Attached is the preliminary schematic for an amp I'm naming after an old school colleague. The schematic came out of a discussion on this list regarding some old Baldwin iron I've been characterizing. Strange to relate, none of that iron will be used in this amp. Power and output iron will come from the guts of a Fisher 400 I picked up earlier this year from a local surplus store (just the iron). I'll start out using some small toroidal power transformers with dual primaries for the interstage autotransformers. I may award the amp with some better iron if the sound warrants it. Another possibility would be to use the primary windings of some cheap P-P transformers like the Edcor XPP10-8-8k.
The schematic looks a little busy because I've included the bias strings. The output stage uses a mixture of fixed and cathode bias. The amount of cathode bias is kept relatively small to allow me to squeeze out some extra watts, and will be used to trim the output stage balance. Those who insist on absolute simplicity could try omitting the source follower.
On the input side, one half of the 6BN11 dual pentode is shown, implying that the other half would be used in the other channel. There is a shield between the two pentodes in the tubes that I have (pretty much all GE manufacture, though of various brands), so I don't expect a problem with crosstalk between channels. If you feel queasy about that, I suspect a pair of 6EW6 or 6EJ7 would work as well, as they have similar transconductance for the same amount of plate current. I find that using the dual pentode makes the top panel layout more pleasing and a little less cluttered.
If you do use the 6BN11, remember to connect the suppressor grids to the cathode at the socket - this connection is not shown in the schematic, and the suppressor grids are brought out in the 6BN11 rather than being connected internally to the cathode.
The schematic looks a little busy because I've included the bias strings. The output stage uses a mixture of fixed and cathode bias. The amount of cathode bias is kept relatively small to allow me to squeeze out some extra watts, and will be used to trim the output stage balance. Those who insist on absolute simplicity could try omitting the source follower.
On the input side, one half of the 6BN11 dual pentode is shown, implying that the other half would be used in the other channel. There is a shield between the two pentodes in the tubes that I have (pretty much all GE manufacture, though of various brands), so I don't expect a problem with crosstalk between channels. If you feel queasy about that, I suspect a pair of 6EW6 or 6EJ7 would work as well, as they have similar transconductance for the same amount of plate current. I find that using the dual pentode makes the top panel layout more pleasing and a little less cluttered.
If you do use the 6BN11, remember to connect the suppressor grids to the cathode at the socket - this connection is not shown in the schematic, and the suppressor grids are brought out in the 6BN11 rather than being connected internally to the cathode.
Attachments
The pain, the pain. All those wasted afternoons watching Lost in Space reruns. 😀
Actually this really happened. One semester I dropped all my courses and went to the bursar’s office and asked for all my tuition back. To my surprise they did it. Picture that happening today.
Actually this really happened. One semester I dropped all my courses and went to the bursar’s office and asked for all my tuition back. To my surprise they did it. Picture that happening today.
Last edited:
Upon reflection, this setup, with some changes in the output bias, may be just the ticket for kicking the living snot out of some of those stubby little 17-18W horizontal deflection tubes in screen driven mode. The P-P topology is just what is needed to take advantage of their high peak current capability. An all-Compactron amp would look pretty neat. Having said that, though, I'd probably want output iron with a little hotter turns ratio for that experiment, so I'll stick to the 6P3S-E for this go-round.
Scott - it's interesting to hear the real story - I always thought it was a play on the name, plus undergraduate whimsy... BTW, I can't even imagine paying the tuition these days, much less trying to get it back.
I had a friend in college who did pretty much the same thing with "Space 1999" - but unlike you he did not have the forsight to drop all of his classes and managed to pull a GPA of just 0.2 for the quarter. I still remember the look on his Dad's face as he helped him pack up after he flunked out.. 😱
My GPA was based on something somewhat resembling the cosine waveform with significant positive offset and some pretty big peaks AND valleys.. 😀
Tuition cost these days is pretty appalling..
Last edited:
No zero's for me one "D" (pass no credit) from Millie Desselhaus herself with the promise that I would never be a physicist. A brilliant wonderful woman BTW, I didn't appriciate it at the time.
Dhaen -
I did some rough back-of-the-envelope calculations, and it looks like the amp has only 7-8 dB of feedback. The current value of the feedback cap (C3) is probably too small - I would start with around 1000- 1500 pF and adjust depending on the square wave response.
As for stability, I know from experience that the output transformers are the main limiters on the bandwidth of the P-P circuits I've built thus far. The Fisher transformers are pretty good, but not exceptional. A lot will depend on where the extra poles from the autotransformer are located. If the autotransformer has similar to or wider bandwidth than the output transformer, then there is a chance to roll off the gain gracefully before trouble starts. I'm going to see what I can accomplish with the cheap autotransformers first, as I already have everything in hand - there is always the option of better iron, and the Electra-Print autoformers are only $30 a pop.
I did some rough back-of-the-envelope calculations, and it looks like the amp has only 7-8 dB of feedback. The current value of the feedback cap (C3) is probably too small - I would start with around 1000- 1500 pF and adjust depending on the square wave response.
As for stability, I know from experience that the output transformers are the main limiters on the bandwidth of the P-P circuits I've built thus far. The Fisher transformers are pretty good, but not exceptional. A lot will depend on where the extra poles from the autotransformer are located. If the autotransformer has similar to or wider bandwidth than the output transformer, then there is a chance to roll off the gain gracefully before trouble starts. I'm going to see what I can accomplish with the cheap autotransformers first, as I already have everything in hand - there is always the option of better iron, and the Electra-Print autoformers are only $30 a pop.
Yes, my mental estimator suggests that less than 10dB of feedback is likely to be safe with 2 inductive loaded stages.
My reason for asking is that I really like iron-coupling. With a stiff bias supply it allows the possibility of favourite class: A2...
My reason for asking is that I really like iron-coupling. With a stiff bias supply it allows the possibility of favourite class: A2...
I have an old Delco car amplifier that uses vibrator power supply running a pair of 12V6 pushpull. It uses the center tapped input choke as the splitter and sounds just awesome. Had to replace the vibrator circuit using a 24V transformer into the stock transformer. It uses a 0v4 (not totally sure its not where im at atm) gas rectifier. I was amazed at how good this little thing sounded.
When i used toroidals as autotransformers there were huge capacitive losses to be overcome. The amp sounded nice when I pushed 30mA of current through the paralleled 5687 driver triode - but the high frequency roll off was still little better than 20khz. So don't expect to much.
Br Cornelius
An interesting design indeed. With the drive possibilities the design presents, I would suggest a more capable screen supply for the output stage. That would maximize power output, minimize distortion, and make the most of the autotransformer's potential -- whatever class you end up driving the output stage under.
Jerluwoo -- That was most likely an OZ4 in your Delco radio. They were famous for using that tube.
Dave
Jerluwoo -- That was most likely an OZ4 in your Delco radio. They were famous for using that tube.
Dave
I'm hoping that the source follower will add enough authority to bull its way past the transformer capacitance. If not, I have several pairs of cheap P-P transformers, the primaries of which could be pressed into service as autotransformers.
One of the things that has stopped me so far from proceeding on the "Bursar" is the thick steel panels on the Hammond case I'm using for everything. I'm not looking forward to attacking the thick steel with a Unibit to make holes for the tube sockets.
The Fisher 400 iron set aside for this project had a fair amount of scratches, tired black paint, and a touch of rust. I took all the old paint and rust off that I could using a drill press with a wire brush attachment - as long as you can get to the paint, it comes right off, leaving a nice burnished finish. If you're a stickler, one of the tooth brush-sized wire brushes (or some paint remover) could be used to dig out the paint left in the crevices. The hardware (the screws are always rusty) was replaced with nice zinc-plated stuff that goes well with the gray. Stainless steel looks even better, if you can get it - that's what I used on my "Kingfisher" amp.
The output iron (inside the case) got a coat of Rust-oleum gray primer. The power transformer (outside) got some primer, then some cast-iron gray metallic automotive touch-up paint, which should go nicely with the dark blue metallic paint that will grace the case and top/bottom panels. If you try this, get as much of the old paint off the end bells and lams as possible, as the solvent in the touch-up paint reaches through the primer and causes any of the old paint left over to wrinkle up. Not a disaster, but annoying for those really finicky. It still looks loads better than anything Hammond ships these days.
The screws for the power iron are still a little long, so I'll hack them off some and use acorn nuts for that well-dressed look.
Once all the holes are hacked into the case, the wiring shouldn't take all that long. I'm starting off with using a couple of small dual primary toroids for the output drive autotransformers, hoping that the source follower I'm using to drive them will bull its way past any capacitance. If not, I have a pair of small Edcor P-P transformers that will be pressed into service.
The Fisher 400 iron set aside for this project had a fair amount of scratches, tired black paint, and a touch of rust. I took all the old paint and rust off that I could using a drill press with a wire brush attachment - as long as you can get to the paint, it comes right off, leaving a nice burnished finish. If you're a stickler, one of the tooth brush-sized wire brushes (or some paint remover) could be used to dig out the paint left in the crevices. The hardware (the screws are always rusty) was replaced with nice zinc-plated stuff that goes well with the gray. Stainless steel looks even better, if you can get it - that's what I used on my "Kingfisher" amp.
The output iron (inside the case) got a coat of Rust-oleum gray primer. The power transformer (outside) got some primer, then some cast-iron gray metallic automotive touch-up paint, which should go nicely with the dark blue metallic paint that will grace the case and top/bottom panels. If you try this, get as much of the old paint off the end bells and lams as possible, as the solvent in the touch-up paint reaches through the primer and causes any of the old paint left over to wrinkle up. Not a disaster, but annoying for those really finicky. It still looks loads better than anything Hammond ships these days.
The screws for the power iron are still a little long, so I'll hack them off some and use acorn nuts for that well-dressed look.
Once all the holes are hacked into the case, the wiring shouldn't take all that long. I'm starting off with using a couple of small dual primary toroids for the output drive autotransformers, hoping that the source follower I'm using to drive them will bull its way past any capacitance. If not, I have a pair of small Edcor P-P transformers that will be pressed into service.
I've used my unibits to cut up to 1/8" mild steel 😱 They are up to the task
In a drill press at about 200 RPM with really good clamping and lots of cutting oil.
Which particular toroid are you planning to use? I tried the low impedance drive and found that it resulted in massive peaking above 20KHz. Something to look for...
The idea of using an OPT for a splitting autoformer is interesting. Could be a use for those 600 ohm line output OPTs.
Cheers!
The peaking might be tameable with a loading/damping RC network placed between the grids. This is a common dodge used to fix a peaky transformer
The toroids I'm contemplating using were picked up at a local surplus outlet. They have epoxy-filled centers, which makes the mounting convenient. I also have some smallish Magnavox P-P output transformers that might be useful.
The toroids I'm contemplating using were picked up at a local surplus outlet. They have epoxy-filled centers, which makes the mounting convenient. I also have some smallish Magnavox P-P output transformers that might be useful.
Last edited:
It looks like your MOSFET follower does not have enough voltage to power it for an adequate swing.
Why? I need only need around +/- 20V to drive the output tubes into full conduction. The follower is centered at around 100V, and will have about 200V at the top MOSFET when I'm finished. This may require some adjustment of the dropper string. The resistor determing the sink current for the bottom MOSFET of the follower is a place-holder value, and will be initially set for somewhere between 5-10ma for first experiments.
Hi Wrenchone-
Thanks for telling me about this thread at BA. When I get back into my workshop, I would like to try this circuit. It could be a good candidate for my burning man amplifier. Though I still would like to use triodes, I might just go with beam tubes for the amp's first iteration.
Cheers- Kent
Thanks for telling me about this thread at BA. When I get back into my workshop, I would like to try this circuit. It could be a good candidate for my burning man amplifier. Though I still would like to use triodes, I might just go with beam tubes for the amp's first iteration.
Cheers- Kent
- Status
- Not open for further replies.