With a big tip O' the hat to Pete Millett I have jumped feet first into designing a rather simple Push Pull amp using the 829B/3E39 transmitting tube.
First, A little background as to why such a "bizarre" tube choice for a relative "beginner" to tube audio.
My first experience with Tube Audio came purely by accident after having found an old Grunow "Console" Radio in a junk pile. I liked the look of it and wanted to make it work. I found the schematic online and played with it for a while. The "sound" seemed so much better even in Mono than anything I had heard that I became intrigued with tubes.
I started picking up tubes etc at estate sales and built a couple of SE amp using proven designs off the net and "junk box" parts.
I came across an old US Army Signal Corps transmitter with (2) 829B's and a beefy power tranny in another trashpile. Having grown up watching M.A.S.H. it was so much like "Radar's" setup I had to take it home.
fast forward about 2 years and I have these 829B's along with some nice Shielded sockets and I came across Pete Millett's 829B SE and really liked that idea. Since SE Iron is relatively expensive and hard to find in "trashpiles" I held off building anything till I could really design something PP.
So I now have my hands on a set of decent PP output trannies capable of the idle current that I want to use with around 7800R P to P.
Time to start "designing".
The overall concept is plain vanilla since the choice of output tube is interesting enough. One half 6SN7 input/gain stage into the second half "split load" phase splitter. Then a 6CG7/6FQ7 will drive the 829B grids.
I am going with P to P of 7800 ohms into 8.
A B+ of 300V, Cathode Bias of about -30V for ~80mA per tube idle current.
Any comments as to whether or not I am sane?
First, A little background as to why such a "bizarre" tube choice for a relative "beginner" to tube audio.
My first experience with Tube Audio came purely by accident after having found an old Grunow "Console" Radio in a junk pile. I liked the look of it and wanted to make it work. I found the schematic online and played with it for a while. The "sound" seemed so much better even in Mono than anything I had heard that I became intrigued with tubes.
I started picking up tubes etc at estate sales and built a couple of SE amp using proven designs off the net and "junk box" parts.
I came across an old US Army Signal Corps transmitter with (2) 829B's and a beefy power tranny in another trashpile. Having grown up watching M.A.S.H. it was so much like "Radar's" setup I had to take it home.
fast forward about 2 years and I have these 829B's along with some nice Shielded sockets and I came across Pete Millett's 829B SE and really liked that idea. Since SE Iron is relatively expensive and hard to find in "trashpiles" I held off building anything till I could really design something PP.
So I now have my hands on a set of decent PP output trannies capable of the idle current that I want to use with around 7800R P to P.
Time to start "designing".
The overall concept is plain vanilla since the choice of output tube is interesting enough. One half 6SN7 input/gain stage into the second half "split load" phase splitter. Then a 6CG7/6FQ7 will drive the 829B grids.
I am going with P to P of 7800 ohms into 8.
A B+ of 300V, Cathode Bias of about -30V for ~80mA per tube idle current.
Any comments as to whether or not I am sane?
I forgot to mention that along the way I have amassed a collection of (7) 829B's all RCA from about the same years in production.
I happened to pick up a lot of boxes of "ham radio" stuff from Flea Markets and Estate Sales. Since I really have no tester to test these I can verify that the filaments are good, I figured it's a decent chance I will have (4) good ones out of 7.
Pentode would give me much more power but looking at the "triode" curves it has pretty good linearity and I will try the "triode sound" for now.
Snip out the G2 connection to the plate and supply it with 200V or so in the future is no problem if I want to go pentode.
I happened to pick up a lot of boxes of "ham radio" stuff from Flea Markets and Estate Sales. Since I really have no tester to test these I can verify that the filaments are good, I figured it's a decent chance I will have (4) good ones out of 7.
Pentode would give me much more power but looking at the "triode" curves it has pretty good linearity and I will try the "triode sound" for now.
Snip out the G2 connection to the plate and supply it with 200V or so in the future is no problem if I want to go pentode.
I'm not a big fan of LTP.
I should also have mentioned the "Goals" of this amp.
1: Use the 829B along with any other "funky" tubes (going for a "steampunk" theme. Brass, Black Wrinkle finish, Antique look)
2: About 15 watts with 1% distortion or so.
3: Be as weird looking as possible
Now a Powersupply is needed.
This this is starting to become ridiculously over complex. Having lots of funky tubes etc. So might as well continue along that theme.
I am looking at 2 power transformers, one for the 829B's w/ Parallel 5U4GBs a choke etc.
The other for the input, splitter and driver stage.
In keeping with the "too many tubes" theme I am thinking about a VR tube regulated supply.
Heres the question: (2) 0D3's in series gets me the 300V. Can I then add an 0C3 in series with those and the 6.8mA load of the 27's to get my 405V?
If so how would you go about it?
This this is starting to become ridiculously over complex. Having lots of funky tubes etc. So might as well continue along that theme.
I am looking at 2 power transformers, one for the 829B's w/ Parallel 5U4GBs a choke etc.
The other for the input, splitter and driver stage.
In keeping with the "too many tubes" theme I am thinking about a VR tube regulated supply.
Heres the question: (2) 0D3's in series gets me the 300V. Can I then add an 0C3 in series with those and the 6.8mA load of the 27's to get my 405V?
If so how would you go about it?
At this point I am looking at a total of about 16 Tubes! LOL for a measly 15WPC stereo!
The power company is going to LOVE me!
The heater filaments alone are dissipating 2 1/2 times the power output!
As crazy as all this seems, I am kind of liking the idea? Certainly would make an interesting "conversation piece", not very efficient but worth a shot.
The power company is going to LOVE me!
The heater filaments alone are dissipating 2 1/2 times the power output!
As crazy as all this seems, I am kind of liking the idea? Certainly would make an interesting "conversation piece", not very efficient but worth a shot.
I have built an 829B push-pull amp, although it is currently wired in UL. Do note that the 829B has unusually high miller capacitance, so you will want a driver that has very, very low output impedance, unless you don't need your treble. (Notice Millett uses cathode followers).
If you are starting from scratch, use an interstage transformer to do the phase splitting, or at least to reduce the output impedance of the driver.
If you are starting from scratch, use an interstage transformer to do the phase splitting, or at least to reduce the output impedance of the driver.
Yes, The Miller is VERY high, I have actually sought advice from Pete Millett (an AWESOME Guy BTW).
The modified "Williamson" style splitter/driver arrangement was done to bring the output Z of the drivers as low as possible. Using a low Rp driver (6SN7) and low plate load and bypassing the Rk gives me about 7K Zout.(I hope)
Millett used the follower since he was only running 22V Peak to Peak on the grids. I need about TWICE that. Rather than run an additional stage I just brought down the Zout.
I have a PC board for Millett Tube Xover and eventually this amp will power the mid-range of a "Tri-Amped" system. So if I give up a little response on the high end I am OK with it for now.
The modified "Williamson" style splitter/driver arrangement was done to bring the output Z of the drivers as low as possible. Using a low Rp driver (6SN7) and low plate load and bypassing the Rk gives me about 7K Zout.(I hope)
Millett used the follower since he was only running 22V Peak to Peak on the grids. I need about TWICE that. Rather than run an additional stage I just brought down the Zout.
I have a PC board for Millett Tube Xover and eventually this amp will power the mid-range of a "Tri-Amped" system. So if I give up a little response on the high end I am OK with it for now.
Arne,
They don't NEED it, it is just that the curves I have been playing with for 2 years. (thanx to Tom Schlangen and Paul LeClercq-via Pete Millett) have shown me that anything less than 8K and you get higher levels of distortion.
Pete used 5K in his SE design and got 4% with only 3.5Watts.
Since this is a tube and application that is rarely used there is not much info available. My calculations at about 8K8 PtoP using this design give me about 2% distortion mostly (if not all) odd order harmonics.
Arne,
Are you planning to use these tubes?
I once had a chinese amp with 829 (Audioromy), and that little thing impressed me, with only 2 x 829B in push-pull. I think it had 20-25W, but with a lot of "grunt", and played bass on "not tubefriendly" speakers! (better than a Dynaco ST-70).
Arne K
Arne K
Attachments
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.