Lisa Analog Card with no (proper) video

[pentium]

Jump in the Delorean for this one. We're going back to the 80's here.

This has been an ongoing problem since I bought my Lisa two and a half years ago. When you turn it on you get a raster on the screen however you can get an image (albeit out of sync because it's non-standard composite) from the rear video port.





In the troubleshooting process I first was able to rule out both bad cabling between the card cage and the edge connector for the analog board. With a scope I can confirm that I am getting a TTL video signal as well. Through swapping parts with a friends Lisa I can confirm that the problem is on my analog board. With his installed I have no problems with video.




First I installed CR6 which was present on a friends card but not mine. That did not work and after testing it was removed again.
Second, I replaced the TDA 1170N along with the 7824CT (which I accidentally killed some how). This did not fix the problem either.
I have yet to replace any of the capacitors on the board yet (but if these were the cause I'd be first documented case). I'll order replacements for all of them next time I am off work. I'm getting pretty frustrated with this thing.

The vintage apple community is almost useless for repairs. All the recommendations so far was to find one on ebay and buy it but that's upwards of $150. I'm being somewhat forced to repair this. They normally do not break.

[kaboom]

Re: Lisa Analog Card with no (proper) video

The vintage apple community is almost useless for repairs.

That's 'cause they don't work on TVs, which is basically what you have here.

All the recommendations so far was to find one on ebay and buy it but that's upwards of $150. I'm being somewhat forced to repair this. They normally do not break.

No good, either. What if you get one that has problems? There's nothing like fixing it yourself. If you also happen to save money, good for that too!


First, I will break down the video output. Then, I'll give you some things to check.

Working back from the CRT cathode, the unmarked 220 ohm resistor couples the cathode to the video output. Q2 and Q3 are the video output. Q6 is the preamp. Q1 is part of the load Q6 works into. It's biased by the TTL contrast control bit. Normally, when a pixel is to be displayed, just "video" is active. But when a "high brightness" pixel is called for, both "video" and "contrast" are active signals. This leads to a hotter signal, ultimately drawing more cathode current. At least, PC-based TTL video seemed to work that way.

The video output is powered by a winding on the fly, via R16, CR3, CR4, C7, and C8. R14 is the video output load resistor. L1 is a peaking coil; R13 is its damper resistor. CR2 is a DC restorer- keeps background brightness the same regardless of what's on the screen or how bright it is.

C6 and R11 form another peaker for enhancing image sharpness.

Now some things to check...

If open, no pic:
R7, R8, R9
R10, R12
R14
R16
R51
CR4
CR3
The combined pair of L1 and R13
Q1, Q2, Q3, Q6

If shorted, no pic:
Q1, Q2, Q3, Q6
CR3, CR4
C7, C8


You should have between 70 and 95 volts across C7 and C8, with C7 being one diode drop lower. R16 is probably a fusible resistor, as typically found after flyback secondaries.

Look for cracked solder on L1, R13 and R14, as well as the two output transistors. With a 10X probe, you can trace everything from the video inputs to the cathode.

You are closer than you realize to fixing this! For now, check that 22 ohm resistor (R16). If C7 or C8 short, the resistor opens. Not that it's any surprise, but lots of TVs were the same way.

Hang in there, you'll be quite happy when it's done!

More later,
-Paul

[pentium]

Re: Lisa Analog Card with no (proper) video

Checked what you wanted, plus did a little more.

I checked all components to see if open. Nothing found.
I checked L1 and R13 in series. No problem.
Checked listed components to see if shorted. Nothing found.

Transistors were tested by putting DMM (in diode check mode) positive lead on base and negative on either emitter or collector, plus checked from emitter to collector. Base to either Emitter or Collector read fine with no reading on polarity reversal. No emitter to collector shorts found.

Q1 and Q3 were completely replaced as I had spares handy. Q2 and Q6 were tested out of circuit.
All the 'lytic capacitors on the board excluding the 100V 10uf caps (which read fine on a capacitance tester and no resistance shorts were found) were replaced.

Components you suspected might have cold joints were reflowed. The card was reinstalled after all this and the problem persists. Assuming the tube must be connected, I can't check C7 or C8 or even easily poke around with a scope while the card is installed because it hides behind the CRT. With a bit of work I can leave it attached but it's a mess and takes up the whole kitchen table. I don't want to do it if you don't need me to.

[kaboom]

Re: Lisa Analog Card with no (proper) video

You actually can run it without the CRT. However, the horizontal yoke must remain connected, it tunes the collector ckt of the horiz output.

I would also keep the vert yoke connected. The vert amp is a current feedback type, it's sort of a distant cousin to a typical audio amp. I'd still keep the v-dy connected, though.

Obviously, make sure to insulate the anode lead. A small plastic container works well.

Also, check pin 2 on P3. If open or cracked, power for the video amp (from winding on fly) will not get to the board.

Check for +5 at the base of Q2.

I was wrong about that contrast control line. It's not used for altering the TTL video level- there's a cap in that ckt. It just varies the bias on Q1, changing the load Q6 works into. If R5 has a dirty wiper, Q1 will be cut off. In that case, the video preamp will not pass signal. If the trace for the contrast control pin is open, anywhere from the pot to pin "B," there will be no bias for Q1. The "hot" side of that pot should have a few volts DC on it.

You may end up doing what I had to do with a Panasonic AG-1960- solder leads onto suspect points and lead them out of the chassis, making scoping points accessible. I'd look close at R5 and the surrounding area first, though.

You should be able to AC-couple the scope and follow the signal at least through the preamp.

Finally, check R4, feeding zener diode CR1. The 5 volt supply, used for part of the horiz sync and Q2, originates here.


-Paul

[pentium]

Re: Lisa Analog Card with no (proper) video

Thank you for giving me all the reason in the world to take a friend behind the shed and beat the teeth out of him with a bike chain.

I had never touched the adjustment pots on the card because they were all dabbed with epoxy and I was told never to touch them. I read the contrast pot and it's infinite. I mark the spot with a pen and swivel it a few times with contact cleaner and I get a reading again. Reassembly was finally met with sccess.

[tom66]

Re: Lisa Analog Card with no (proper) video

Glad you fixed it. Seeing the retrace lines made me think of the CRT RPTV I had to fix once, it was on its last legs. It had an overblown picture with retrace, no raster visible. Turned out screen pot was bad in some way (don't remember how) but just touching it would bring the raster back.

[kaboom]

Re: Lisa Analog Card with no (proper) video

Thank you for giving me all the reason in the world to take a friend behind the shed and beat the teeth out of him with a bike chain. I had never touched the adjustment pots on the card because they were all dabbed with epoxy and I was told never to touch them.

Sometimes, bad advice is worse than none at all!

I read the contrast pot and it's infinite. I mark the spot with a pen and swivel it a few times with contact cleaner and I get a reading again. Reassembly was finally met with sccess.

I'm very glad to hear this!
Now give yourself a nice pat on the back.




Looks good!


I'm not sure how hard it is to adjust those pots with everything back together, but if you want to adjust brightness and contrast you can.

First, turn contrast to minimum. Then, adjust brightness up to a point where the raster is -just- visible. Then drop it back -slightly-. Finally, turn contrast up to desired level.

Ask me about the other adjustments if you are curious. You could make the raster bigger if you want, but I would just leave things for now...




again,
-Paul

[kaboom]

Re: Lisa Analog Card with no (proper) video

Glad you fixed it. Seeing the retrace lines made me think of the CRT RPTV I had to fix once, it was on its last legs.

This may have "soft blanked" the signal during the VBI, by the actual video signal being interlocked with the vert sync. In such a case, the horiz scan/retrace would be visible during vertical retrace. I've seen other monochrome computer monitors do the same. They'd always be OK once hooked to their computers. Unlike CRT TVs which sometimes derive horiz blanking from both the fly and the vertical ckt. (Both H and V blanking.) Those would never show the lines unless the CRT was biased to the point the cathode, grid, or other blanking means could no longer cut it off (during retrace.)

It had an overblown picture with retrace, no raster visible. Turned out screen pot was bad in some way (don't remember how) but just touching it would bring the raster back.

If there's scan, HV and beam current, there's a raster. You can have a raster and no video, but you cannot have a pic without the raster!

This

is a raster.



-Paul