9100A with no display

[Richard Wilder]

Hi Group!
I recently got a 9100A into my lab for repair. Symptom is no display. So far I've verified all supply voltages, fuses are all OK, the register display lights come on but the error light doesn't come on with the RUN switch set, pressed STOP, STOP, 1, CHS, SQRT. The error lamp tests OK. I downloaded Tony Duell's incredible schematics for the A model and have tested the display board signals and all the FF outputs with the following results. At this point I don't know which FF's should be toggling. I've also tested all the main ROM and control ROM outputs and I can post those results if anybody thinks that might be usefull. For now here are the FF states. Any assistance troubleshooting this vintage machine would be greatly appreciated! ;^)

9100A

Symptom: No display

Fuses: OK

Register illumination lights: Light

-15V, 220V & 3500V all OK

RUN mode, press STOP,STOP,1,CHS,SQRT: No error light, lamp tests OK

On the Vertical amp board:

ZDHD and ZDRD signals are OK, ZDMV and ZDRV are no signal

On the Horizontal amp board:

ZDHL is ON, all of ZDRL, ZDSL, HDRR, ZDFO are no signal

ZACE (blanking) on the gating board is stuck high (0V) (crt trace blanked)

Checked FF's on the TP connectors:

(0 = -V, 1 = 0V, T = toggling)

LH FF	CKT	STATE

F1	ROMA8'	T

F2	ROMA7'	T

F3	ROMA6'	T

F4	ROMA5'	T

F5	ROMA4'	0

F6	ROMA3'	0

F7	ROMA2'	0

F8	ROMA1'	T

F9	ROMA0'	0

F10	F50	0

F11	F51	T

F12	F52	1

F13	F53	1

F14	F54	0

F15	BADDRA'	1

F16	BADDRB'	1

F17	BADDRC'	T

F18	BADDRD'	T

F19	BADDRE'	T

F20	BADDRF'	T

RH FF	CKT	STATE

F1	F25	0

F2	F24	T

F3	F23	T

F4	F22	T

F5	F21	T

F6	F20	T

F7	F30	T

F8	F31	T

F9	F32	T

F10	F33	T

F11	F34	1

F12	F73	T

F13	F72	T

F14	F71	T

F15	F70	T

F16	F40	T

F17	F41	T

F18	F42	1

F19	F43	1

F20	F44	1

[Eric Smith]

Hopefully Tony will offer more authoritative advice, but from your description I'd guess that the logic is working correctly and that the fault is in the display electronics. IIRC, that circuitry is covered in the official service manual.

[Richard Wilder]

Thanks Eric! ;^)
I've rung out the display boards but problem is they're missing several of the signals as pointed out in the service manual. Also, the error sequence doesn't turn on the lamp so I'm concentrating on the logic now.

[Tony Duell]

I would have said this is a logic (rather than display) fault. The fact you can't get an error light from sqrt(-1) suggests the logic is not doing the right thing. And many of the inputs to the display driver boards are not toggling.

On the other hand, most of the FFs are doing something. So, alas, it's not an obvious fault like the clock generator.

My schematics cover the -B model (since that's all I own :-)). the flip-flop boards are identical, the control logic/microcode board (far left) is similar (same schematic, different ROM programming), the other areas of logic are different. How different I do not know.
A fault in almost any part of the logic (even the keyboard!) can kill the display, which makes life a little harder, particularly as I don't have the schematics.

I think what I would do next is to dismantle the machine totally and clean all the edge connectors. Test the diodes on the gating board (it doesn't take long with a DMM :-), they can all be tested in-circuit). Maybe check the transistors on the ROM PCB itself, they can fail, although normally a loss of a ROM drive line will kill the ROM outputs (and stop many of the FFs toggling).

The problem could be core memory related. Have you checked any signals in that area?

[Richard Wilder]

Thanks Tony! ;^)
I've dis/re-assembled the thing but I haven't tested all the logic diodes yet. Today I'll test the KB & also see what's coming out of the core board. On one of your previous posts you mentioned that all of the ROMAn outputs should be toggling. Mine aren't all toggling - should I be concentrating on those FF inputs? (Or maybe the idle state is not necessarily branching all over the ROM...?)

[Tony Duell]

Something struck me last night....

The display task is synchronised to the mains (power line) -- there's an AC feed from the secondary of the power transformer to a circuit in the logic assembly. I wonder if this is missing. On the -B, the circuitry is on the right hand 'sideboard', but I have no idea where it is on the -A.

May be worth looking at this, since much of the logic is working.

[Richard Wilder]

I noticed some wierd timing on some of the core ROM outputs. Here's the output from TBS:

And here's what I would consider a normal looking signal out of SC4:

That < 500ns glitch on TBS doesn't seem right.

[Richard Wilder]

I checked the KB today: All the strobes are changing with a key press as well as all of K0 - K5. I checked the ST output of the STOP transistor. It's normally -10V and goes to -15V when I press the STOP key. The STOP key generates strobe as well as a keycode on K0-K5. The -10V doesn't seem right - shouldn't it be close to 0V? I tried to follow the ST output from the KB connector on to the logic board but can't find it on any of the schematics. It becomes ISTP on the logic connector but I can't find it anywhere else. Any clues as to where this line goes?

Thanks! ;^)

[Tony Duell]

ISTP certainly goes to the rear peripheral connector on the gating board. I am not sure if it goes anywhere else...

[Eric Smith]

Apparently I misread it, as I thought he was claiming he did get an error indication. If not, then I agree that it is most likely a logic fault.

[Richard Wilder]

Any comments on that glitchy looking TBS signal or the ISTP high level being at -10V instead of 0V?

[Tony Duell]

If ISTP doesn't go anywhere other than the I/O connector (and it may well not), then there's no pull-up on that signal. I am not sure what you are measuring it with, but I'd not worry too much about the the odd voltage level. I will check the 9125A plotter schematics (the only peripheral I have) and see how it's used there.

Remember that the inputs to logic circuits on the 9100 are often current, not voltage, driven. This can make signals look odd.

For the TBS signal, remember, again, it's an open-collector signal. It's used (in the 9100B) on sheet 11 of the gating board schematic, were it's effectively ORed with TBA (via diode J1) and then used as an input to one of thr JA5 AND gates. Your glitch may well be the TBA signal.