Help me get my head around this....

[tracynorton]

Hi
I'm usually a lurker, but have been getting deep into SS amps for awhile now. I have HH's and a Trace Elliot, and an old Heathkit TA16 I rebuilt. Love them all. My HH MOSFET Studio 60 'Power Baby' is my go to amp.

But I found something peculiar....I've attached the schematic and before I state what the amp model is...I need to know what this is a derivative of. It appears to modulate the rails...
I have checked, double checked, and compared both ch's (it's a 2 ch amp). It's a single sided pcb, very spartan and spread out (altho the layout and routing are worse than amateur).
It is a guitar amp. Has anyone seen anything like this? How does it work. It does work, I had it running and it's 100W into 4ohms!

Once I have complete schematic done (incl power supply), I'll be happy to contribute it to schematic section...and reveal make/model (it's well out of production).
Thanks
all.

[g1]

  That is what Enzo calls "flying rails" design.  I believe Roly has referred to it as "tail wagging the dog"  .
The key is what you have omitted, the power supply drawing.  In this case you will see that what is normally grounded (filter caps) is not ground.  It is the output hot.  They have just reversed the supply tie point with the output device tie point.
  As far as being derivative, the oldest example I have is from 1980.  But it may not even be the original concept.

[Enzo]

what he said^^^^

This is not new at all.  Flying rail, grounded emitter, other names too I am sure.  Not at all a new idea, and not at all derivative.  It is a standard form, just not what we might call conventional.

As g1 said, the common of the filter caps, the power transformer CT, does not go to chassis, it goes to the speaker.

In a conventional design, the speaker is between the emitters and the grounded CT/filters.  In the flying rails the speaker is between the grounded emitters and the CT/filters.  The only real difference is which end of the speaker is grounded.

Yes, the rails move around with respect to the ground, but the voltage across each filter cap itself is stable.

[J M Fahey]

The idea behind it is that a plain Op Amp, usually fed from +/- 15V rails and being able to swing just +/- 12V, can fully drive an output stage such as this one, fed from +/-40V rails (and the examples I saw were +/-75V rails or higher).

And it can do so, because output transistor base and emitters are always close to ground, piece of cake.

[g1]

  JM outlined the advantage of this design above.
The downside is that you need separate power supplies for multi-channel amps (as shown by the separate transformer winding in the QSC example).

[tracynorton]

wow. amazing. thanks! what a great forum.
I thought for sure I was crazy or was missing something. I have to put this thing back together and take some voltage measurements to add to the sch. I'll post the completed sch after that.

Incidentally, this is Lee Jackson's Perfect Connection SP-1000 design. I saw some guys over at Ultimate Metal forum had cloned it, but never published anything.

Yes, the power supply has two sep rectified secondaries..but no more filtering after bridge (just like QSC design posted).
Thanks again!