Old PT's and High Filament Voltage

[boots]

This has probably been talked about before, but I'm hoping you guys can give me some advice (or direct me to a thread?).

Being the old parts scrounger that I am, I occasionally come up with some old iron I'd like to try using. The old PT's were designed to work with lower primary wall voltages, so I am ending up with loaded filament voltages on the order of 7 VAC. This can't be good for tubes. I've always understood that proper filament voltage is critical for the long life of a tube.

Is there an easy way to make these old PT's work with today's higher wall voltage? Is there a suitable rheostat or something that could go in series with the PT primary to bring the secondary voltages down to the proper level?

Big radio transmitters sometimes have an onboard variac to tweak the filament voltage, but I am looking for something compact enough to mount inside a standard chassis. It's easy enough to use a variac on the bench, but that doesn't help much when we want to plug the amp into the real world.

Have you guys found a good way to deal with this, or do we all just cook our filaments with too much voltage? What do you do when you are repairing or restoring an older amp that was designed to run on lower wall voltage?

[TUBEDUDE]

I've tried diodes and low R power resistors for dropping voltage. The resistors worked, but made me nervous as they get hot inside. The diodes were a little noisy, and as i had 7 diodes in each side, I didn't want to put a cap across them and I was out of real estate. If I thought I could pull it off, I'd remove a few windings from the secondary and try to drop the voltage.

[Invertiguy]

I would try adding a bucking transformer before the power transformer. A 12.6V filament transformer wired in such a fashion would give an input voltage to the PT of roughly 110VAC, which should suit old gear just fine.

For info on bucking transformers: http://sound.westhost.com/articles/buck-xfmr.htm

[Zippy]

If you're going to add parts, you might just add a filament transformer. A bucking transformer for the entire power load would have to carry much more current.

[Phil_S]

You can insert a wirewound resistor of the appropriate wattage rating between the PT and the first connection. Typically, it will be a very small value like 1/2 Ohm.

Let's say your filaments add up to 3.0A @ 6.3V. You are seeing 7.5V. You need to drop 1.2V. R=V/I. 1.2V/3.0A = 0.4 Ohms. Try it with a .50 or .51 Ohm 5W resistor. See if that gets you in the relevant range. This is the simplest solution. The resistor gets hot. Make sure it has good free air or better, put on the chassis as a sort of heat sink.

[matt h]

(deleted)

[ampfab]

I recently solved a similar problem by using two .25ohm 5 watt resistors. I installed each resistor on each filament output wire, in series with the filiment circuit. Dropped it right into the correct voltage. I was powering a pair of 5881s and three 12ax7s. The resistors don't get hot.

[TUBEDUDE]

Where as one .47 ohm @ 5 watts gets quite hot!

[Phil_S]

Where as one .47 ohm @ 5 watts gets quite hot!

Aha! That's really interesting. I wouldn't have expected that to be true. If it is so, then a 10W would probably be a better choice if heat is a concern. I did this in the last amp I built with a .51 (what I had on-hand) and it worked like a charm. I didn't take it's temp (need one of those infrared gizmos) but assumed it was hot.

[Structo]

I seem to recall using a zener diode, across the winding?

Anybody tried that method?

[TUBEDUDE]

Wouldnt a zener across an A.C. winding clamp peaks and create noise as it dumps current?

[Structo]

Another way is to use a zener on the center tap of the heater coil. (or artificial CT)

I did this to drop B+ on an old amp.

But I am not sure if there is a zener rated low enough voltage for this purpose.

If there are, then get a stud mount zener.
If noisy, hang a .01uF cap across it to ground.

[David Root]

A 5W sandbox does run fairly warm, but OTOH they will fit where a 10W sandbox might not. Or use one 5W on each side, again if they'll fit.

I seem to find the area around the power tube sockets is pretty busy and around the noval sockets is even worse.

[jeeptechfred]

Another way is to use a zener on the center tap of the heater coil. (or artificial CT)

I did this to drop B+ on an old amp.

But I am not sure if there is a zener rated low enough voltage for this purpose.

If there are, then get a stud mount zener.
If noisy, hang a .01uF cap across it to ground.

Correct me if I'm wrong. (-:
This works on a typical B plus because it is d/c and the diode sets that amount of voltage off of the chassis. it would not work to reduce the voltage to the heaters because typically each outer leg of a 6 volt heater winding is connected directly to the heater pins. A zener is a diode and the heater circuit is typically a/c. I'm not sure how that would play out, I don't believe any thing would be accomplished. maybe some added noise?

[SoulFetish]

A different approach would be to add silicon diodes in series with the secondary heater taps. This would give you the flexiblity to determine the voltage drop by adding/subtracting diodes in series or increasing the current handling capacity by placing additional diodes in parallel. Here is a simple schematic illustrating the circuit.

*edited schematic to show correction - explanation in following posts*