I want to parallel 6-8 el84's SE UL.
bias is fixed bias.
driver stage a mosfet connected to a 1:10 step-UP.
i need to figure out what kind of input capacitance we are looking at on the sec of the 1:10 and therefore what the highest input impedance should be to that stage.
this will determine where to bias my mosfet.
what kind of power per el84?? 3W or so?
looking for ~25W.
finally, do i need to put feedback on this circuit?
Thanks!
bias is fixed bias.
driver stage a mosfet connected to a 1:10 step-UP.
i need to figure out what kind of input capacitance we are looking at on the sec of the 1:10 and therefore what the highest input impedance should be to that stage.
this will determine where to bias my mosfet.
what kind of power per el84?? 3W or so?
looking for ~25W.
finally, do i need to put feedback on this circuit?
Thanks!
I can't imagine why you would need to drive any quantity of EL84 with a 1:10 step up transformer, you don't need the gain as the EL84 is easy to drive, and such a step up ratio would increase the source impedance of the driver stage by 100X. I think a 1:1 would be ok.
Miller capacitance in UL will not be a big deal, I doubt the total including strays would be over a few hundred pF. This unfortunately will not be the case in triode connection where 10 tubes might be as bad as 2200pF assuming cgp around 9 - 10pF with a mu of 19.5
Even in UL connection the number of tubes required is going to be at least 8 - 10 in parallel to get this kind of power. You would need a custom opt with a primary impedance in the range of 350 ohms.
A pair of 6C33 would net you about the same amount of power in PSE triode connection. 600 ohm transformers for single 6C33 are readily available, two in parallel would require a custom made opt.
Finally a single pair of EL34 or 2 pairs of EL84 in pushpull UL would also give you the power you require with an off the shelf output transformer - would be much cheaper and probably perform better than anything based on a lots of EL84 in parallel.
Miller capacitance in UL will not be a big deal, I doubt the total including strays would be over a few hundred pF. This unfortunately will not be the case in triode connection where 10 tubes might be as bad as 2200pF assuming cgp around 9 - 10pF with a mu of 19.5
Even in UL connection the number of tubes required is going to be at least 8 - 10 in parallel to get this kind of power. You would need a custom opt with a primary impedance in the range of 350 ohms.
A pair of 6C33 would net you about the same amount of power in PSE triode connection. 600 ohm transformers for single 6C33 are readily available, two in parallel would require a custom made opt.
Finally a single pair of EL34 or 2 pairs of EL84 in pushpull UL would also give you the power you require with an off the shelf output transformer - would be much cheaper and probably perform better than anything based on a lots of EL84 in parallel.
yeah but
i really like the sound of the el84's.
thats the main reason. UL is very linear too.
i did not want to go with a tube voltage gain stage on front.
just use a buffer and a 1:10 to get the gain to drive the El84's.
I am also thinking about cathode feedback to reduce THD and decrease output impedance of output stage.
What tube can i use which has say 10-15 mu that i could use with a 1:1 IT to drive 8-10 // el84's and has the low distortion of a transformer?
i really like the sound of the el84's.
thats the main reason. UL is very linear too.
i did not want to go with a tube voltage gain stage on front.
just use a buffer and a 1:10 to get the gain to drive the El84's.
I am also thinking about cathode feedback to reduce THD and decrease output impedance of output stage.
What tube can i use which has say 10-15 mu that i could use with a 1:1 IT to drive 8-10 // el84's and has the low distortion of a transformer?
I like the sound of EL84 as well so I can certainly understand that.
When I made the reference to a 1:1 transformer I assumed you would use it with a mosfet or jfet in common source configuration.
The problem with a 1:10 transformer is the impedance reflected to the grids of the EL84 would be pretty high unless driven by a follower of some description, but even then with say a source impedance of 100 ohms that equates to a source z of about 10K which is not horrible of course. More to the point a really good 1:10 transformer is going to be relatively expensive.
The 5687 would do this job quite nicely actually and you could do it with a single tube for a stereo amp. The 5842 is another good choice,
and both of these have low enough rp that choke loading (>50H) on the plate would work pretty well..
When I made the reference to a 1:1 transformer I assumed you would use it with a mosfet or jfet in common source configuration.
The problem with a 1:10 transformer is the impedance reflected to the grids of the EL84 would be pretty high unless driven by a follower of some description, but even then with say a source impedance of 100 ohms that equates to a source z of about 10K which is not horrible of course. More to the point a really good 1:10 transformer is going to be relatively expensive.
The 5687 would do this job quite nicely actually and you could do it with a single tube for a stereo amp. The 5842 is another good choice,
and both of these have low enough rp that choke loading (>50H) on the plate would work pretty well..
- Status
- Not open for further replies.