Filament voltages - an idle curiosity

[kruesi]

Does anyone here know why we settled on 6.3 V of all things? -Why not 6.0, or 3.1416 V? Some of the other higher "unusual" voltages I can almost understand, being more relating to current than voltage for series strung filament applications ("All American 5" design, Muntz TV receivers etc). But the unusual, seemingly random 6.3V seems an odd choice.

There has just got to be a good story behind this one...?

[johnr66]

I'm guessing, but I think it has to do with battery voltages. Some portable tube radios and other equipment had two or more batteries. One battery was the cathode heater, called the "A" battery. Another battery, the "B" battery was a high voltage battery for the plates. It made sense to make the tubes have the same filament voltage rather than make another set of tubes for mains power (aside from the transformerless sets).

Many receivers/amplifiers had an isolated 5volt winding for the rectifier tube. Not sure of the "why" on this voltage selection.

[HollowState]

Just a guess but it may have something to do with running tube filaments from a standard car battery. The early car batteries were nominally rated at 6 volts with the actual voltage being very slightly higher.

[kruesi]

Hmmm... you may be on to something there- Lead-acid cells are nominally 2.1V when charged- could be as high as 2.2. So this is a possibility. Still the "C" battery back in the day was, I believe, Zinc-Carbon. Or maybe the use of lead-acid predates this...?

[Tom Bavis]

5.0V heaters were used with early battery tube, using 6.3V batteries and a rheostat. This allowed a little more life per charge, since the rheostat could be adjusted as the battery voltage dropped... RCA later standardized on multiples of 2.5V for AC tubes: 2.5, 5, 7.5, 10V - they may have been thinking about using an integer number of turns on transformers, and all voltages could be EXACT. A radio often had three transformer windings - 2.5V for small-signal tubes, 2.5V or 7.5V for output tube(s) and 5.0V for the rectifier. The higher power tubes like 10, 50 were 7.5V, no matter since they had their own heater winding... but the small-signal tubes could still add up to 10A or more! Philco had other ideas - they built car radios using 6.3V tubes and decided to use the same ones in their AC powered radios. They bought from Sylvania, not the enemy (RCA), and Sylvania was happy to provide whatever they wanted. Fewer parts to stock and all that... guess whose idea worked out?

[tubelab.com]

Early radios that used Directly Heated tubes did have a rheostat (pot with only two terminals) in series with the filaments. There were three reasons for this. Some times a radio would use a "wet cell" (rechargeable lead acid) for filament power, and other times the same radio could be powered by a "dry cell" (primary carbon zinc) in areas where there was no line voltage to recharge the batteries. The rheostat was used to adjust for battery charge, battery type, and even radio volume. The early DHT's would handle this.

Hmmm... you may be on to something there- Lead-acid cells are nominally 2.1V when charged-

By the time indirectly heated tubes were becoming popular most people had access to electricity to recharge batteries, and radios found their way into automobiles. Although there were some oddballs just about every automobile produced from the late 20's to the early 50's used a 3 cell lead acid battery, which produces 6.3 volts nominally. So if you wanted to pick a voltage that would make everybody happy........

[Iain McNeill]

Great thread kruesi.

Knowing how early electronics was controlled by practical concerns, I would expect that there were requirements for heater power. Bigger tubes needed more cathode heating - more heating needed more amps, more amps caused more losses. It's a power transfer thing?

Any ex-Sylvania employees out there?

[kruesi]

It's really sounding like the 6.3V choice was from lead-acid cell voltages- dunno why this never occurred to me. Although many pre-war radio circuits I've looked at are powered from A, B and C cells, I guess I'm too used to powering everything from the line.

In my post above I mention the C cell battery pack... I of course meant the "A" cell- we're talking filament voltage here, not grid bias(!) -My oversight.

Tom- I had no idea all the stuff about 2.5, 5, 7.5 ... ever existed! I do recall old transformers having a number of different filament windings, but never put two and two together. Thanks!

And tubelab- you've cleared up the mystery of that huge rheostat- I've seen TRF sets using type 12A directly heated triodes with a large rheostat in series with the filament supply- and here I thought it was just a particularly strange way of doing the volume control. Makes sense to be able to adjust for whatever supply one had available in the days before the completely-standard power grid we have today.

[kruesi]

Bad form for me to post two replies in a row- but Iain brings up a valid point, one that Tom touched on- I'd expect higher-power tubes (ones that require more emission) to perhaps have higher filament voltages than small-signal stuff- it only makes sense.

Indeed, any rememberances from ex-Sylvania (or indeed ANY tube manufacturer employees) would be gladly welcomed!

[SoNic_real_one]

6.3V comes from the acid-lead days. 3 elements will provide exactly 6.3V nominal voltage.
It is also a series of TV tubes that work with constant curent (eg 300mA), all the heaters connected series directly to the power main. The bigger tubes (higher power for the heater) had a bigger voltage drop on the heater, smaller ones had a smaller voltage drop. The rest was disipated with power resistors. In this way the TV set didn't need a transformer, just a rectifier and thus was able to work on DC and AC power networks (was called "universal"). And was way lighter.

In european system, the tubes with 6.3V heaters are having the first letter E, the ones with 300mA have P (eg. ECL versus PCL). Some of the smaller 6.3V tubes where manufactured to draw a 300mA curent, so they could work on any place (I saw ECC83 series with P type of tubes). Type P was very popular in TV sets working directly at 220Vac.
First letter in tube numbering in european system:
http://www.duncanamps.com/technical/tubenumber.html