DC Heaters for V1 & V2

[Gregg]

Elevation of DC heaters is still desirable if the circuit includes cathode followers or cascodes, etc., that run the risk of otherwise exceeding the maximum heater-to-cathode potential permissible for the tube.

+1
However if you really want to protect the CF that 22k resistor should be changed to 68-82k.

[jmaf]

The rectified DC voltage available to the heaters is likely to be less than 6.3VDC due to the voltage drop of the pn diodes.

Are you sure about that?

[Merlinb]

If the heater transformer really is giving 6.3V then you will indeed get close to 6V DC after rectification and smoothing, because the AC waveform gets clipped by the rectifier, reducing the peak voltage. You don't get the ideal 1.414 multiplication when working at low voltages. Of course, everyone will be familiar with heater transformers giving somewhat high voltages, so you might get 7V DC or something, depending.

One error in that schem: There should only be ONE pair of 100R balancing resistors- ideally the ones on the AC side of the circuit.

[jmaf]

The only time I did this, which I mentioned above, I rectified the 12.6 V, which gave me around 17 VDC, which I then regulated down to 12.6 VDC. I didn't do it on the 6.3 line, but that is an interesting gotcha to keep in mind.

[jmaf]

One error in that schem: There should only be ONE pair of 100R balancing resistors- ideally the ones on the AC side of the circuit.

I thought about that but wasn't 100% sure. When the AC pulse is going negative it's a short, right?

[diagrammatiks]

just to chime in...

elevation was never really a noise issue...it just happened to help that out.

there are a variety of different circuits that require different heater elevations. The cf is one of them but there are others too. Most of the time the cathode follower can get away with having an elevation but that isn't true of all circuits.

As for DC heaters, as jmaf said it is a pain to learn how to do it but you only gotta learn it once.

[trobbins]

I think the author's intent was to have two pairs of midpoint dividers. Rectification noise may couple in to the valves with DC heaters, and would be alleviated by capacitive bypassing the DC heater to ground (either via the midpoint divider, or a simple capacitance from one arm of the heater).

[diagrammatiks]

ah yes i see now. unregulated dc heaters.

well then let big bubba do the work for you!



the vht schematic linked above has 12000 i think. 4700uf is really too little for unregulated heaters.

[Gingertube]

Elevating the heaters with respect to the cathode DOES impart some sonic benefits, usually not all that important for guitar amps but significant in HiFi design.
Allen Wright always recommended +25V minimum regardless of whether AC or DC heating was used in his HIFI preamps.
The old tube designer bible (RDH) recommended +70V.
I have traditionally used DC heaters sitting on top of +45 to +50V for all my HiFi designs and use similar elevation of AC heaters in guitar amp design.

A big hint here if using vintage power trannies and not a tube rectifier (i.e. you have a free 5V AC winding), a bridge recitifier using schottky diodes into about 10,000uF off the 5V AC winding gives 5.95 to 6.0 V DC, just right for heaters for the first stage or 2 of the guitar amp.

Cheers,
Ian

[trobbins]

Measureable benefits?

[Gingertube]

Don't actually know for Guitar Amp work - to be honest!!!

In my latest HIFI Amp Design I ran out of heater current and used that shottky rectification of the 5VAC supply for the input stage. Sitting that on +50V gave audible reduction in hum level. This was observable on an Oscilloscope - also the noise went from being a "buzz" (that is 100Hz residual supply ripple plus lots of harmonics) to more of a genuine 100Hz hum, that is, it seemed to be effective in removing/suppressing the higher frequency harmonics of the hum.

It is so easy to implement that I "just do it" for guitar amps (whether DC or AC heaters) and haven't really bothered to make measurements (except as above) to justify the extra $1 worth of components.
I note that some other designers (Kevin O'Connor for example, to name just one) also do it routinely.

I'm afraid that deciding what of the things I do for HiFi tube amp design is a waste of time, or worse, counter productive, for guitar amp design is one of those things I'm still coming to grips with. The design goals are quite different and so it is not surprising that what works for one may not work for the other, the difference being that a guitar amp is a musical instrument in it own right whereas the HIFI Amp is less so. I say less so (rather than is not), as I can not bear those 0.001% THD+Noise SS HiFi Amps, cold sterile and boring.

Cheers,
Ian


Cheers,
Ian

[Danelectron]

I have elevated the AC heater voltage on a fender build and had favorable results. Though not a big improvement I left it. Where would a person want to do this in a 2204? I recently built a 2204 that is plenty quiet aside from a slight buzz wide open. It's basically a stock build with the PPIMV master volumn. I'd be curious to see if it improved that. Thanks, Dan

[soundmasterg]

I used the DC filament circuit from a Sovtek MIG 100 when I rebuilt that amp into a Vox AC100 clone. I used DC on the first preamp tube for each channel (I added a second channel on the AC100 with more gain) and the amp is extremely quiet. It is a regulated circuit and the input voltage is around 13v AC. It rectifies around 6.3v and is perfectly stable with two preamp tubes. Any more than that and it wouldn't regulate well.

Greg

[trobbins]

Ian, I like the idea of using the 5VAC heater - that may be a useful work around.

Heater elevation is an interesting topic - I recently did some 12A?7 heater cathode resistance measurements and the results suggest that most amps would show very little hum benefit, if discernable at all (Heater-cathode DC conduction resistance - diyAudio). The most effective single change is probably going to be a humdinger pot, and modification of poor heater wiring - there have been some good examples of how effective this can be, including when rectified DC related noise is noticeable (AX84.com - The Cooperative Tube Guitar Amp Project).

A technique I found neat was AWA's use of input valve(s) heater as a major part of the cathode resistance in cathode biased PP amps. They put a failsafe bypass resistor across the heater(s) to cold bias the output tubes if a heater failed, or a valve was pulled. Although not suitable if you want a standby power switch, it also minimises the heater winding current rating and doesn't require heater rectification to get DC heater current, or careful heater wiring layout.

Ciao, Tim

[diagrammatiks]

this is from morgan jone's valve amplifiers. please don't sue meeee



I haven't implemented it yet because I don't have a place to mount the transistors but it's supposed to give a rock solid voltage reference off the b+.