IBM 5155 Power supply schematics or info

[Tony Duell]

> > > Maybe IC3001 or IC3002 defective ?
> >
> > Or the transistor is open-circuit?
> 
> The transistor (tested with multimeter) seem to be good (the two junctions
> are ohmnicalli corrected)
> I can try to substitute it. Now I haven't the board near me, I not remember
> if is a standard NPN

I think it is just a standard NPN transistor. From what I remember, it 
had a non-standard 'house number', but something like a 2N3904 will 
probably work.

> 
> > Still, you're now down to one simple-ish stage that presumably you can
> > debug.
> 
> Yes , the only question is : what is the tolerance of these protections ?

As yuo know, there are 8 comparators on this board, 4 for overvoltage and 
4 for undervoltage (no, it's not, IIRC, one chip for overvoltage, etc). 
You can trace back the sircuit from the base of the each of the 2 
transistors to find out which comparators are for each function.

I think what I might do next is get a spare comparactor chip of the same 
type (LM339)  and connect the power and ground pins to the power and 
ground pins of the chip on the board (I am not _sure_ they both have the 
same power and ground connections, check this!). Then link the inputs -- 
right way round -- to each of the 4 comparators in the undervoltage 
circuit in turn. Put a suitable pull-up resistor (to the comparator power 
connection on the output). See if any one of the 4 undervoltage 
comparators appears to have an output that's always low (remember, a low 
output from the comparator will cut off the transistor and therefore 
trigger the SCR).

You may find one of the output voltages is not coming up properly.

-tony