DC Filaments, any cons?

[duffy878]

I'm considering changing one of my amps to use DC filament wiring.

Are there any downsides to using DC rather than AC filament wiring?


Thanks in advance guys

[loudthud]

Most builders only run one or two preamp tubes on DC. Many have tried, many have failed. Do a search. No, you can't just rectify the 6.3V heater winding and connect a three terminal rectifier. ROFLOL

[olddawg]

You know, I built an 18 watt Marshall clone a while back. I was so disgusted with the hum that I considered using DC filaments. It was a second hand build. I picked it up cheap from a guy who had given up. It took me a couple of weeks to figure out the problem. It turned out there were several. The ground scheme was wrong, the preamp B+ supply was coming from the wrong point in the power supply, and the filament dress was sloppy. Now it is the quietest amp that I own. I guess what I am trying to say is that there is no bandaid fix for bad design and sloppy work. If you do it correctly you will not need a dc filament in a guitar amp.

[tedmich]

It is sometimes easier to do than a proper lead dress, but often not as rewarding. My Peavey Bravo has DC heaters on its 4 gain stages (27VDC in series!) and then 6.3VAC on the 12ax7 driver tube and 2x el84 PP.

[duffy878]

interesting... thanks for the comments.


I converted an old shortwave radio to a guitar head over the summer but its got some hum and whatnot.

I know the heaters are actually wired in series, which Ive never seen before, and that is probably the source of the noise. Ive read about DC heaters (somewhat) but never really a full explanation of when its good and when its bad and all that.

Its really not a problem I'm just looking to improve it.

[R.G.]

Ive read about DC heaters (somewhat) but never really a full explanation of when its good and when its bad and all that.

In a tube, practically everything matters at least a little.

Hum from heaters comes in two flavors - capacitive pickup and stray emission.

Anything with a varying voltage on it can couple capacitively. You pays your money and takes your chances. The dual heater tubes like the 12A?7 series have the heaters wound reverse directions from each other so that the radiated E field and M field should cancel somewhat.

Heaters in tubes with oxide-coated cathodes (that's all of them you're ever likely to mess with) can pick up stray emission oxides in the tube and put out their own pulsating batches of electrons. This will cause hum as well.

DC prevents capacitive coupling and stray emission hum entirely. The heaters don't wiggle around, so they don't emit. However, DC causes its own set of problems, largely in that we no longer use batteries to generate DC very much, we use rectifier/filter/regulators. You have to filter the DC you feed your heaters pretty well, or the hum just moves to twice the line frequency. And you can't simply rectify and filter the 6.3Vac, because the DC you get that way has so much in the way of losses that the result is not a particularly good DC heater supply. So DC heater supplies get complicated and difficult to do from the same power transformer that did your AC heaters.

There is also a DC failure mechanism. The heaters get hot (well, duh... ) which makes the metal try to "emit" metal atoms, simply boiling them off. Any thinner spots get hotter, and so they emit faster. That makes them hotter, and that makes them emit more... AC power on heaters reversing the polarity of the heater voltage at line frequency seems to make this matter less. It's a subtle effect, and usually only a consideration on non-oxide cathodes which run hotter and which you're unlikely to ever use, but it's there.

Another option is to simply elevate the heaters by a few volts positive. That way any emitted electrons are attracted back to the heaters instead of straying off to cause hum. That's why you see elevated heaters.

Frankly, DC and elevated heaters are more trouble than they're worth. Proper grounding and wire routing will make the hum so low that these tricks are not necessary. I've done this personally on some amps. In that sense, DC on heaters is a coverup for other problems which don't get fixed.

[guitician]

The dual heater tubes like the 12A?7 series have the heaters wound reverse directions from each other so that the radiated E field and M field should cancel somewhat.

Do you mean the spiral filament? And do you mean that they cancel from one half of the tube to the other?

[mooreamps]

Most builders only run one or two preamp tubes on DC. Many have tried, many have failed. Do a search. No, you can't just rectify the 6.3V heater winding and connect a three terminal rectifier. ROFLOL

Ya, well yes you can. True, you can't use a conviential DC bridge rectifier part with the four lugs on the same device. The forward voltage drop on those parts is roughly 0.7 volts, and you lose too much voltage across it. But, you can use a pair of fast recovery diodes, which the forward voltage drop is appox 0.1 volts, and it works fine.


The "con" is ; the FRED diodes are more expensive ; I forget how much I paid for mine {$ 1.50 each maybe} ; verses a 1n4007 ; which I get for about a dime a piece.


But even with dc filaments, I use a hand drill to twist a pair of wires together, real tight, to minimize any radiation coming from the wiring.




-g

[loudthud]

Ya, well yes you can. True, you can't use a conviential DC bridge rectifier part with the four lugs on the same device. The forward voltage drop on those parts is roughly 0.7 volts, and you lose too much voltage across it. But, you can use a pair of fast recovery diodes, which the forward voltage drop is appox 0.1 volts, and it works fine.

The "con" is ; the FRED diodes are more expensive ; I forget how much I paid for mine {$ 1.50 each maybe} ; verses a 1n4007 ; which I get for about a dime a piece.

But even with dc filaments, I use a hand drill to twist a pair of wires together, real tight, to minimize any radiation coming from the wiring.

-g

DC current in a wire doesn't induce any more noise than a magnet would.

Looking at the Vishay spec sheet for a 1N4001 series diode, I see on a graph that the forward voltage drop at 1 amp @25C is about 0.9V. At a temperature rise of 50 degrees C allowing -2.2mV per degree thats about 0.79V. Since in a bridge rectifier there are always two diodes in series, the voltage loss is about 1.58V.

Searching on Mouser for "Fred", the cheapest diode is the HFA15TB60. From the IR spec sheet the forward voltage drop at 1 amp is about 1.125V at 25 degrees or roughly 0.9V at 125 degrees. Again there will be two diodes in series when a bridge rectifier is used.

Using the PSUD simulator, 6.3V winding with 0.1 ohm bulk impedance, 4700uF with 0.1 ohm ESR, looking at the numbers in the little window on the left the loss across the bridge rectifier is 1.7346V. The peak voltage on the cap is 6.9284V, average voltage is 6.4547V, minimum is 6.0055. An LM317 regulator requires about 1.8V between input and output at 600mA so you could make a 4.2V power supply if the line voltage never goes below the nominal line voltage rating of the transformer (a cold day in hell). Note that the peak current from the transformer through the diode bridge is 3.5449 AMPS when the load current is 600mA. The RMS transformer current is 1.2895 AMPS.

Actually, the unregulated voltage would run a pair of preamp tubes just fine but there would be almost a volt of sawtooth ripple.

Happy Halloween!

[loudthud]

interesting... thanks for the comments.

I converted an old shortwave radio to a guitar head over the summer but its got some hum and whatnot.

I know the heaters are actually wired in series, which Ive never seen before, and that is probably the source of the noise. Ive read about DC heaters (somewhat) but never really a full explanation of when its good and when its bad and all that.

Its really not a problem I'm just looking to improve it.

The series heaters connection is a different problem. Does the amp have a power transformer? Does it have tube types like 50C5, 50L6, or 35W4 ?

[tedmich]

DC current in a wire doesn't induce any more noise than a magnet would.

Looking at the Vishay spec sheet for a 1N4001 series diode, I see on a graph that the forward voltage drop at 1 amp @25C is about 0.9V. At a temperature rise of 50 degrees C allowing -2.2mV per degree thats about 0.79V. Since in a bridge rectifier there are always two diodes in series, the voltage loss is about 1.58V.

Searching on Mouser for "Fred", the cheapest diode is the HFA15TB60. From the IR spec sheet the forward voltage drop at 1 amp is about 1.125V at 25 degrees or roughly 0.9V at 125 degrees. Again there will be two diodes in series when a bridge rectifier is used.

Using the PSUD simulator, 6.3V winding with 0.1 ohm bulk impedance, 4700uF with 0.1 ohm ESR, looking at the numbers in the little window on the left the loss across the bridge rectifier is 1.7346V. The peak voltage on the cap is 6.9284V, average voltage is 6.4547V, minimum is 6.0055. An LM317 regulator requires about 1.8V between input and output at 600mA so you could make a 4.2V power supply if the line voltage never goes below the nominal line voltage rating of the transformer (a cold day in hell). Note that the peak current from the transformer through the diode bridge is 3.5449 AMPS when the load current is 600mA. The RMS transformer current is 1.2895 AMPS.

Actually, the unregulated voltage would run a pair of preamp tubes just fine but there would be almost a volt of sawtooth ripple.

Happy Halloween!

I believe schottky diodes have a lower V drop across them, the ST STPS20L45C dropping only 500mV max, and there are regulators which have lower drop out voltages than the LM317; the Sharp PQ7RV4J0000H will give 6.3VDC at 4A with only 6.8VDC input. Combining these and you might be able to get a usable DC heater voltage from a 6.3VAC winding.

[guitician]

The series heaters connection is a different problem. Does the amp have a power transformer? Does it have tube types like 50C5, 50L6, or 35W4 ?

Right, I've got an old Magnatone amp that runs a 12ax7(4>5), 35Z5 and a 50L6 in series with a 10w 150 ohm resistor and 6V Pilot lamp across the AC line. The thing cranks when it's turned all the way up, with surprising little hum.

[raiken]

But, you can use a pair of fast recovery diodes, which the forward voltage drop is appox 0.1 volts, and it works fine.

The "con" is ; the FRED diodes are more expensive

-g

I'd like to see the FRED that has a 0.1V drop at rated current! No way in hell!

A Schottky is your best bet for low forward voltage drop.

Randall Aiken

[mooreamps]

I'd like to see the FRED that has a 0.1V drop at rated current! No way in hell!

A Schottky is your best bet for low forward voltage drop.

Randall Aiken

Well, it's been awhile when I bought those. "happy now" ???

[duffy878]

thanks for the explanation R.G. that answered a lot of questions.

loudthud, yes the amp has a power transformer. it has a 6X4 recto, a 12AX7, and a 6AK6. I posted about it in the conversion forum a few weeks ago.

I guess then if DC is not the way to go, what would be the best solution for series wired heaters? Its rather tight in the chassis so I don't think going to parallel heaters is a reasonable option.