1K 25W power supply R?

[Phil_S]

I've been contemplating a few alterations to the power supply ladder in my variant of the Express. I built it with a pair of 6AQ5. To accommodate the lower voltage ceiling of the power tubes, I used a lower voltage transformer. I kept the power supply ladder almost true to the Francesca schematic. The only change I made was to use a 9k1 instead of 18K2 for the drop between the plate and screen supply nodes. This kept the preamp voltages about where you'd see them in the original design.

Plates are at 333V and screens are at 326V. I know this is high for PP 6AQ5's. I've got a pair of NOS Sylvania that seem to be doing OK, though. I've also got a couple of other amps running this tube at 310-315 range.

I've been thinking about this for a long time, maybe a year, and believe that I'm probably going to be OK at 333V on the plates, but I really should lower the screen voltage to around 300V. Please be gentle here as I know this may be considered enough of a deviation to have the amp no longer regarded as anything resembling an Express.

First, I want to change the 1K 25W dropping resistor. Is there any magic mojo to using a 25W resistors? This is clearly over-spec'd. Anyone know the reason for a 25-watter? I'm thinking I can get by just fine with a 10-watter. I haven't calculated the correct value to drop the screens just yet, and I expect I will also have to reduce the value of the next dropping resistor to keep the preamp voltages where they are supposed to be. My stumbling block is that I don't understand why such a big R was selected and whether I'm going to get an unexpected change in character when I use a 10W resistor.

The other question is, whether anyone has a sense or opinion about what happens to the tone when the screens are running so much below the plates.

Thanks for whatever you've got to offer here.

Phil

[rooster]

Really, I think at this point you shouldn't be so concerned with being true to the Express circuit. ?? I mean, I think the preamp gain stage of the Express has been maintained in your amp and that may be close enough for an Express spin.

So a tweed Deluxe uses a 10K/2 watt carbon resistor as a choke. Hm, a BJr uses a 2K/2watt MF resistor. ?? So I think your path is wide open as to what you want to try, value-wise, and wattage-wise. I have read accounts, mostly from Aikens, that the ultimate sonic product of using a resistor instead of a coil is signifigant... But, well, a wire wound resistor is a coil of wire, albeit without the inductance. SO - to answer that other question, 25W or 10W, maybe Ken was going for longevity, or maybe he considered that the 25W version was closer to an actual wire coil choke. I can't say with certainty. And, well, don't forget, Fender used carbon at one time.

How will it sound, going higher in resistive value? Well, everything downstream will be reduced voltage-wise and that will change things, dramatically if you go with a 10K value, right? But a small value of change, say 2K up from 1K will not be so dramatic downstream. As to the changed sonics at the output tube, well, typically you will have more compression and a slightly reduced volume. This is what happens when you increase the screen resistance. ...Er, but this is only a simple explanation that deals with, in your case, a starting point of 1K. Which is to say that, on the opposite extreme, tying the screens to the plate directly (6L6), with no resistive value between plate and screen, will increase compression and reduce volume also. OK, that's another day..

So go ahead and experiment, that's my input. Look at some other amps besides the ones I mentioned, too.

[Phil_S]

The amp is not a static thing. When I change that dropping R, other things will change, too. I'm thinking I can compensate some by reducing the value of the R between the next two caps and I'll probably have to go down to about 1K. However, to get a 30V drop, I'm going to need a fairly high value R. I'm guessing something in the 20K-30K range. I suppose there's no harm to giving it a try. Thanks for the encouragement.

[Gaz]

You could also use a smallish sag resistor(s) before or after the rectifier. I think 50-120 ohm/10W after the B+ would probably get you there, and I actually think the compression will be less noticeable than with a huge screen resistance, but that's just my opinion.

[geetarpicker]

I'll mention for the record that at least in a stock cranked Express that 25 watt resistor is perhaps not that over spec. In my original and two clones that resistor gets hot enough that I make sure that the two B+ wires that run under it do not touch the resistor on it's underside, to minimize any chance of melted insulation. That said you are talking a lower B+ than standard and different tubes that probably don't draw as much current.

[Gaz]

Have your ever measured the actual dissipation? "Hot" to the touch may not really be that hot to the resistor, plus, either a 5W or 25W will dissipate the same heat, one will just take it easier, if I understand correctly.

[Phil_S]

You could also use a smallish sag resistor(s) before or after the rectifier. I think 50-120 ohm/10W after the B+ would probably get you there, and I actually think the compression will be less noticeable than with a huge screen resistance, but that's just my opinion.

Thanks very much for your insight. A sag R will take the full current draw as it load, won't it? My estimate of load is Ia=52, Ig2=12 (both for the pair), and for preamp tubes, let's call it 70mA total at idle -- we don't need to split hairs here. I imagine under load, maybe it gets up around 80-90mA. Let's call it 90mA. To drop 30V before B+1, where V/I=R, 30/.08 = 333 ohms. I'm also assuming that current draw will rise when voltage drops. I'm not very comfortable with the tube math, having muffed several calculations, I think, due to my non-technical background. I'm not quite clear why you think a ~100 ohm R will get me there.

BTW, I don't have any problem at all with the idea of lowering both plate and screen voltage, but I didn't want to jump through a lot of hoops either. I do follow your logic that this is probably the most transparent way to get it done and hadn't really considered it before. I did think on it for quite a while before I selected a PT. My goal was to get the preamp right and this did force a compromise that ended up with the power section running at higher voltages than I expected. The PT is 275-0-275 and may run a bit higher due to line voltage.

[Gaz]

Phil, your math looks right - I was just too lazy to do it myself, however I did recently try using a 120R between center tap and ground in a similar amp, and thought it dropped B+ around 30vdc. I only clipped it in for a second, so I don't really remember - You know how it is going a little mad late night with the jumper wires, right?

You should try it between the CT and ground first, i you do - Kevin O'Connor writes that it will only effect the power tubes in this position, not the whole B+ line.

Either way, it may be worth a shot, even at 330 ohms, and I'd be curious to know if you find it more tonally transparent. I was pretty surprised to see two 1k resistors in series with each leg of the full wave recto in the X10 in the Xits X10 thread.

[Phil_S]

Gaz: That's a really interesting idea, the OT CT. Should I just be using a big old sandblock resistor, or should I find a zener?

[gearhead]

Have your ever measured the actual dissipation? "Hot" to the touch may not really be that hot to the resistor, plus, either a 5W or 25W will dissipate the same heat, one will just take it easier, if I understand correctly.

There have been numerous threads on this one component. The static current is ok, but you get a surge when the screens start sucking electrons.

Larger value components spread out the heat better. A 10 watter mounted directly on the board has been known to blister the board.