Thinking about building a 3-channel Fisher...

Sometimes, if I have too much time to myself, I think up crazy things. This time around, it's building a 3-channel Fisher EL84 amp (either 30A or SA16 3-channel clone). I have some OPTs from R20 mono chassis (and already know the optimal feedback loop for a 30A form-factor thanks to Dave), as well as a power transformer from an X-202. Based on some calculations I've done, it looks as if the PT will have sufficient HV current to handle (3) 12AX7 / 12DW7 and (6) EL84 / 7189 tubes. The heater tap will be insufficient for my purposes, but an Edcor 6.3V / 5A PT is pretty cheap. I'm assuming that I can simply build three channels, wire up the hum balance accordingly (heaters for the EL84s will be on the Edcor, 12AX7s will be on the X-202 PT), build the PT section as for a 2-channel amp and let 'er rip.

Please tell me in excruciating detail why my assumptions are wrong (before I waste a bunch of money and time), and this isn't as simple as I think it will be. :yes:

-D

p.s. I've even laid out the FPX template for the top of the amp...it's a sickness I tell you!!!

Can't tell you if your assumptions are wrong - past my pay grade, but looks like a really interesting project.

Really like that PT Current Draw program through your link - v cool.

The plan is to look somewhat like an SA-16, just with an extra channel grafted on to the far side. Hopefully Larry or Dave or Fisherdude can talk some sense into me. :thmbsp:

-D

Knock some sense into ya??? NAH!!! DO THE CERTIFIABLE!!! Wish like hell i had the time, $$$, and ability to do this.

larryderouin said:

Knock some sense into ya??? NAH!!! DO THE CERTIFIABLE!!! Wish like hell i had the time, $$$, and ability to do this.

Click to expand...

$$$ shouldn't be too much. I've got a shit-load of spare parts to use. The only real costs will be in the top plate and the filament transformer.

So no possible technical hitches in my giddyap as far as you can see?

-D

using the 3rd channel as a bass channel, make sure of the freq. cutoff's. You might want to look at some of the Motorola schematics for ideas.

Derek -- By using a separate heater supply transformer for the output tubes, that leaves the X-202's transformer as being able to deliver significantly more power out of its HV winding than if its heater winding were fully loaded. Between this, not powering the preamp heater tubes, and eliminating the inefficient screen bleeder circuit of the X-202, there should be adequate volt-amps made available from the X-202's power transformer to allow production of full power output from all three channels at once if need be. Therefore, from a B+ supply standpoint, the X-202's power transformer should be fine. If you elect to use fixed bias, the current draw to supply that from the X-202's transformer is nil and can be disregarded from this exercise. Fixed bias operation is highly recommended to minimize the quiescent current requirements of the output tubes, and therefore minimize transformer operating temperature.

As for the screen dropping elelment, consider using the EFB screen dropping element I used in my Fisher clone and in that of the EFB modified SA-100. It can handle the extra screen current of the third amplifier with ease, and is highly efficient -- wasting little power versus that generated and lost to heat by the original design of the X202/SA-100 screen supplies. For that matter, you could build in the complete EFB circuit of the SA-100 modification, which would then insulate all three channels from any effects they might collectively have on the power supply. Of course, taking this approach, not only do you get more power output and less distortion, but extended tube life as well -- important in a design using six output tubes.

The 5AR4 can also handle the full current needs of the three channels as well, although only premium tubes need apply for the job as full power requirements will operate the tube near it's maximum current ratings. Again, this tube would welcome EFB or straight fixed bias operation of the three output stages as well.

Other than that, and having entirely too much time on your hands, I applaud your efforts to keep yourself out of trouble with such projects! Keep us posted!

Dave

larryderouin said:

using the 3rd channel as a bass channel, make sure of the freq. cutoff's. You might want to look at some of the Motorola schematics for ideas.

Click to expand...

Not bass. This is for L/C/R duty in my home theater. All will be full-range (or at least down to ~ 80Hz).

-D

dcgillespie said:

Derek -- By using a separate heater supply transformer for the output tubes, that leaves the X-202's transformer as being able to deliver significantly more power out of its HV winding than if its heater winding were fully loaded. Between this, not powering the preamp heater tubes, and eliminating the inefficient screen bleeder circuit of the X-202, there should be adequate volt-amps made available from the X-202's power transformer to allow production of full power output from all three channels at once if need be. Therefore, from a B+ supply standpoint, the X-202's power transformer should be fine. If you elect to use fixed bias, the current draw to supply that from the X-202's transformer is nil and can be disregarded from this exercise. Fixed bias operation is highly recommended to minimize the quiescent current requirements of the output tubes, and therefore minimize transformer operating temperature.

Click to expand...

Good to know - the only loading on the heater winding of the PT will be for the three 12AX7s, so the load should be minimal.

As for the screen dropping elelment, consider using the EFB screen dropping element I used in my Fisher clone and in that of the EFB modified SA-100. It can handle the extra screen current of the third amplifier with ease, and is highly efficient -- wasting little power versus that generated and lost to heat by the original design of the X202/SA-100 screen supplies. For that matter, you could build in the complete EFB circuit of the SA-100 modification, which would then insulate all three channels from any effects they might collectively have on the power supply. Of course, taking this approach, not only do you get more power output and less distortion, but extended tube life as well -- important in a design using six output tubes.

Click to expand...

I was already planning on using global EFB, although I'm a little puzzled as to how to calculate the appropriate resistor values with 6 power tubes (I'm assuming the cap, the pot and the LM337 remain unchanged). Needless to say, I don't think I'm up to trying EFB on a per-channel basis.

The 5AR4 can also handle the full current needs of the three channels as well, although only premium tubes need apply for the job as full power requirements will operate the tube near it's maximum current ratings. Again, this tube would welcome EFB or straight fixed bias operation of the three output stages as well.

Click to expand...

Not sure if the NOS Toshiba 5AR4 I have qualifies as premium. Is there a rectifier that goes to 11 so to speak (other than the GZ37, which costs waaaaay too much). Perhaps I should consider solid-state rectification.

Other than that, and having entirely too much time on your hands, I applaud your efforts to keep yourself out of trouble with such projects! Keep us posted!

Dave

Click to expand...

Thanks, Dave!

-D

bhamham said:

Can't tell you if your assumptions are wrong - past my pay grade, but looks like a really interesting project.

Really like that PT Current Draw program through your link - v cool.

Click to expand...

+1 I bookmarked that handy tool