What?!

[Kilowatt]

What's with this schematic?! It looks to me like the outputs are shorted directly to ground, and I see no connection from the speaker to the OPS! I don't believe that this schematic could be accurate. It's for the QSC 1700.

http://schematic.narod.ru/Files/Audio/semi/QSC/1700.jpg

 

[Damon Hill]

It's capacitor coupled. Odd way to design an amplifier these days, given that direct coupling works so well with complementary output devices and bipolar supplies.

The relay switches the speaker to ground and isolates it
from the output stage. I think it's right but it's certainly not
the way I would build an amplifier.

 

[djk]

You guys don't get around much, do you? The grounded output stage has many nice feaures.Crown used it in their best amp, the PSA-2. In the QSC note that the cases of the outputs are grounded, no insulators required.If you want DC coupling just hook up the center tap of the power transformer to the speaker hot lead.Particularly nice for MOSFET amps.With the 2SJ50/2SK135 types the driver circuit needs to swing an extra +/- 7V in the front end.The Hafler XL-600 had a +/- 100V tier for the front end.The new TransNova Haflers ground the outputs, just like the QSC.Real easy to build a front end with a regulated supply that only has to swing +/- 7V instead of +/- 100V. http://k-amps.8m.com/cgi-bin/i/PowerAmps/Semicond/hafler_915c.jpg

 

[djk]

Get a TIFF viewer and then click on the images http://164.195.100.11/netacgi/nph-P...1&S1=5029299.WKU.&OS=PN/5029299&RS=PN/5029299 Those with a skilled eye will spot this as the same with the grounds swapped. http://164.195.100.11/netacgi/nph-P...,321,554.WKU.&OS=PN/4,321,554&RS=PN/4,321,554 Just to fry your brain, a Sumo Model 9 http://164.195.100.11/netacgi/nph-P...1&S1=4229706.WKU.&OS=PN/4229706&RS=PN/4229706 The op-amp runs on +/- 20V, the outputs on two floating 40V supplies.

 

[Damon Hill]

Nope, I don't get around much. I'm still skeptical, but I see
some practical benefits of the design and it's an opportunity
to learn. If it works for Crown, et al, it's worth a close
examination.

 

[Chucko]

AFAIK Pat Quilter (the "Q" in QSC Audio) was the first to come up with this output stage topology. It's used in most of their power amps. You can download a bunch of schematics as PDFs from the QSC support web pages if you're looking for more of this kind of thing.

Quilter's name is on a couple of interesting patents for switching power supplies too (the PowerLight patents). Check 'em out.

 

[mrfeedback]

Floating Bridge Amplifier

This is called floating bridge amplifier - allows easy stereo/bridge mono switching.
Notice that the power supply centers are not hardwired to ground.
This technique has been around a long time in Pro Audio.
Amcron VZ-5000 buts 5000w into 1 ohm in bridge mode.

Regards, Eric.

 

[Damon Hill]

It's giving me a little bit of a headache. I think it works by
modulating the power supply...

 

[Kilowatt]

I think I'll stick with normal OPS's.

 

[djk]

You guys are making this too hard! There is no difference between this and a "normal" output stage. In your "normal" amp the outputs are tied to one terminal of the speaker and the other terminal of the speaker is tied to the center tap on the power transformer. In the "wierd" amp the outputs are tied to one terminal of the speaker and the other terminal of the speaker is tied to the center tap on the power transformer. Uhh, that's exactly the same! What's the difference? The ONLY difference is which side of the speaker the feedback is taken from. One side is inverting, the other is non-inverting.

 

[Nelson Pass]

Just because a topology is pulled inside out doesn't
mean it doesn't work. I first saw this sort of thing
in Strickland's Accoustat TNT amp.

 

[HarryHaller]

Common Source

Well put from a man who designed a highly successful range of common source output stage amplifiers...... the Alephs. The coolest thing about this topology is that the front end doesn't have to swing very large voltages.

H.H.

 

[jam]

I believe Mr.Pass is correct in that Jim Strickland ,of Acoustat, invented this topology. Hafler currently use this output stage and a number of variants have appeared in MJ magazine.
I wonder if Mr.Pass or anyone would like to comment on the distortion characteristics of this topology.

Jam

 

[Nelson Pass]

The advantage of this topology lies mostly in its ability to
be driven by a low voltage source, such as an op-amp.

The output devices themselves have no idea what's going
on, and they see the same voltage/current relationships
as a similarly biased follower.

I have admired the ingenuity of Strickland's design, but I
always felt that it was underbiased. What would be very
interesting is this circuit operated Class A.

It is very easy to do.......

 

[Helix]

Even still, it doesn't bother showing the speaker! just a 'speaker bus' This is a very bad example of a schematic, and even though to someone who finds schema simple to follow, think of the poor newby repair guy! At a first glance, there isn't any power suppy caps drawn. not helpful thanks

Also, is that a OK way of fitting a clipping LED? what if it is clipping due to overloading (from current) as opposed to overdriving.

 

[djk]

Since the drive voltage requirements are very low, about one volt at the driver transistor inputs,any signal above a couple of volts is clipping.Because of the loop feedback the voltage at this point is forced high whenever the amp output cannot follow the input.This is true whether you run out of voltage on a high impedance load, or current on a low impedance load.

 

[Kilowatt]

Actually, there is a speaker hookup shown on the schematic. It's the square symbols on the end. The main signal path is drawn out, and it even tells the speaker impedance. Also, I don't remember seeing many schematics that do show the power supply caps. It's a fine schematic, I just found the topology rather strange since I had never seen it before.

 

[subwo1]

It looks to me like the output is taken through the filter caps. Pretty slick.

 

[FBJ]

Thanks for explaining that djk,
It took me a while to understand it also. Like most DIYers, we only really understand what we have seen before. And when something comes along we have not seen before, it's takes a while to get use to it. But with guys like you I am sure you will keep us straight.

 

[Helix]

the outputs isn't taken throught the filter caps, it is capacitivly coupled