliability; repeaters were expected to lie on the ocean bottom for twenty years. The cost associated with the repair due to component failure was prohibitive. All of the electrical components were subject to rigid reliability requirements; however, most fragile were the vacuum tubes, which were the only means of amplification. Development of these super reliable tubes was initiated in 1932. They were life tested for a period of eighteen years. When installed they were significantly below the current state of the art, a mutual conductance of 1000 vs. 6000 micro mhos. They were manufactured under conditions that presaged those of modern semiconductor fabrication.

The repeater design features also contributed to reliability. In order to increase tube life, the signal levels into each stage of the amplifier were a level lower than that causing grid current to flow..."

The complete article is at http://www.ewh.ieee.org/reg/7/canrev/cr52/CR52_TAT.pdf

Martin