Hi all,
I'm thimking of starting a new project using a pair of 6L6's in push pull, triode strapped into a 5k (A-A) output transformer.
Does anyone have any suggestions as to driver/splitter stages I might use, and likely tubes for the circuit. I have been looking at 6SL7,6SN7 and 12AX7,12AU7 stages but am wide open to other suggestions, as long as I can freely get the tubes at reasonable cost, and the circuit isn't too complex. Links to possible schematics would be much appriciated too.
I was looking at modifying the original williamson circuit to suit but I presume someone must have already done this who might show me a lead.
Are there advantages to adjustable self bias (williamson) over the more common fixed self bias (mullard/leak) of the output tubes?
Any help or comments appreciated.
Thanks in advance, Bob.
I'm thimking of starting a new project using a pair of 6L6's in push pull, triode strapped into a 5k (A-A) output transformer.
Does anyone have any suggestions as to driver/splitter stages I might use, and likely tubes for the circuit. I have been looking at 6SL7,6SN7 and 12AX7,12AU7 stages but am wide open to other suggestions, as long as I can freely get the tubes at reasonable cost, and the circuit isn't too complex. Links to possible schematics would be much appriciated too.
I was looking at modifying the original williamson circuit to suit but I presume someone must have already done this who might show me a lead.
Are there advantages to adjustable self bias (williamson) over the more common fixed self bias (mullard/leak) of the output tubes?
Any help or comments appreciated.
Thanks in advance, Bob.
Hi Bob,
6CG7 is another candidate for a phase splitter 🙂
I experimented with a triode strapped 6L6 fair exstensive (SE mode though) and found fixed bias the only way to seriously go. Clearer sound and more power.
6CG7 is another candidate for a phase splitter 🙂
I experimented with a triode strapped 6L6 fair exstensive (SE mode though) and found fixed bias the only way to seriously go. Clearer sound and more power.
With the Hafler mod (staggering the low frequency rolloffs), a Williamson is a good classic choice. I'm in agreement with Geek, fixed bias would also be an excellent mod to that circuit. It will require the driver stage to work a bit harder on peaks, but that's not a terrible tradeoff for the increased headroom.
For crystal clear sonics.........
I design neutral sounding tube amps that let the music flow. If you like fuzzy & warm sonics, many other options are available than stated below.
I like a direct coupled preamp tube to twin-tail phase splitter design. A 76 tube or 6P5GT to 6SN7 is a great combination. The best sounding tubes in a few amps I made are Tung-Sol 6P5GT to a Sylvania 6SN7 'T' shape plate tube.
Any 6P5GT with round plates are good sounding tubes, hard to go wrong here. It is quite a different story with the 6SN7. 1960s production 6SN7s are not great sounding tubes. I also had no luck with any GE 6SN7 or RCA no matter what the years of manufacture. Different circuits does bring forth different results.
The Sylvania 6SN7GTA with 'Y' plates & clear micas are also quite nice sounding. The Sylvania 6SN7WGT also sound good. It is available in either 'T' or 'Y' plate with the 'T' plate being more common.
One can also use a 6J5GT (not the metal type tubes) to replace the somewhat rare 6P5GT with a slightly different resistor circuit values.
As for 6CG7, one (and only one) stellar performer are the Raytheon 6CG7 with black plates & center shield. Unfortunately, these are quite rare. Thus, the 6SN7 is a better bet due to availibility. Many other 6CG7s I tried just do not perform well sonically with slight smearing in the audio.
I consider the AuriCap the least expensive coupling cap that will not smear (un-do) the clear sonics of the above design with extensive tube rolling. An inexpensive Sprague 'orange-drop' will quickly un-do the clear sonics.
My 2 cents.
I design neutral sounding tube amps that let the music flow. If you like fuzzy & warm sonics, many other options are available than stated below.
I like a direct coupled preamp tube to twin-tail phase splitter design. A 76 tube or 6P5GT to 6SN7 is a great combination. The best sounding tubes in a few amps I made are Tung-Sol 6P5GT to a Sylvania 6SN7 'T' shape plate tube.
Any 6P5GT with round plates are good sounding tubes, hard to go wrong here. It is quite a different story with the 6SN7. 1960s production 6SN7s are not great sounding tubes. I also had no luck with any GE 6SN7 or RCA no matter what the years of manufacture. Different circuits does bring forth different results.
The Sylvania 6SN7GTA with 'Y' plates & clear micas are also quite nice sounding. The Sylvania 6SN7WGT also sound good. It is available in either 'T' or 'Y' plate with the 'T' plate being more common.
One can also use a 6J5GT (not the metal type tubes) to replace the somewhat rare 6P5GT with a slightly different resistor circuit values.
As for 6CG7, one (and only one) stellar performer are the Raytheon 6CG7 with black plates & center shield. Unfortunately, these are quite rare. Thus, the 6SN7 is a better bet due to availibility. Many other 6CG7s I tried just do not perform well sonically with slight smearing in the audio.
I consider the AuriCap the least expensive coupling cap that will not smear (un-do) the clear sonics of the above design with extensive tube rolling. An inexpensive Sprague 'orange-drop' will quickly un-do the clear sonics.
My 2 cents.
The Williamson is a good starting point.....
You can then change things from there....based on your taste..
I am not a big fan of the cathodyne phase splitter... Just not my thing.... I prefer Long-Tailed cathode coupled .....with bootstrapping...this way you maintain balance by way of the feedback.... or you can put a CCS in the tail....whatever you prefer....
I don't agree with Amperex about the 6SN7 tubes.... With carefull circuit design the 6SN7 will perform exceptionally well...but on the other hand will sound miserable if not executed well....
I do prefer direct coupled as much as possible....since it eliminates influence of the caps....no matter how expensive the caps....
Chris
You can then change things from there....based on your taste..
I am not a big fan of the cathodyne phase splitter... Just not my thing.... I prefer Long-Tailed cathode coupled .....with bootstrapping...this way you maintain balance by way of the feedback.... or you can put a CCS in the tail....whatever you prefer....
I don't agree with Amperex about the 6SN7 tubes.... With carefull circuit design the 6SN7 will perform exceptionally well...but on the other hand will sound miserable if not executed well....
I do prefer direct coupled as much as possible....since it eliminates influence of the caps....no matter how expensive the caps....
Chris
I'm thimking of starting a new project using a pair of 6L6's in push pull, triode strapped into a 5k (A-A) output transformer.
OK, that sounds like a good idea. I've been giving some thought to using 6V6's, possibly in triode mode, for a future project.
As for phase splitters, your best choice is the LTP (differential amp) with a current source in the tail. The 6SL7 would be very good in that capacity. The next best choice would be the cathodyne. The A Number One problem with other active splitters is asymmetrical harmonic distortion. Avoid these, and a really good single plate-to-PP grids interstage would be better than any splitter that's not a LTP or cathodyne. 6SN7's and 12AX7A's are also useful, but I would avoid the 12AU7A. That's more of a digital VT, and isn't so good for analog amplification as its characteristic is highly non-linear. For medium-u, dual triodes you'd be better off going with a 6FQ7 or a 6CG7: same u as the 12AU7A, but characteristics that look to be far more linear.
As for drivers, I'd suggest using SRPP's to drive each 6L6. (Again, best choice for that would be the 6FQ7, 6CG7, 6BQ7A, or the 6SN7. The SRPP is more of a power amp type circuit, and you will need some drive capability to charge up the input capacitance to increase the slew rate. Triode mode, thanks to Miller Effect, will have a larger input capacitance. If you're going to push the 6L6's into class AB(2), you might consider using a power MOSFET as a common drain driver. The MOSFET can source more current, has higher gain than any VT, and is linear when operated with heavy feedback (as a common drain naturally provides).
OK, that sounds like a good idea. I've been giving some thought to using 6V6's, possibly in triode mode, for a future project.
As for phase splitters, your best choice is the LTP (differential amp) with a current source in the tail. The 6SL7 would be very good in that capacity. The next best choice would be the cathodyne. The A Number One problem with other active splitters is asymmetrical harmonic distortion. Avoid these, and a really good single plate-to-PP grids interstage would be better than any splitter that's not a LTP or cathodyne. 6SN7's and 12AX7A's are also useful, but I would avoid the 12AU7A. That's more of a digital VT, and isn't so good for analog amplification as its characteristic is highly non-linear. For medium-u, dual triodes you'd be better off going with a 6FQ7 or a 6CG7: same u as the 12AU7A, but characteristics that look to be far more linear.
As for drivers, I'd suggest using SRPP's to drive each 6L6. (Again, best choice for that would be the 6FQ7, 6CG7, 6BQ7A, or the 6SN7. The SRPP is more of a power amp type circuit, and you will need some drive capability to charge up the input capacitance to increase the slew rate. Triode mode, thanks to Miller Effect, will have a larger input capacitance. If you're going to push the 6L6's into class AB(2), you might consider using a power MOSFET as a common drain driver. The MOSFET can source more current, has higher gain than any VT, and is linear when operated with heavy feedback (as a common drain naturally provides).
I'd go for a differential driver for the front end and what I'd use would depend on what I had to hand and more importantly the gain required. Choices are (moderated gain) ECC99 and 6H6P and for higher gain Allen Wright's cascode front end off the PP1C or EL84's in pentode a la Thorsten's Legacy amp.
I've built them all and like them for different reasons but I'd probably go with the PP1C as you're at about the correct supply voltage for it already and it has enough gain in one stage to drive the 6L6's to clipping from a CDP source.
I've built them all and like them for different reasons but I'd probably go with the PP1C as you're at about the correct supply voltage for it already and it has enough gain in one stage to drive the 6L6's to clipping from a CDP source.
My own PP triode amp is all differential, starting with a 6SL7 LTP phase splitter with a 6AU6 pentode as a CCS in the tail.
The splitter is coupled to a 6SN7 differential driver via a step network and plate-to-plate negative feedback. This arrangement gives excellent bass and balance, with sufficient stability to apply global NFB and without the headache of DC imbalance that can be caused by direct coupling with imperfectly matched tubes.
The OP stage is triode-connected EL34s but there's no reason why it couldn't use 6L6s. Sound is great. Hum and noise are inaudible. A schematic is attached FYI.
The splitter is coupled to a 6SN7 differential driver via a step network and plate-to-plate negative feedback. This arrangement gives excellent bass and balance, with sufficient stability to apply global NFB and without the headache of DC imbalance that can be caused by direct coupling with imperfectly matched tubes.
The OP stage is triode-connected EL34s but there's no reason why it couldn't use 6L6s. Sound is great. Hum and noise are inaudible. A schematic is attached FYI.
Attachments
Thanks for all the replies, good advice all round!
Right then, I'm going to draw up a preliminary schematic of my plan when I find the time. Here's what I have in mind.
Very similar to Ray Moth (above) plan, but a little less complex.
Splitter/input 6SN7 differential pair with 6DX8 constant current sink, direct coupled to 6SN7 driver stage (heaters elevated as required).
Solen Polyprop (cheap) coupled to 6L6 triode mode with adjustable bias (to play with!).
I might use a 6SL7 for the front end to give a bit more gain, but like the idea of using the same tube for all. Don't know yet.
Thanks again for the suggestions... Bob 😀
Right then, I'm going to draw up a preliminary schematic of my plan when I find the time. Here's what I have in mind.
Very similar to Ray Moth (above) plan, but a little less complex.
Splitter/input 6SN7 differential pair with 6DX8 constant current sink, direct coupled to 6SN7 driver stage (heaters elevated as required).
Solen Polyprop (cheap) coupled to 6L6 triode mode with adjustable bias (to play with!).
I might use a 6SL7 for the front end to give a bit more gain, but like the idea of using the same tube for all. Don't know yet.
Thanks again for the suggestions... Bob 😀
- Status
- Not open for further replies.