Help diagnose misbehaving aleph-x prototype

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I finally received my transformers from Victoria Magnetics. They are extremely wonderful. I recommend them to anyone building this amplifier. Now I can get on with the rest of my construction...

So, I have an Aleph-X front end and one side of the amplifier built for a few months, but I haven't been able to work with them since my last transformer developed a short. Now that I can apply power, I'm having a lot of problems, and I hope you will help me diagnose.

I have tested in several ways. If I hook up only the one side of output devices that I have built, with the front end disconnected, most things run fine. The rails are +/-12V, the current drawn is 4A. The drop across the current source power resistors is around 550mV, just like designed. The only problem is that the output tends toward the positive rail, but I expected that without the controlling circuitry.

When I hook up the front end with the inputs shorted, I get all sorts of bad problems. The output hovers around 5.5V and the amplifier draws so much current that I can only operate it for about 5 seconds before the power resistors unsolder themselves. Yes I am serious. The amp draws so much current that the rails sag to +/-7.5V and the transformer hums like the HVAC system in the Death Star.

I have made measurements and annotated them on this schematic in magenta type. This is basically the schematic I am using, except my front-end current source uses a buried zener reference with resistor to V- and I am using two parallel output devices instead of one.

As you can see it is drawing about 16A. Also you can see there is a lot more current through one leg of the input diff than through the other: 15mA on the right and 6mA on the other. Also note the unexpected measurements around the Dc offset feedback R36.

Is it even possible to operate half-an-aleph-x this way? I've been staring for a long time but it doesn't make me less stumped.

Cheers,
jwb

An externally hosted image should be here but it was not working when we last tested it.
 
I'm not sure I'm clear on how much of the amp is assembled. For me, your schematic ends just to the right side of the input differential. You've got just one side of the output stage installed?
If so, I think you've run afoul of the criss-cross nature of the Aleph-X front end. The feedback (which includes some arbitary amount of DC) is presented to Q4...that signal appears across R16. If the DC is positive, then the signal across R16 will drive more negative.
When Q4 goes negative against R16, the see-saw nature of a differential will tip the Q8/R17 side more positive (which you see as -1.7V across R17...a [relatively speaking] more positive value than the -5V across R16). This is Q3's drive. Expressed from the negative rail--and assuming that we're talking 7.5V rails--the Gate of Q3 is seeing +5.8V relative to the Source. In other words, it's biased <i>way</i> into conduction.
Okay, so what's up?
Well, the circuit is hooked up like an Aleph with the feedback going into the 'wrong' side of the front end...in other words, it's positive feedback instead of negative. Any little twitch will drive the circuit nuts.
You've got two choices. You can 'uncross' the drive, using the output from the other side of the differential. This will, in essence, give you a Mini-A. Or you can install the other side's outputs, and things should become more manageable. If you've got two sides running off the same differential, the balancing act is more realistic because the front end differential has a...well, a counterweight on the other side.
If I missed where you've got the whole channel put together, then we'll start over, but I'm assuming that you're running the circuit I see on my screen.

Grey
 
Yeah, I had somehow talked myself into the idea that I could test half the amp, as an assembly checkpoint. I thought the positive feedback I was seeing was a flaw in my assembly. But I just need to realize that the positive feedback wouldn't really be positive feedback If I added on the other end of the amp. Which, presently, I am doing.

Thanks for the reply.

Ever seen 3W resistors glowing bright red?
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.