Hopefully a simple question.

[a95sonoma]

This a hopefully a simple question. What damage, if any, does leaving a heater always on, on a tube such as a 12ax7, with B+ only applied when in use, do to the tube? The voltage for the heater would be DC.

I'm not thinking of using some rare Telefunken just the cheap everyday run of the mill Chinese 12ax7b. The heater supply would be a external 12volt switching supply with a old but very lightly used 6800uf computer filter capacitor, bigger than most output tubes.

This is a mod so it would be easier to keep it external. I've already done the mod and the circuit is now dead silent. The problem is wiring in the heater power supply so it turns off and on with the amp, would be be a slight pain.

Thanks James,

 

[Vinylsavor]

Hi!

Not recommended. The cathode can develop an interface layer which impacts the performance of the tube

Thomas

 

[20to20]

This a hopefully a simple question. What damage, if any, does leaving a heater always on, on a tube such as a 12ax7, with B+ only applied when in use, do to the tube? The voltage for the heater would be DC.

I'm not thinking of using some rare Telefunken just the cheap everyday run of the mill Chinese 12ax7b. The heater supply would be a external 12volt switching supply with a old but very lightly used 6800uf computer filter capacitor, bigger than most output tubes.

This is a mod so it would be easier to keep it external. I've already done the mod and the circuit is now dead silent. The problem is wiring in the heater power supply so it turns off and on with the amp, would be be a slight pain.

Thanks James,

In the last days of tube TV's the manufacturers sold "Instant On" sets that kept the filaments warm while the set was off. Not sure if the filament voltage was full or maybe 75% but it actually had the affect of reducing tube failure due to hot/cold cycles of the filament.

 

[Kay Pirinha]

In the last days of tube TV's the manufacturers sold "Instant On" sets that kept the filaments warm while the set was off. Not sure if the filament voltage was full or maybe 75% but it actually had the affect of reducing tube failure due to hot/cold cycles of the filament.

Very often in the standby mode a diode was inserted, thus reducing the power consumption of the heater chain to somewhat more than 50 %.

Greez!

 

[DF96]

Yes, full heater power with no cathode current is likely to lead to an interface layer. However, this might take a few years.

 

[Merlinb]

Interface poisoning was largely cured by the 1960s, so it probably isn't a problem.

 

[DF96]

Interface was only cured for computer valves and certain other special quality types. The data sheet normally mentions this for those valves which have the cure. It is unlikely that other valves have the special nickel cathode sleeve material which avoids it. If this were the case, there would be no need for computer grade valves.

 

[Merlinb]

Computer grade valves had double-super-extra-special quality cathodes, but ordinary receiving types still benefited from the nearly-as-good nickel and oxide used in the 1960s.

 

[disco]

Those computer types hardly turn up in audio designs (I know of) .. is the reason obvious?

 

[DF96]

Some computer types were the same as other valves, but with a special cathode which can cope with periods of being hot but with no cathode current.

Others were specially designed, but as the aim was switching they didn't worry too much about linearity.

 

[Kay Pirinha]

Others were specially designed, but as the aim was switching they didn't worry too much about linearity.

Please have a look at the E 90 CC and the E 92 CC curves. In my eyes these computer valves (designed for Windows 58...) don't seem to be too un-linear, do they?

Btw: They are cheap, too, especially if you consider the RFT counterparts ECC 960 and ECC 962, respectively.

Best regards!

 

[DF96]

Almost any triode will not be "too un-linear", but some are more linear than others.

 

[Adolf Corkscrew]

Well, if you had a CCS at the anode, and ran them at about 5mA and 300V B+, the result might actually be quite acceptable. Then again, the same operating conditions with a truly linear triode like the ECC99 from JJ Electronics results in very low distortion.

One thing often mentioned in the datasheets of these computer triodes is that they're prone to microphony. That's a thing to take into consideration when designing audio equipment.

 

[20to20]

Well, if you had a CCS at the anode, and ran them at about 5mA and 300V B+, the result might actually be quite acceptable. Then again, the same operating conditions with a truly linear triode like the ECC99 from JJ Electronics results in very low distortion.

One thing often mentioned in the datasheets of these computer triodes is that they're prone to microphony. That's a thing to take into consideration when designing audio equipment.

Find the RCA 6201 (12AT7 type) data sheet. It has very interesting design notes, for use under cutoff, on/off controls, and high vibration environments. Sounds like the perfect tube.