Hello, I want to build "more powerful SE" than 45 tube can give me.
And I found out lots of people are building with the 807 tube. Is this schematic good?
I have only 3200 ohm transformer. Hope that with changing the operating point I will be able to use 807 with this tranformer.
Thank you very much for help. Best regards, Michal
And I found out lots of people are building with the 807 tube. Is this schematic good?
I have only 3200 ohm transformer. Hope that with changing the operating point I will be able to use 807 with this tranformer.
Thank you very much for help. Best regards, Michal
3K2 is too "small" for 807.
Rather this is in 2A3 or 300B territory.
If you have efficient loudspeakers, the 4W (2A3) .. 6-7W ("classical" 300B) power range is usually enough.
Rather this is in 2A3 or 300B territory.
If you have efficient loudspeakers, the 4W (2A3) .. 6-7W ("classical" 300B) power range is usually enough.
Hmm. Do you think it is too small even for 807 in triode? Of course, I can get two paralell, but Im worried about bad effect on the sound, because the tubes will not be identical in any case... that would yield even higher output power.
I can not use 2a3/300b because these tubes are very expensive and have complicated heating.
I want to create something simple and cheap, which yieldshigh output power in SE so people would understand what it is all about. Something like an starting amp in SE
But this is another story, because up to this moment I designed amps just for myself and tweaked them exactly to my preferences. Which is not the case now.
I can not use 2a3/300b because these tubes are very expensive and have complicated heating.
I want to create something simple and cheap, which yieldshigh output power in SE so people would understand what it is all about. Something like an starting amp in SE
But this is another story, because up to this moment I designed amps just for myself and tweaked them exactly to my preferences. Which is not the case now.
3.2k would work with the 807 in tetrode and give you around 7-8W, but you have a LOT of harmonics to deal with.
http://www.tubezone.net/pdf/807stc.pdf
Local feedback around the 807 may be a solution to that, and you are getting pretty close to the RH807 at that point.
http://www.tubezone.net/pdf/807stc.pdf
Local feedback around the 807 may be a solution to that, and you are getting pretty close to the RH807 at that point.
3k2 is perfectly fine for 807 at lower voltages (about 300V plate and 250V screen in tetrode mode or about 300V in triode mode). The power output will not be that great though.
MikyK,
If you want to parallel a pair of 807 tubes, then use individual self bias resistors and individual bypass caps.
You will be surprised how well good 807 tubes match their self bias voltages (and therefore how well they match their cathode currents).
They will match well enough for parallel single ended.
Be sure to use individual grid stopper resistors for the control grids. You are paralleling an RF tube. And, at RF, the cathode bypass caps are paralleling the cathodes (to ground).
A bad match of self bias voltages and cathode currents usually indicates Either most likely a poor/bad 807, Or less likely an unusually strong 807.
Toss out the poor/bad 807, or put it in your tube museum as a show piece.
If your single ended output transformer has a Ultra Linear tap, you should be able to use UL with a B+ of up to 350V.
Some like to connect the 807 cathode(s) self bias resistor(s) and bypass cap(s) to the output transformer secondary. You have to pay attention to the phase of the primary and secondary, or else the feedback is positive, and is not the desired negative feedback.
Positive feedback makes the amplifier gain go up (but will not oscillate because of the primary to secondary step down ratio).
A 3K to 6k to 16, or 8 Ohms output tap, will not oscillate even if the feedback is positive, but you do want the feedback to be negative (the amp gain will go down, versus tying the cathode self bias to ground).
If you use cathode feedback, and the secondary is 4, 8, 16, then use the 16 Ohm tap to the bottom of the self bias network(s)
On another note, I have been using indirect heated Beam Power tubes in UL and Triode wired modes. I do like the ease of powering the filaments.
I do want to do another 2A3 SE amplifier someday, but this time I will use DC for the filaments (more work than I want to do, I have to stop being lazy).
Happy deciding, building, and listening.
If you want to parallel a pair of 807 tubes, then use individual self bias resistors and individual bypass caps.
You will be surprised how well good 807 tubes match their self bias voltages (and therefore how well they match their cathode currents).
They will match well enough for parallel single ended.
Be sure to use individual grid stopper resistors for the control grids. You are paralleling an RF tube. And, at RF, the cathode bypass caps are paralleling the cathodes (to ground).
A bad match of self bias voltages and cathode currents usually indicates Either most likely a poor/bad 807, Or less likely an unusually strong 807.
Toss out the poor/bad 807, or put it in your tube museum as a show piece.
If your single ended output transformer has a Ultra Linear tap, you should be able to use UL with a B+ of up to 350V.
Some like to connect the 807 cathode(s) self bias resistor(s) and bypass cap(s) to the output transformer secondary. You have to pay attention to the phase of the primary and secondary, or else the feedback is positive, and is not the desired negative feedback.
Positive feedback makes the amplifier gain go up (but will not oscillate because of the primary to secondary step down ratio).
A 3K to 6k to 16, or 8 Ohms output tap, will not oscillate even if the feedback is positive, but you do want the feedback to be negative (the amp gain will go down, versus tying the cathode self bias to ground).
If you use cathode feedback, and the secondary is 4, 8, 16, then use the 16 Ohm tap to the bottom of the self bias network(s)
On another note, I have been using indirect heated Beam Power tubes in UL and Triode wired modes. I do like the ease of powering the filaments.
I do want to do another 2A3 SE amplifier someday, but this time I will use DC for the filaments (more work than I want to do, I have to stop being lazy).
Happy deciding, building, and listening.
Last edited:
Some like to do a one-time finished build of a well reviewed circuit. Others like a pile of spaghetti like clip leads and solder tacked wiring , bits of perf board and dead bug construction until the circuit is understood or is tweaked to the point of satisfaction.
What's your tolerance for experimenting?
3200Ω on the RCA 807 triode curves isn't the best load line in the world but it isn't the worst either and with the two mentioned loads placed arbitrarily at 250V/55mA there isn't a whole heck of a lot of difference. If you're not averse to a hybrid circuit or feedback or both there are ways to get more out of the 807. I wonder if Tubelab's UNSET circuit might not be a good way to use it. Or JC Morrison's , or Frank Blöhbaum's or . . . . . . .
Just trying to expand the range of possibilities a bit.
What's your tolerance for experimenting?
3200Ω on the RCA 807 triode curves isn't the best load line in the world but it isn't the worst either and with the two mentioned loads placed arbitrarily at 250V/55mA there isn't a whole heck of a lot of difference. If you're not averse to a hybrid circuit or feedback or both there are ways to get more out of the 807. I wonder if Tubelab's UNSET circuit might not be a good way to use it. Or JC Morrison's , or Frank Blöhbaum's or . . . . . . .
Just trying to expand the range of possibilities a bit.
Attachments
Thanks for the curves and because I see the positive lines:
what about to:
make high voltage supply (about 450V)
use direct coupling, so voltage accross 807 will be roughly 300V
fry it with current so it has as much dissipation as possible
use cathode winding feedback as suggested by one member here (thanks for that) to linearize it a little bit?
I have previous experiences with direct coupling. My 2x 4P1L SE DC amp sounded just phenomenal.
what about to:
make high voltage supply (about 450V)
use direct coupling, so voltage accross 807 will be roughly 300V
fry it with current so it has as much dissipation as possible
use cathode winding feedback as suggested by one member here (thanks for that) to linearize it a little bit?
I have previous experiences with direct coupling. My 2x 4P1L SE DC amp sounded just phenomenal.
Last edited:
If you want to observe the specified max plate dissipation of 25 Watts then at Vp=300V you get Ip=83mA.
Looking at the curves it's pretty asymmetric. Still, I'm one of those clip lead spaghetti builders I mentioned above (and have an amp covering half a workbench for the past couple of years while I try everything I can think of to make it work the way I want) so I'd say "Why not?" , but definitely not as a finished build. Not yet anyway.
Do you have any idea how much power your output transformer is spec'd for? How much current the primary can take?
Looking at the curves it's pretty asymmetric. Still, I'm one of those clip lead spaghetti builders I mentioned above (and have an amp covering half a workbench for the past couple of years while I try everything I can think of to make it work the way I want) so I'd say "Why not?" , but definitely not as a finished build. Not yet anyway.
Do you have any idea how much power your output transformer is spec'd for? How much current the primary can take?
Attachments
My output transformers are rated for 100mA current ad the manufacturer specifies 40W max power.
Which I think is overkill, but they sure can work to 15 watts maybe more. They are like 4kgs each.
Which I think is overkill, but they sure can work to 15 watts maybe more. They are like 4kgs each.
I guess it depends on your way of doing things. If you want a more certain road to a good amp then Michael's suggested circuit is likely to be a good answer. He certainly has a lot of experience and knows what he's talking about. Though I think he'd probably agree that personal taste might vary from person to person.
My tendency is to put the emphasis on your point of interest and learning from it. Trusting your intuition and following your nose I guess you could say.
If the idea of the 807 and /or direct coupling is holding some attention then by all means give it a go. Just don't spend a lot of time on making it beautiful or permanent.
Well, I could add a bit but I don't know you so will wait to see if you have any more to say.
My tendency is to put the emphasis on your point of interest and learning from it. Trusting your intuition and following your nose I guess you could say.
If the idea of the 807 and /or direct coupling is holding some attention then by all means give it a go. Just don't spend a lot of time on making it beautiful or permanent.
Well, I could add a bit but I don't know you so will wait to see if you have any more to say.
Hmm. I got used to the sound of 4P1L direct coupled amp, which is the best amp Ive ever made. The sound was so pure, so revealing in terms of detail, simply amazing.
So from that point I started hating coupling capacitors in the circuit. I tried 3 stage SE and didnt like the sound. I was inspired by the schematic you offered me but I would use high only one stage, something like E180F in triode mode or simillar which has low output impedance (but probably not so low as has the cathode follower).
I simply want to search tube with normal socket (no magnoval, etc), with high anode dissipation and replaceable. And triode curves linear as possible 😀 Im currently using 6P3S-E in my breadboard amp. Not ideal choice but easily replaceable by 6L6 which is still in production. I think at normal price tag I will not be able to find something better than 6L6.
I like 807 tube because it is large, which means it will radiate heat better than smaller 6L6. Also 25W dissipation is better than 20W of the 6P3S. But the triode curves are not much linear...
As I said I can linearize it with feedback from the secondary winding, but I havent any experience with this design. Maybe I should be less lazy and take the risk that it wil not sound good and just try it 😀
So from that point I started hating coupling capacitors in the circuit. I tried 3 stage SE and didnt like the sound. I was inspired by the schematic you offered me but I would use high only one stage, something like E180F in triode mode or simillar which has low output impedance (but probably not so low as has the cathode follower).
I simply want to search tube with normal socket (no magnoval, etc), with high anode dissipation and replaceable. And triode curves linear as possible 😀 Im currently using 6P3S-E in my breadboard amp. Not ideal choice but easily replaceable by 6L6 which is still in production. I think at normal price tag I will not be able to find something better than 6L6.
I like 807 tube because it is large, which means it will radiate heat better than smaller 6L6. Also 25W dissipation is better than 20W of the 6P3S. But the triode curves are not much linear...
As I said I can linearize it with feedback from the secondary winding, but I havent any experience with this design. Maybe I should be less lazy and take the risk that it wil not sound good and just try it 😀
Yup, maybe you should.😀 If it doesn't sound good you'll at least have the sockets and power supply set up to start making it sound good. You clearly know more about this than your first post indicated.
I'm presently very interested in feedback.
I've built a lot of circuits using feedback that didn't sound good and came to equate feedback with bad sound. I've since learned it's way more complex than that and that there is something more general in the way the designer went about incorporating it into their design.
I've been especially interested in the work that Frank Blöhbaum wrote about in his Linear Audio (Vol.6 and Vol.8) articles on his Multiplied Transconductance Amplifier. He uses a hybrid input circuit to get super high gain and then drops most of it with feedback to end up with a harmonic spectrum almost entirely 2nd order. I've built two of his circuits and I think they are worth trying.
Look also at Spread Spectrum's work in a similar vein.
I mentioned Tubelab's UNSET , and JC Morrison's stuff before too. Worth reading.
Also interesting is Rod Coleman's contention that the cadence of steadily diminishing harmonics is what makes for the most listenable sound. (My paraphrase, not his exact words.) So I hope to learn enough about feedback to try shaping with it.
I have a couple of old 807's in my drawer I've been wondering what to do with. Please keep posting on where you go with yours.
. . . . .
. . . . . . . . .
. . . . . . . . . . . .
Member
Joined 2009
Paid Member
what about lowering plate impedance of the output tube by using feedback from the plate of the 807 (in tetrode mode) to cathode of input / gain tube so that it can drive the 3k2 OPT. The input tube needs to produce high gain, such as 6f12p russian tube with mu=100 butstill decent output impedance.
Want more powerful than 45 and still have GREAT sound?
Go for 2a3. Get cheap china 2a3 tubes for it. Easy. Problem solved. Want to go a bit fancy? Get a matched pair of sovtek 2a3's. They're good too.
PLUS your 3k2 OPT will be perfect for 2a3. This is a no-brainer decision.
Also. learn to draw your own load lines. 😉
The 807 in triode is NOT linear. Looks nice though.... but look at those triode curves.. sheesh... not worth "experimenting" with (just my opinion).
Ian
Go for 2a3. Get cheap china 2a3 tubes for it. Easy. Problem solved. Want to go a bit fancy? Get a matched pair of sovtek 2a3's. They're good too.
PLUS your 3k2 OPT will be perfect for 2a3. This is a no-brainer decision.
Also. learn to draw your own load lines. 😉
The 807 in triode is NOT linear. Looks nice though.... but look at those triode curves.. sheesh... not worth "experimenting" with (just my opinion).
Ian
Last edited:
E180F can drive 2a3 - I use gyrator load. Surprisingly good results and not expensive. Sure, C3g is a bit better but its also more expensive now... E280F is an excellent choice too.
If you are already building DHT amps then why go for triode strapped tetrodes?
Also, feedback won't fix those non-linear triode curves. nope. Feedback can definitely helpwith harmonic distortion but not with non-linearity.
But I have no doubt the 807 amp sounds nice. but I bet your 4P1L will beat it. 🙂
Ian
If you are already building DHT amps then why go for triode strapped tetrodes?
Also, feedback won't fix those non-linear triode curves. nope. Feedback can definitely helpwith harmonic distortion but not with non-linearity.
But I have no doubt the 807 amp sounds nice. but I bet your 4P1L will beat it. 🙂
Ian
Last edited: