Salvage parts revD build

[whitelabrat]

Hi all

I've got one of the rev D ST35 boards and I'm planning to put together something using salvage and leftover parts from my parts bin. Here's my plan so far.

I have an old Calbest EL84 amp that I'm going to salvage the iron from. The power transformer is an enormous filiment ps intended for tube rectification. Unloaded the voltage is about 350-0-350. I intend to use a 5U4GB for rectification. I've already got a 33uf 500V cap ready for C9. The OPT's are 8k ohm with 8-16-32 ohm taps. Pentode only. I've also got a JJ 250uF 500V (for C12?), two Panasonic 120uF 500V caps, two .1uf caps of unknown origin, mabey a .1uF orange drop and lots of older (than me) carbon resistors. I bought some tube sockets for point-to-point wiring, but I'll fudge them in somehow.

I'd like to replace the 12AU7 with a 6CG7. I won't be able to solder the tube sockets to the board so it shouldn't be difficult to wire the heaters for 6.3V.

Anyone try a 12AT7 instead of the 12AX7?

[EWBrown]

I haven't tried a 12AT7 for the VA stage, there is no reason it shouldn't work, though it will have about 40% lower gain, which isn't a real problem in most cases. Others have used a 5751 as a direct plug in substitute, the 5751 has a gain 70 vs the 12AX7's gain of 100.

As a "plan B" you might consider using 12AT7s as both VA and PI stages.

The NFB may have to be fiddled a little bit, initially set the NFB pots as in described in the instructions, then any "tweeking" can be done afterwards, as needed.

The plate and cathode resistance values would have to be changed to accomodate the 12AT7s, and to keep the PI's grid at the right voltage level (this determines the actual current through the 33K and 27K resistors in the PI stage).

I'll try a tubeCAD run and see what I get and post it here later.

With the vintage trannie and tube rectifier, the main "prime directive" is to get a B+ voltage somewhere between 360 and 380VDC at the OPT center taps. A little more or less won't hurt, I've used betwen 330VDC and 400VDC with no probs...

/ed B in NH

[whitelabrat]

I just had an even more sinister idea. My output transformers have a 32ohm tap on them. Grado RS2 headphones are rated for 32ohms. What if I replaced the EL84's with something less powerful and then ran in triode mode? ST35 headphone amp?

I guess the magic question would then be, what pentodes could fit into the circuit and put out less than 1 watt in triode operation?

[whitelabrat]

Assuming I could rewire the sockets as needed how about push-pull using something like a 12BY7 or 12GN7

[whitelabrat]

Forget all the other madness I mentioned earlier. I'm going to build a 6N1P OPTL headphone amp. Someone also made a good point that I should just do point-to-point if I want to seriously deviate from the PCB circuit. So for the most part I'll keep things as is.

I have a couple variations that I'm interested in trying though. First is I got a Hammond 373BX dirt cheap. It's 350-0-350v instead of the PA774's 330-0-330. I intend to use a 5U4GB tube rectifier so that should introduce some voltage drop. I'm also going to use a Triad C-14X choke which provides 6H with 150ohms resistance which should also get the voltage down to the target. C9 will be a 33uF with otherwise the same specs as the 120uF's to be 5U4GB friendly.

I'd like to calculate the voltages. Does anyone know the current draw on the that I could expect? If I needed to adjust the voltage I assume that R34 and R35 are the guys to mess with.

Also, what is the appropriate way to mount the variable resistors? Top or bottom?

[whitelabrat]

What are U1, U2, U3, and U4 for on the board?

[whitelabrat]

Ok. I've done a lot of digging and I believe I've got some answers for myself and some new questions.

U1, U2, U3, and U4 are part of some CCS deal, but they require either triode or ultralinear operation, correct? I intend to operate in pentode with my old OPT's so I guess I can't use the LM317's?

If going with a 5U4GB rectifier D1 should be jumpered, D2 unpopulated, D35 and D34 removed, and the choke attached at the two unmarked points in between R35 and R34. Right?

[Ty_Bower]

I'm talking out my rear end, but I think it's OK to use a constant current source on a pentode output stage as long as you have a tightly regulated screen supply.

I also seem to recall there is a thread around here where someone actually tried out the CCS option. I believe they didn't find much audible difference.

[mesherm]

I used this on a 6V6 PP amp I built and it worked fine. I don't see why it wouldn't work of EL84s. If you don't have UL taps just connect the screens through a UF4007 and a 250 to 1k ohm resistor to the either the plate (for "triode" mode") or B+ (for pentode mode). You don't HAVE to use the diode. You can use the traditional a 100 ohm resistor to the plate or have a separate B+ for pentode operation.
The TL783s will work better than the LM317s. You can used a larger value bypass cap also if you wish. I just had 200uf handy at the time.

[whitelabrat]

Thanks for the schematic! I'm just going to skip it for now.

I got a good couple of hours last night for stuff'n. I still need to figure out the trim pots. Heads or tails? Anyway photos:




My cheap sockets were too small for the PCB so I had to bend the pins to fit. I used Cardas solder on most of the parts. They all look like cold solders, but I know they aren't. I had the soldering iron cranked up pretty high and I cooked the board pretty good.



You'll notice the 33uF 500v cap at C9. I couldn't find 716P's at 400V so I went with 600V's. The massively oversized .22uf's were just laying around wanting for something to do so I used them.