DC preamp heater supply, are these tubes polarity sensitive?

[2thick4uni]

I’m considering adding a DC filament supply to a preamp I'm building, which has 12AX7, EF86 and 6SN7 tubes in it.

I know that the usual advice with 12AX7 is to connect positive to pins 4&5 and negative to pin 9. Does anyone know if EF86 and 6SN7 heaters are polarity sensitive with DC heater supplies? I'm grounding negative of heater via a 100nf cap and using 4,700uf smoothing cap as suggested in "Inside tube amps" by Dan Torres.

Thanks,
Allan

[UR12]

There is no polarity on the heaters otherwise you wouldn't be able to run them off AC. If it was a directly heated cathode then you may want to worry about it but it should still work either way.

[2thick4uni]

There is no polarity on the heaters otherwise you wouldn't be able to run them off AC. If it was a directly heated cathode then you may want to worry about it but it should still work either way.

Thanks Dana, that what I'd hoped to hear

Any idea why Dan Torres states that positive should go to pins 4&5 and negative to pin 9 on 12a_7 series tubes?

[skyboltone]

I guess I just wonder why. A properly dressed set of AC heater wires need not cause any hum. If hum is a problem, change to humbucking pickups. Perhaps that's a little simplistic.

The only cautions that I know of regards voltage on heaters. The tubes are designed, and manufactured for AC voltage. 6.3 VAC has a peak +/- voltage swing from 10.72+ to 10.71 -. Has anyone seen a literature that describes how we best arrive at a good DC voltage? Is RMS good enough?

I remember reading an article long ago about the bad effects of either too low or too high voltage on filaments. Imagine a set of Black Plate RCAs with a 300% shorter life due to wrong filament voltage.

What are exactly the benifits of DC on the filaments? Can somebody point me to an article. I think Fuchs uses DC, also maybe many others. Aiken? I dunno.

Help on this one would be appreciated

Thanks
Dan

[UR12]

I guess I just wonder why. A properly dressed set of AC heater wires need not cause any hum. If hum is a problem, change to humbucking pickups. Perhaps that's a little simplistic.

The only cautions that I know of regards voltage on heaters. The tubes are designed, and manufactured for AC voltage. 6.3 VAC has a peak +/- voltage swing from 10.72+ to 10.71 -. Has anyone seen a literature that describes how we best arrive at a good DC voltage? Is RMS good enough?

I remember reading an article long ago about the bad effects of either too low or too high voltage on filaments. Imagine a set of Black Plate RCAs with a 300% shorter life due to wrong filament voltage.

What are exactly the benifits of DC on the filaments? Can somebody point me to an article. I think Fuchs uses DC, also maybe many others. Aiken? I dunno.

Help on this one would be appreciated

Thanks
Dan

I don't profess to being a expert on dc heaters but the DC voltage on the heaters should be 6.3vdc. As I understand it running them a little high shortens the life of the tube and running them a little lower, say 5.7vdc wont hurt anything .

The only reason I can think of by todays standards, in a guitar amp, to run DC heaters is so that you can run some of the older Octal preamp tubes like the 6sn7, 6sl7, 6sj7. They tend to produce a lot of hum when run off of AC. Maybe this goes back to the days when tubes were powered off of batteries. Can you imagine what happens to your heater battery voltage as the battery runs out of juice?. It's voltage goes down and this never seemed to be a problem in the old days.

So I personally don't see any benifits , unless you are using tube types that produce hum when run off an AC heater supply.

I also don't understand why Mr. Torres would prefer pin 9 for the ground unless he just didn't like the idea of having a ground pin (4&5) right next to his plate pin (pin6). Grounding pin nine just puts the ground next to pin 8 of a 12a_7 which is the cathode. this is complete speculation on my part.

Another wierd thing is that Ken Fisher seems to have ran his heaters alternating from pin 9 to pin 4&5. In other words on his preamps he runs pin nine of the first tube to pin 4&5 of the second tube and pin 4&5 of the first tube to pin 9 on the second tube. Must be mojo or can anyone explain the benifits of this wiring scheme?

[philmanatee]

Doesn't Fender run their heater wires that way too? Phil

[sliberty]

I always heard that Fender was semi-random in their heater wiring. They apparently never told the assembly ladies to keep it consistant, so they just did whatever happened.

[Chris333]

I tried an unregulated DC heater supply in a build with a couple octals in the preamp a while back. Saw the Torres thing and it looked so simple I had to try it. Noisier than AC (I put in a switch so I could compare AC to DC in real time). I ended up with a separate filament transformer, rectified and filtered but not regulated, with two 100 ohm Rs as a virtual center tap. That helped. It's quieter than AC but still noisier than a battery. The difference is pretty noticeable with the amp cranked, but at lower than police-visit volumes, it doesn't seem to matter much. Maybe the problem is going from 60hz hum to 120hz ripple. 120 sounds more up-front, at least to my ears... Not sure if a regulator can match the quietness of a battery, but if it can, it would be worth it.

[2thick4uni]

Thanks for all of the comments so far. Reason I'm interested in dc heaters is that I rewired my champ 5E1 clone with dc heaters and it did make a considerable difference to hum level, but not sure how much was down to bad lead dress rather than dc supply. I think dc heaters are more common among hifi tube builders because they are in persuit of very quiet amps, but it would be interesting to compare dc with properly dressed ac wiring to see if there really is any difference.

From what I understand with dc heaters the negative is not grounded, it is connected to earth via a 100n cap which ac couples it to ground.

Another wierd thing is that Ken Fisher seems to have ran his heaters alternating from pin 9 to pin 4&5. In other words on his preamps he runs pin nine of the first tube to pin 4&5 of the second tube and pin 4&5 of the first tube to pin 9 on the second tube. Must be mojo or can anyone explain the benifits of this wiring scheme?

Yep, thats got to be Mojo as its ac but maybe someone will point out the phase difference if alternative wired and tell us that it makes a big difference to how the tubes sound.

It's interesting to look at the heater wiring in the Franchesca TWreck pix, it is very loose and hardly twisted at all, certainly compared with most boutique builds which are very tightly wrapped and neatly routed. How hum free were the original Trecks against modern boutique stuff?

[David Root]

I second the comments about heater current on 6SL/N7s. My current amp uses 4-6SL7s and one 6SN7. My PT is copper strapped and has the internal grounded screen between pri/sec windings, but I still had to go to DC heaters because of massive hum & buzz.

I took the 6.3 VAC supply through a 25A 50V bridge rectifier, followed by two 10,000 mF 16V caps in a PI with a 1.5mH 5A DC choke. Cap off the CT and ground the DC through a pair of 1% 100 OHM resistors. The 1.5 mH inductance takes out the DC overvoltage from the BR perfectly, giving 6.1 to 6.3 VDC depending on your AC line. (See K'OC TUT 2, fig. 2-76). Choke I used is Hammond 156B. Results were fantastic, night and day difference.

I put the DC feed on all five octals plus the reverb driver 6K6 and four 6V6s, since I had 9.6 VAC from the PT, which translates to about 5A DC. I went into the power tubes first, but I'm told you should start at the preamp V1 position with the 100 ohm pair right before that point, so I have a little wiring to do.

[Bob-I]

Another wierd thing is that Ken Fisher seems to have ran his heaters alternating from pin 9 to pin 4&5. In other words on his preamps he runs pin nine of the first tube to pin 4&5 of the second tube and pin 4&5 of the first tube to pin 9 on the second tube.

He claims that any hum injected in one stage will be canceled at the next. Doesn't make since to me because the 2 triodes would cancel it out in one tube anyway.

As long as the heaters are done right with a ground reference I've never had hum from heaters. The simple way to check is to connect the heaters to a 6V lantern battery and see if the hum is reduced. I had one case where imbalanced 100/100 ohm false centertaps caused a hum, one resistor was about 88ohms the other was almost 120 (no more Radio Shaft components). Putting in a decent pair of resistors made the amp almost silent.

[FUCHSAUDIO]

As it's not exactly a trade secret, I felt I'd share: In lieu of messing with AC hum balance controls or lead dress, since we don't use the 5.0 volt winding on the PT's we buy, we voltage double it and use a packaged 7806 regulator. Mounting on the aluminum chassis saves heatsink cost, and all three 12AX7's run on DC filaments. Basically near silent, as far as hum is concerned. We tried numerous brands of 12AX7's, with varying success as far as hum rejection. If you use tube rectifiers, add an extra transformer and have fun.

Tubes are not polarity sensitive (but we maintain polarity anyway), tubes sound no different than on AC (despite some guru's professions gibberish to the contrary), and they are certainly quieter. Observe proper grounding techniques, and you should be fine. Single point ground !!! For everything/everywhere.

[UR12]

As it's not exactly a trade secret, I felt I'd share: In lieu of messing with AC hum balance controls or lead dress, since we don't use the 5.0 volt winding on the PT's we buy, we voltage double it and use a packaged 7806 regulator. Mounting on the aluminum chassis saves heatsink cost, and all three 12AX7's run on DC filaments. Basically near silent, as far as hum is concerned. We tried numerous brands of 12AX7's, with varying success as far as hum rejection. If you use tube rectifiers, add an extra transformer and have fun.

Tubes are not polarity sensitive (but we maintain polarity anyway), tubes sound no different than on AC (despite some guru's professions gibberish to the contrary), and they are certainly quieter. Observe proper grounding techniques, and you should be fine. Single point ground !!! For everything/everywhere.

That's cool, Thanks for sharing! Myself and a few others were looking at just this solution a couple of weeks back but I was a little worried about running the 7806 as it is rated at 1A continuous and 1.5 amps for short periods. We figured worse case would be about .3a for each 12ax7 heater @6v which would come out to .9 amps for 3 12AX7s. We also found that if you outboarded a 12.6 volt heater tranny and used a 7812, the 12ax7s would only pull .15 amps each at 12v and get us down to .45 amps @12v which is well below the rated output of the 7812 regulator. Unfortunatly we were also using a couple of 6sl7s for a second channel and decided to go with the 25w bridge and 10000uf cap. It's nice to know that you have already tried this and it works with 3 12ax7s and the 7806. It might be nice to get the PT wound with a 12.6v winding instead of the 5 volt next time around since we aren't using a Tube recto.

Thanks again!

[Ron]

The centertapped heaters in a 12AX7 are humbucking, so you don't have to worry about which wire is connected to pins 4/5 and 9.

However, the filaments in 6L6-type tubes and EL34s are not humbucking. In their case, you have to mind which filament wire goes to which pin, so that the push-pull action of the output stage provides a hum-cancelling action.

For a pair of output tubes, connect pin 2 of one tube to pin 2 of the second tube, and similarly with pin 7. This may be what Ken Fischer was referring to, and not the preamp tubes.

[UR12]

Ron

Not only are the El34s wired up alternating 2 and 7 so are the preamp tubes alternating 4,5 - 9. Personally I cant see any beniifit in doing this but I wired my Express and Liverpool this way and have no problems with hum. I'll have to try wiring it up normally on my next build and see how it compares to Kens's method. I don't expect to hear any difference. So what would be your recomendation on a parallel output EL34 SE design?