The scenario where a SS amp does more power into 8 ohms than 4

[nsubulysses]

What is the scenario that causes this?  I have seen it twice. One Sunn Concert Bass that did 112W into 8 ohms and only about 84 into 4 ohms (at clipping) and now I have a Sunn Concert Lead that will do again 30V into 8 ohms for 112W but only 16V into 4 ohms for a pathetic 64W at clipping.

Is this a scenario of driver transistors not being able to supply enough current to output transistors? I really don't get it unfortunately 

Sunn Concert Lead schematic is attached

[Loudthud]

The Sunn Concert series is a very primitive early 70's design. There is no current limit other than the beta of the output transistors and the drive available through the drive transformer. The driver transformer has a primary and four secondaries. On some versions there is a trim pot in the driver stage, but no info anywhere I can find on how to tweek it. Probably a clipping symmetry adjustment.

There are two basic versions. The earliest is a "Flying Bridge" with a floating power supply of about 35V. Check that the four transistors and all the resistors are good. There should be very little DC on the output with no speaker connected, perhaps a Volt or two. If there is too much DC, a transistor is blown or a resistor has drifted or smoked. It is not uncommon for a short to develop in the driver transformer between the primary and one of the secondaries. This puts big DC on the output.

The second version is a conventional totem pole configuration with +/- 34V rails, but it still has the same driver transformer. This version has a 5 wire power transformer, the earlier version has an 8 wire power transformer.

If everything seems OK, a low beta output transistor is suspected. If you don't have a way to measure beta at high current, you'll need to replace all of them.

I'm not sure any 240V export versions of the Concert series were ever made, but some 120V ones were made in Mexico.

Edit: Schematic you posted is the early version.

Edit2: Note that the wire common to R305 and R307 connects to ground at point 33 on the PCB.

[nsubulysses]

Just to clarify, both versions that I have seen this scenario with are the silverface versions that I believe were made from 1972-1975. The one with interstage transformer

No big DC on the output. There is about a volt as you say with no speaker attached. I have some 2N3055 I could swap some and see if there is a difference.

Will check into this amp more tomorrow thanks LT

[g1]

  It's not always a fault, it can be designed.
There is a yorkville powered mixer, 160W per Ch. at 8 ohm, 100W per channel at 4 ohm.  (see specifications on linked page)
  I can't recall why, maybe power transformer limitation?
If anyone cares to speculate, service manual w/schematic available on this page:
http://yorkville.com/legacy/mixers/product/mp10ds/

[nsubulysses]

The thing is I usually clock these amps in at about 130W or so at clipping into 4 ohms. I actually have a concert lead that does 132W at clipping into 4 ohms. There are these weird stragglers every once in a while, hmmmmm

[J M Fahey]

Maybe the ones which put out more power into 4 ohms are a version with driver trnsistors like all modern power amps.
The one you showed is, asnoted above, a transformer driven one.
Single direct driver for all 4 transistors is Q1, and since you have no outpu (discrete) Darlingtons output transistor Hfe on their own is very low.

If power transistors are current limited, then available power will be lower if you lower impedance.

We usually prefer to use voltages, because they are easier to read (you don´t have to break any connections to get your meter in) but expressed as current, output power is: I(squared)*R (or Z)
Those are 2N3055 so maximum current they can pass (in practice) is 10A (peak) ... and at that value Hfe drops like a sack of potatoes ... not that you had much to begin with.

2N3055 has current gain of only 20 ... at 4A , so at 10A it might have Hfe as low as 5 or 7 ... needing, say, 1A base drive to put out 7A, let alone 10A, which must be supplied by Q1 through the transformer .... and it has to drive 2 transistors at the same time, since it´s a bridge out, so we are current limited on all sides.

Suppose the transistor is limited to 7A peak.
*  @8 ohms it may supply (best case yadda yadda) 7*7*8=392W peak= 196W RMS
Not bad at all.

* @4 ohms: 7*7*4= 196W peak=98W RMS 

In fact 4 ohm loads are not recommended at all, since you have 70V available (in a brdged out load gets connected from +35V to -35V rails) , peak current woud be 70/4= 17.5 A, way beyond max datasheet value (15A) and even worse, real world practical rating of 10A peak.

While into 8 ohms it´s more reasonable.

[nsubulysses]

Of course you are all more knowledgable than me but I've worked on a couple dozen of these amps and I really do believe it should do about 130W into 4 ohms if it works right. Maybe I am hardheaded or LT can verify

These all (earlier silverface ones and later red knob ones) had insterstage transformer to couple the output transistors. I do however have the early silverface one that does not have the symmetrical clipping trimpot on the output board like LT mentioned.

My power supply rails at output transistors are +/-24V. LT could you please elaborate on what you mean by 35V flying bridge and the signifigance of R305 and R307 connection to ground? They do go to ground on the amp and check fine for resistance.

I will have to take my Concert Lead apart at home and compare but I am pretty certain it clips at 23V into 4 ohms. I am really missing something, unless interstage transformer could be faulty?

I have replaced Q1-Q5 with new RCA2n3055 and there was no difference.

[nsubulysses]

My Concert Lead at home does 87W into 8 ohms and 132W into 4 ohms. Power supply rails are +/-24 as well. This is what I usually see.

Thank you all for your input so far.

[J M Fahey]

I bet there are diferent versions.
+/-35V is a very different beast from +/- 24V , that´s for sure.

the "24" will definitely need bridged outputs for any signficant power into 8 or 4 ohms; the "35"  might drive 4 ohms directly, no bridging.

Also even if they look imilar on theutside, drive transformers can be wound different.
If 1:1 turns ratio, peak drive current is (best case) same as idle Class A driver current, this can be easily changed by varying turns ratio and pre-design max output current; in fact transformer driven amps had a built-in kind of short circuit protection if designer cared about that.

[nsubulysses]

Wow, I even did the Sunn Concert Series power amp stability mods at the end of this service manual and the amp performs exactly the same !!

I'll let you know if I ever figure it out

[Loudthud]

Have you checked the four 10W emitter resistors?

[nsubulysses]

.27 ohm emitter resistors all measure between .38 and .42 ohms

Hmmm

[Enzo]

So none is open.   Short your meter probes together on resistance.  What reading do you get?  That is the resistance of your probe wires, subtract that from your resistor measurements.


Or try this.  Grab a 10 ohm resistor, and measure it as close as you can.  The value doesn't matter, but you need to know what it is exactly.   9.873 ohms or whatever.  Use clip wires so the moving about of the probe tip doesn;t affect things.   Now measure the resistors in your unit that are 0.27 ohms but with that 10 ohm resistor in series.  Whatever they add to that 9.873 reading will be the resistance of the resistors under test.

Ever try to weigh your cat?  She won;t sit on the bathroom scale, so you weigh yourself, then weigh yourself holding the cat.  Do the math.    it's like that.