Compaq Portable 1 VDU repair help - schematics? replacement?

[mykrowyre]

After many hours of tracking down bad caps, I finally got my new Compaq Portable 1 to post, and give me error codes on the screen! Woohoo!

So I inserted my floppy controller card, and dang no post. Ok no problem. Removed it and SADLY I now have no video. My victory was very short lived.

I checked the horizontal and vertical sync and instead of 60hz and 18khz, I now have 5V on horizontal sync, and a perfect square pulse of 65khz (??) on vertical sync. What the heck!? I'm slowly working my way back thru the logic from H/V sync.

The mainboard still seems to be ok, I see normal activity on the bus.

I almost purchased the photofact, but I don't think it includes CSCS15-A, CSCS15-B, or CSCS15-C which is the video, floppy, and monitor sections.

Does anyone have a schematic for the VDU or maybe an extra VDU board they would like to sell?

Thanks

 

[mykrowyre]

I found a replacement board on ebay, but would still be interested in a schematic for the first revision of the VDU board.

Board: 000008-0000

 

[mykrowyre]

I checked the horizontal and vertical sync and instead of 60hz and 18khz, I now have 5V on horizontal sync, and a perfect square pulse of 65khz (??) on vertical sync. What the heck!? I'm slowly working my way back thru the logic from H/V sync.

Sorry I mixed up horizontal and vertical when I typed that.

I have 65kHz on horizontal sync (MDA), and solid 5V on vertical instead of 18.43kHz and 50Hz

Wondering if the shift registers are used to divide the frequency. There are three rather warm 74S299s.

 

[Hugo Holden]

If you could find the VDU schematic I could help to repair it.

Whether a VDU is at risk of being damaged by abnormal sync or H drive pulses, depends a lot on whether it has its own independent H scan oscillator, where it is safe from any abnormal sync, or if it uses the H pulses from the computer board as the scanning frequency source, then it is at risk. (in these cases VDU's such as the IBM5151 and the monitors in the PETs are at risk). I'm not sure about your one.

Its easy to tell, if the monitor has an H hold preset, if it has one, it is safe as it contains its own H oscillator that is merely synchronized to the incoming H pulses from the computer If these go grossly abnormal or even turn to noise, the H. AFC circuit simply drops out of lock and the H scanning frequency remains fine. This is why TV's would not be damaged by snow, or just noise for sync pulses, they all have their own H scan oscillators. Some Computer VDU makers decided they could do without the H oscillator, but its risky.

In the case where the VDU is at risk, the worst alteration in the H pulses you can get is when their frequency goes abnormally low. Going to a higher frequency actually causes less trouble. This is because the in the lower frequency case, the HOT is kept in conduction longer and the current can rise very high and the EHT too, so something can get damaged.

 

[mykrowyre]

Hi Hugo, thanks for the reply. I managed to get hold of the 1987 photofacts for the Portable, the VDU card schematic is there, although it's a slightly newer revision. Doesn't matter though because the primary chips are covered, so I think I have a good chance now of repairing it.

The VDU in this case refers only to the video card and the monitor is not connected. Will update when/if I find the dead component.

 

[djg]

The H and V sync are generated by the MC6845 U39. Have you checked if the signals are bad coming out of that chip? Looks like this is one of the annoying chips you can't read back the register values to check them.Looks like the cursor low R15 is read/write so I would check if you write various patterns testing all bits if the read back is correct. If readback correct I'd manually program the timing registers and see if H and V sync output matches what you program. If you have values and expected outputs I can check on my machine if needed.

Looks like the IBM 5150 table 6 and Compaq 80286 technical reference table 7-2 has the 6845 register settings.