heater voltage

[Bengali]

is there a general consensus of using lower heater voltage?

ex. using 6.1V for 6.3V heater. I read some where using lower voltage, the tubes sound better, last longer , not sure if this is tube specific (6sn7, 12au7,etc..)

some other site show driving the heater slightly higher give better performance at cost of tube life.

[DF96]

I think the general consensus is to use the rated voltage, except where reducing it a little can reduce noise a little. A healthy cathode has plenty of spare emission, so increasing heater power will simply shorten life.

[Wavebourn]

Here is an interesting picture, from a great book.

Curves for 6P15P tube, dependence on filament voltage of:

___ anode current
-.-.- steepness
- - - screen grid current

I assume 6.0V stabilized is quite acceptable.

[ChrisA]

Originally Posted by Bengali

is there a general consensus of using lower heater voltage?

ex. using 6.1V for 6.3V heater. I read some where using lower voltage, the tubes sound better, last longer , not sure if this is tube specific (6sn7, 12au7,etc..)

some other site show driving the heater slightly higher give better performance at cost of tube life.

The tubes are spec'd to perform within 10% of 6.3 volts. That would mean the range of 5.67 to 6.93 is within the manufacturer's spec's. But I've done some experiments with a variable voltage bench supply and looked at the audio waveform on a scope. I can typically go down to about 5V before I could see a difference.

Also tube life is strongly related the heater voltage. I've been using a resistor in series with the heaters to lower the voltage to 6.0V

[wa2ise]

Most tubes you want to run at their rated heater voltage, though some small signal level tubes can be more linear at around 85% their rated heater voltage.
Like the 12AV6 and 12AX7, and detector diode tubes like 6AL5.

[DF96]

Interesting plots. One thing to watch is that a cooler cathode may be quieter at low currents, but if operation starts to stray into the temperature-limited region then space charge smoothing of shot noise no longer operates so noise gets worse.

[Captn Dave]

The Fisher, in their 7591A designs, used only 10 volts on their 12AX7s. They wired the dual triode heaters from the phono stage and line stage in series (four of them) and used them as a cathode resistor for the four power tubes. By doing so they got DC heater voltage for those stages. Significantly, they use a 500R resistor in parallel with the heaters for another 80ma more cathode current. (They sure ran them hard)

The bias is set correctly when that voltage is 39 or 40 depending on the model. That, of course, works out to about 10 volts each for those 12.6 volt heaters. They draw about 130ma at that voltage.

I think it's safe to say they specifically desired lower voltage because they could have simply specified 50.4 volts and increased the resistance of the parallel resistor to draw less current. They might have been going for the improved linearity to which wa2ise referred.

I know of at least one person who went so far as to change a resistor in a modern amp to reduce heater voltage (a lot) because she better preferred the sound quality.

[rman]

Hi.

I have an amp using four 6EM7 dissimilar triodes. They where purchased at a local store new old stock. They are heated with just under 6 volts with a series resistor. After less than two years, I switched on the amp to find one Chanel was dead. The heater on one of the hi mu sides was not glowing.

So it seems running lower heater voltage does not guarantee long life. Inrush current still takes it's toll.

Cheers.
Rolf.

[payitforwardeddie]

Eating a healthy diet, exercising well, abstaining from drink and drugs does not guarantee a long life.