Tube Heaters

[Faber]

Okay, after looking through a few schematics, I have seen tube heaters running on DC and others on AC.  I'm confused about which one it is supposed to be.  Help please?

[ambulancevoice]

it can be either
people say in some tube amp cases that giving the filaments a dc supply reduces hum and noise
for example, the Epiphone Valve Jr

[bancika]

DC heaters will usually reduce hum if done properly (with heavy filtering and regulation), but just as effective is to elevate heaters some 20-80V above ground.
like this

[Gus]

DC can make thing worse if you don't understand what is needed from the design.

As posted floating them up a litle can help as well as having a good layout.

[MarcoMike]

DC can make thing worse if you don't understand what is needed from the design.

can you explain this a little more?

[zachomega]

Tube heaters draw lots of current.  This means you have to have a VERY robust rectifier...especially given that it will get hot and reduce its current rating even further.  Additionally, the voltage needs to be filtered quite a bit. or else the ripple in the power supply is going to play havoc with the tubes. 

Personally, AC heaters have never steered me wrong.  There are a host of things you can do with AC to help reduce the hum...and lets face it, Fender Twin Reverbs at 100 watts are virtually dead quiet - with regard to tube heater noise anyway. 

-Zach 

DC can make thing worse if you don't understand what is needed from the design.

can you explain this a little more?

[mojo_hand]

Tube heaters draw lots of current.  This means you have to have a VERY robust rectifier...especially given that it will get hot and reduce its current rating even further. 

(This is true of tube rectification, if you spent $2.25 for a 35A silicon bridge rectifier and small heatsink, and another $1 for filter caps, you'd have conquered the problem at less cost than than even the cheapest rectifier tube.  But sometimes this isn't convenient, like when you already have a 12.6VAC secondary, since rectifying and filtering that would give about 17 volts... then you either need to wire three 6v heaters in series -- assuming you have multiples of 3, which you probably don't -- or use a regulator to drop it to 12v... it just gets messy.  DC usually works out much better when it's planned for from the beginning, as a mod it may be more trouble than it's worth.)

[petemoore]

  Just wired up a Valve Jr. type amp.
  ROX BTW.
  Heaters @ 6.3vac works great, hum is really very low, output is more than I though of course...
 

[MarcoMike]

DC usually works out much better when it's planned for from the beginning, as a mod it may be more trouble than it's worth.)

Ok, now I get it. of course if someone wants to convert to DC an existing circuit there is the risk of messing up things, probably getting a too high or too low voltage. I was considering this few days ago: if you want to rectify and filter an existing 6.3V ac supply you get less than 9 volts after diodes and caps, which is not enough for a 7806 (diode boosted to 6.3) to work properly...
In the tube amp I'm using a 7Vac supply for this reason (I know a guy who makes transformers.. )

[earthtonesaudio]

Is it true that some tubes are designed for AC heater voltages and some are designed for DC?

[mojo_hand]

Is it true that some tubes are designed for AC heater voltages and some are designed for DC?

In the beginning, tubes were made pretty simply, and usually worked quite nicely with AC.  When tubes with high gain and very high heater currents started to be made (for industrial purposes, mainly), they really could not be used with AC, because those are the exact worst circumstances for hum contribution from heaters.  It was noticed that there was a problem with using DC, which was that, as voltage dropped going from one end of the heater to the other, one end of the filament would end up carrying a disproportional amount of the current.  At that point, some of the tubes which really needed DC were given different sort of winding configurations for their heaters, so that emission would be better distributed.  Some tubes were later designed for use in portable radios, and these were set up the same way, since it was assumed that AC would be unavailable.  But few, if any, older tubes were redesigned with DC in mind.  If you're dealing with a very weak input signal, like a phono front end, you're just kinda SOL, because the usual tubes for the purpose were designed for AC, but heater noise can really put the hurt on your signal when it's that puny.  So you are forced to use DC anyway.

I only like DC because it's so easy to regulate compared to AC, and I like regulation of heater voltages for purposes of prolonging tube life.  I believe that this benefit usually outweighs the problems with uneven emissivity.  But it's a PITA, and arguably not worth it a lot of the time.

Enough of my wandering off topic...

AFAIK, that is the answer.  Some were specifically designed for DC, and even for a particular polarity of DC.  Others were designed for AC, and never changed.  Either type will function when used with the other, but DC tubes running AC will be too hum-prone to be usable.

PS - It being Saturday night, there is no guarantee that anything above will make sense.  Apologies in advance if it's messed up, and... cheers!

[earthtonesaudio]

Thanks mojo_hand.  That makes sense to me.  I just bought an LM317 from Radio Shack, so now I feel more free to start experimenting with some of my old tube stash.

[Krinor]

I'm sorry if this is a little bit off topic, but would something like this work on a simple Fender Champ (5F1-schematic below):

DC heaters will usually reduce hum if done properly (with heavy filtering and regulation), but just as effective is to elevate heaters some 20-80V above ground.
like this

My amp has an irritating hum which I just can't seem to get rid of.

[mojo_hand]

In a guitar amp, heater noise isn't the most likely of major hum sources.  Is there any sort of electromagnetic shielding between your power transformer + rectifier and the rest of the amp?  Other things to consider would be twisted pair wiring for power and heaters, and shielded cable for carrying signal.  Or if the amp's power supply caps are more than 10-15 years old, by all means replace those, as #1 priority.  They dry out over time, and single-ended amplification is very sensitive to power ripple.

[Krinor]

Is there any sort of electromagnetic shielding between your power transformer + rectifier and the rest of the amp?

The amp is as pr the schematic and "normal" layout using point to point terminals. How would I go about shielding ?

Other things to consider would be twisted pair wiring for power and heaters, and shielded cable for carrying signal.  Or if the amp's power supply caps are more than 10-15 years old, by all means replace those, as #1 priority.  They dry out over time, and single-ended amplification is very sensitive to power ripple.

Power and heater cables are twisted. I haven't used shielded signal cables though. The caps are new.
Thanks for your thoughts. I'll try shielded cables first.

[Gus]

SE like the champ amps have a free connection point to raise the fil above ground.  This is an OLD refinment.

Google or ASK or search at this site
The posts are still missing the problem one can have with DC heater supplies.

[mojo_hand]

How would I go about shielding ?

I happen to have just been given an amp that had chronic hum problems, so here's a good example of the work of some technician from last century.


It's kind of hard to see, but the orange wire is disappearing behind a shiny piece of sheet metal (chromed steel?) that someone put into this amp.  The thing that the orange wire attaches to is the base of the rectifier tube, which is the only thing in that part of the chassis, so they were clearly trying to shield the rest of the amp from the rectifier's electromagnetic noise.  In this particular example it didn't solve the whole problem, but it was a sensible thing to try.  A lot of old tube equipment came with metal shields on noisy tubes, such as rectifiers and oscillators, which may have partly been for cooling or prevention of microphony, but which also served to keep the noise from getting into everything.  In some cases, such as this early Scott radio receiver, they plainly got carried away...
 

In audio frequency equipment that dealt with reasonably strong signals, it was usually adequate to put the preamp tubes at one end of the chassis, and the PT and rectifier at the other, so you'll see that sort of chassis layout quite frequently.  The one which I posted up top was not done in this way, the friggin' rectifier is RIGHT NEXT TO the lowest level preamp tube.  That would not apply to an original tweed Champ, or most other great designs.  It might apply to a reissue or clone.  If yours follows the original layout, with the 12AX7 and incoming signal wiring at least several inches from the PT and rectifier, then it's probably not the cause, and you might want to try other fixes instead.  But if it's anything like the amp I'm working on, it should be high on the list of suspects.

[Krinor]

Thanks for your elaborate answer. My tubes are far apart. I suspect it may have something to do with sloppy wiring. I'll have to clean it up a bit and change some of the cables and maybe get a new rectifier. The one I'm using now is ancient. I read somewhere that there was a subsitute for 5Y3 which sounded better. Not that I think a rectifier can have such a great impact on the sound...
I'm sorry if this thread got somewhat hi-jacked... I'll make my own thread if I have further questions.