Tube heater question/idea

[KCman]

First, I am putting a bunch of parts I collected together to build something.

I have a PT from a Peavey Classic 30 to use.
I am thinking 2 el84 and 2 12ax7 tubes.

The secondary for the heaters on the classic 30 run ac to 4 el84 in series. (it's also rectified for other low voltage stuff, I don't want to do that)

I was thinking of running the 4 tube fillaments ac, but that would be 6 fillaments to heat. Can I put each 12ax7 in parallel with itself and then run all tubes in series to equal the voltage of the 4 el84s in series?

Seams to me it should work, but any negatives?

I was just thinking of keeping it ac for all heaters to save space. (SS high voltage rectifier, there is no 5v tap)

Thanks
Kirk

[Helmholtz]

No, but..

To run 4xEL84 in series the heater supply voltage will be around 25Vac. Each EL84 and each 12AX7 with parallel filaments wants to see 6.3V. But the supply voltage will only split evenly between the 4 tubes if the hot filament resistances are equal. The EL84 filament corresponds to 8.3 Ohm but the parallel 12AX7 filament to 21 Ohm. So each 12AX7 will get a much higher portion of the supply voltage than the EL84s.

To even out heater voltages you need to wire appropriate resistors (Rx) in parallel with the 12AX7s' heaters, so that 21Ohm//Rx = 8.3 Ohm. I get Rx = 14 Ohm. Wattage rating at least 5W, wirewound.

[Malcolm Irving]

Sorry, it wont work. When you put things in series they have equal currents, but don't necessarily share the voltage equally. A 12ax7 with its two heater in parallel requires 6.3 volts and 0.3 amps (resistance = 6.3/0.3 = 21 ohms), while an EL84 requires 6.3 volts and 0.76 amps (resistance = 6.3/0.76 = 8.3 ohms).
The overall resistance of your series connection would be 21 + 21 + 8.3 + 8.3 = 58.6 ohms. The voltage from the Classic transformer must be 4 x 6.3 = 25.2. The current in your series connection would be 25.2/58.6 = 0.43 amps. This would give 0.43 x 21 = 9 volts across each 12ax7 and 0.43 x 8.3 = 3.6 volts across each EL84.

[Helmholtz]

Sorry, it wont work. When you put things in series they have equal currents, but don't necessarily share the voltage equally. A 12ax7 with its two heater in parallel requires 6.3 volts and 0.3 amps (resistance = 6.3/0.3 = 21 ohms), while an EL84 requires 6.3 volts and 0.76 amps (resistance = 6.3/0.76 = 8.3 ohms).
The overall resistance of your series connection would be 21 + 21 + 8.3 + 8.3 = 58.6 ohms. The voltage from the Classic transformer must be 4 x 6.3 = 25.2. The current in your series connection would be 25.2/58.6 = 0.43 amps. This would give 0.43 x 21 = 9 volts across each 12ax7 and 0.43 x 8.3 = 3.6 volts across each EL84.

Seems you missed my post above .

[Malcolm Irving]

Yes - simultaneous posting!

[Malcolm Irving]

Helmholtz suggested parallel resistors would work well. It presents the same load on the transformer as the original 4 x EL84 circuit. (Actually an easier ride for the transformer as the overall 'cold' resistance will be slightly higher.)

[KCman]

Thanks for the quick reply. Makes total sense. I wasn't thinking of the resistance and mismatched current.

I could regulate DC to all. Any benefits to either way? (Besides maybe less hum) adding resistors seems easier. I will have to buy parts now either way.

Thanks again, I've been repairing and adding simple mods for a while. I am new at building / designing. You guys have made me think in a new way about a couple things!!

Kirk

[mozz]

Why not run 4 of the el84 and 2 12ax7 as the original? You can switch off 2 of the output tubes if needed but leave the filaments lit.

[J M Fahey]

You canīt use "ohms" in this case since filaments are not standard resistors but strongly variable PTCs.

They can be considered somewhat "equivalent" resistors as they will have some current clue and voltage drop ... but such calculated value will hold only at *that* point and nowhere else.

That filament winding expects 4 EL84 filaments in series ... and nothing else.

2 series EL84 ones, will drop 12.6V and pass 0.76 A .

You need another 12.6V 0.76A load in series or values will be anywhere but needed.

2 12.6V filament 12AX7 in parallel will pass a total current of 0.3A (0.15A each), so you need a parallel resistor passing 0.46A when dropping 12.6V so 27.4 ohm , and dissipating 6W .

[Helmholtz]

2 12.6V filament 12AX7 will pass a total current of 0.3A , so you need a parallel resistor passing 0.46A so 27.4 ohm , and dissipating 6W .

Or use the 6.3V/300mA wiring for each 12AX7, wire the tube heaters in series and place a 14 Ohm resistor in parallel with each 12AX7 heater, as I suggested above.
Or you could wire a single 27.4 Ohm (=~ 2x14 Ohm) resistor in parallel with both 12AX7(6.3V) heaters.

You canīt use "ohms" in this case since filaments are not standard resistors but strongly variable PTCs.

Why not? We use ohms with PTCs as well. Here we are dealing with the hot (operating) resistance, as defined by the tube datasheet values of nominal heater voltage and current.

Apart from that, tubes like EL84, ECC83, 12AX7 are designed/recommended for parallel (constant voltage) heating, while some TV tubes (in Europe "P" series) were specially designed for series heating.
I.e. for the tubes in question, series heater wiring is not the best solution anyway.

[Malcolm Irving]

2 12.6V filament 12AX7 in parallel will pass a total current of 0.3A (0.15A each), so you need a parallel resistor passing 0.46A when dropping 12.6V so 27.4 ohm , and dissipating 6W .

I think O/P meant the 12ax7 filaments are parallel within the tube (i.e. 6.3V) and then two 12ax7s are in series.
Helmholtz suggested two resistors, each one in parallel with one of the 12ax7.

[Malcolm Irving]

You canīt use "ohms" in this case since filaments are not standard resistors but strongly variable PTCs.

They can be considered somewhat "equivalent" resistors as they will have some current clue and voltage drop ... but such calculated value will hold only at *that* point and nowhere else.

Yes, true. If I knew the 'resistance' of the filaments as a function of the current passing through them, I could have made an accurate calculation of the voltages and currents resulting from O/Ps original proposal. By assuming nominal 'resistances' my calculation would have worked out with the appropriate voltages and currents, if the proposed circuit was valid. The idea was just to show that the voltages and currents don't work out.

[KCman]

Thanks again.
Im not really a fan of fillaments in series anyway. I should wait for a better PT.
I have an OT for 2 el84 from an old Baldwin organ amp. I have one for 4 el84 as well, but I thought to save that for something better. I just wanted to use what I have to get rid of some crap and learn a little!

My goal was to only buy caps. Oh well

[loudthud]

Since you only want to run two output tubes, why not use 12V6 or 12AQ5 or even 25L6. You can probably find NOS tubes cheap.

[J M Fahey]

Im not really a fan of fillaments in series anyway.

Tell that to Peavey
Too late now to complain