• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Posted new P-P power amp design

Carbon films are less noisy and I've measured 1.2:1 SWR at 470MHz with a 2-watter. More than enough HF range for audio ;)

Just enough... I've had oscillation at 120MHz. I was afraid that I'd have airliners crashing nearby :eek:

My only worry with the CF's would be that some may be better than others. CC's are a known quantity.

Actually I would expect a ferrite bead would work OK as well.

Pete
 
Got my board today. It's very... red..... What, you don't like red?

I saw the output tubes today. They are very.....RED! I was planning to buy one of these boards sooner or later just to experiment with. I saw that there was only 7 left, so I just ordered one.

I got 20 6GV5's when they were $1 each so I can afford to "test" a few. Do you really think that I will run this amp at 340 volts. Nah. I'll be disapointed if I can't get 50WPC.

I probably won't get to build it until I get back from up north, but I do expect to complete the build for under $75!
 
I've got some UTC LS-54s that were just made for this amp...

I got a pair of LS57's but I think they will go into a fully differential class A P-P amp that is slowly developing in the 6L6GC in AB2 thread. The LS54's and the LS57's are only rated for 20 watts.

I had the same thought- the B+ seems a bit puny.

If I read the schematic correctly there are a whole bunch of different sweep tubes that will plug into this board including 6LR6's (I got some of those too). I have several P-P OPT's to try, so I will leave the power supply stuff off of the board at first, hook it up to some bench supplies, and experiment until I find something I like. Then I will figure out how to make the power supply work.

Pete calls this an "engineers amp". I am sure that he would expect a fellow engineer to do some engineering with his board.
 
I had the same thought- the B+ seems a bit puny.

Let's just say it's conserative :eek:

You can definitely go higher and get more power - I did on the breadboard. But keep in mind I wanted to make this fairly safe for somebody with limited experience, so avoiding glowing plates and exploding electrolytics was on the "design goals" list.

So with 450V electrolytics on the board, you can certainly up the steady-state B+ to 400V without going over 450V at cold start. If you go higher you'll need to kludge some series-connected caps in.
 
This is a push-pull amp pentode using sweep tubes. Not screen driven (sorry SY) but pentode mode, with plate-to-grid feedback. Using 6JN6 or 6JM6/6GV5 tubes it does about 18W.

How did you manage that? At Vpp= 350Vdc, I'm getting 37W from a PP pair of 6BQ6GTBs, and those have a rated Pd= 12W (Though I run 'em hotter at Pd= 17.5W -- getting them closer to Class A helped the sonics noticeably, and being TV HD's, the spec busting doesn't bother these a bit.) Are you using a particularly shallow loadline for some reason?

I'd push 'em harder. :D
 
I got my PCB in the mail today. Looks great, nice and sturdy and red. I have a couple thoughts floating around now that I'm close to firing up the soldering iron:

So with 450V electrolytics on the board, you can certainly up the steady-state B+ to 400V without going over 450V at cold start. If you go higher you'll need to kludge some series-connected caps in.

So based on the above, would I be OK running a power transformer thats 720VCT as opposed to the 560V of the spec'd Edcor with no other changes to the schematic?

On the mechanical side, does anyone have any thoughts on using chassis mount tube sockets and running a bunch of jumpers down to the PCB, the idea here being to eliminate any strain on the solder joints when installing and removing tubes?

John
 
So based on the above, would I be OK running a power transformer thats 720VCT as opposed to the 560V of the spec'd Edcor with no other changes to the schematic?

That'll be pushing it for the 450V electrolytics,. I would expect you'll hit 500V at startup with 720VCT...

On the mechanical side, does anyone have any thoughts on using chassis mount tube sockets and running a bunch of jumpers down to the PCB, the idea here being to eliminate any strain on the solder joints when installing and removing tubes?

I guess I wouldn't worry too much about stress... there are lots of mounting screws holding the board and the sockets are beefy. Of course, you do get some flexibility if you wire to they sockets... you can use octal tubes, for example.

Pete
 
How did you manage that? At Vpp= 350Vdc, I'm getting 37W from a PP pair of 6BQ6GTBs, and those have a rated Pd= 12W (Though I run 'em hotter at Pd= 17.5W -- getting them closer to Class A helped the sonics noticeably, and being TV HD's, the spec busting doesn't bother these a bit.) Are you using a particularly shallow loadline for some reason?

I'd push 'em harder. :D


Yup, low voltage and shallow load line. The goal was not max power - minimum distortion, and safe & simple, was the idea. 8k plate-to-plate and about 340V, biased @ about 45mA...

Pete
 
So based on the above, would I be OK running a power transformer thats 720VCT as opposed to the 560V of the spec'd Edcor with no other changes to the schematic?

As pete said, a 720 VCT transformer is likely to be close to the limit for a 450 volt cap. Electrolytics are particularly nasty when the vent their goo all over the PC board, or worse just explode. Panasonic makes some 500 volt caps, but I have no idea if any of them will fit the PC board.

Any time you modify someone elses design to go where the original designer never intended or tested, you must be prepared to accept the possibility that it won't work initially, and may not work at all. This means that some experimenting will be required, and the probability of blown parts increases with the amount of deviation from the original design. Any time you increase the power output of a given design, you will reduce the reliability due to the increased voltage and heat dissipation. This does look like a conservative design with some room for reliable improvement, but I don't have my board yet.

I strongly advise builders of my amps (and anybody elses) to build it exactly as it was intended at first, test it, make sure it works, then listen to it for a while before making any modifications. There are two good reasons for this.

First, lets say you build it with all of your cool ideas in place, flip the switch and it all goes BANG. What went wrong? Was one of your ideas the source of the big bang, or was it a simple wiring error. Debugging and troubleshooting this is about the same as debugging a brand new virgin idea since it bears only a mild similarity to the original design.

Second, suppose it makes 1000 watts but sounds like disco played through a cell phone, what did you do to effect these changes. Was it one thing you did, a synergistic combination of many changes, or did the original design just suck. You have no way to tell. What if it sounds magical, how do you duplicate it. What gave it the "magic?"

I plan to build the PC board pretty much as it is intended with one exception. I will stuff the board with the highest voltage rated caps that will fit, just so I don't need to change them later. I plan to test it at the 18 to 20 watt level, listen for a while, then experiment. As they have pounded into our heads over the last 20 years here at work, make ONE change at a time and evaluate it carefully. Document it, and then try something else.

A few years ago I looked at the "Tabor" schematic, reduced it to the simplest possible design, and breadboarded a very simple amplifier using 6AU6's and 6AQ5's. It sounded very nice but I never got the time to take the breadboard design to a complete amplifier. I thought that it would be really cool to build a big one of those. This design is very similar to the "big one" that I envisioned. All of my successes and failures will be posted here, but it looks like it won't happen for a while.
 
I have a 47uF 500v Panasonic that seems to fit, but it would have to be mounted on the other side of the board or it will bump into other components. I'll have to double check that. It's fatter than the spec'd part. I didn't check all the positions cause I'm planning to wire in some big film cap's offboard. I can't find any 10uF 500v electrolytics although 630v film caps can be had at a steep price.

It's tempting to want to push up the voltage, especially with 6GV5's with their 700v anode rating, but I would also expect a whole lot of resistor values needing some tweaking with a B+ change. That and at least Q1 will need to dissipate more heat. A beefier power transformer would be needed too. I'm not sure a Hammond 374BX would have enough juice. I've already got a cranky 270HX that makes a racket. I could probably dig up a 100vct transformer for the bias, but the design matches B+ drift with the bias drift so I may get into trouble there. I've got a 372HX in the Simple SE but I won't mess with that. It's my reliable daily driver.

My goal here is a sort of build with what I've already got challenge. The big question is to borrow the 100w OPT's I've got, and rig up some inefficient speakers for 32 ohms to get 8088ohms on the primaries. Or do I use the little Edcor XPP15-8-8K's with the 96+db efficient full rangers that sit 4 feet from my head.
 
Last edited: