GC SuperSymmetry

[nar]

The resistors to GND from each output give more stability against absolute DC offset too

Maybe like in the XA200 put some other "magical resistors" ? 4 of them on it if I can remember

[metalman]

I attempted throwing away some gain this weekend, using voltage dividers after the differential pair to drop 6dB before the LM3875's. There is still a stability problem, although the amplitude and frequency of the oscillation seems to have decreased.

Nelson, any idea how much of a reduction in open loop gain will be required to stabilize these circuits? I need to place an order for resistors to continue and was wonderng if you had any insight as to a ballpark figure...

Getting closer

[gengcard]

The resistors to GND from each output give more stability against absolute DC offset too

I would like to know I must add some resistors at the output of LM3875 or differential pair.

I just changed two resisters as Metalman's suggestion. it's better.

[metalman]

OK, tried my last set of suitable resistors before I have to order in more. Have thrown away 15.5db of gain, and things have improved again, but not there yet. With no signal at the inputs, the outputs (positive and negative speaker outputs) are reading about 1V above ground. I can play music, but it just doesn't quite sound right, very hard sounding. Lastly, when I switch the volume (using a stepped attenuator) I get a bloody big pop each time. I think my next attempt will be -21dB.

Terry

[Nelson Pass]

Originally posted by gengcard
[BI just changed two resisters as Metalman's suggestion. it's better. [/B]

You mean your simulation looks better.

[metalman]

originally posted by gencard
I just changed two resisters as Metalman's suggestion. it's better.

Actually that was a suggestion Nelson gave to me. I was just passing it along
I have been quite surprised by how much gain I have had to "throw out", much more than I expected. But then again, I was warned.

originally posted by Nelson Pass
One of the key issues is the enormous open loop gain of the
chips, which can give rise to some difficulty.

[gengcard]

You mean your simulation looks better.

Yes, it is.

Now, I'm making PCB. I have one question for heatsink of three Mosfets. Are they able to be applied with small size heatsink especially MOSFET of differential pair? It'll help me to make PCB easiler.

[gengcard]

If they can't be applied with small size heatsink, I will find a big heatsink for them include LM3875T two pcs.

For capacitor at power supply, how much should capacitance at power supply be???? Is 20,000uF enough?

[metalman]

At last a working circuit with no apparent problems. No hot chips, no clicks, pops, strange voltages, nothing that I can tell is wrong. Plays music quite well too, although I have only tested it with my portable CD player and el-cheapo bookshelf speakers. Is this circuit optimal? I have no idea, but it does work.

Terry

[Nelson Pass]

Looks very nice.