Yet another one...

Greetings,

Last Friday another Spanish unit appeared. With both overCLK 's and mine's this is the third known unit to appear. This one is like mine and different from overCLK 's. As both units were found in Catalonia while overCLK 's was found in another autonomous comunity the conclusion is that the units sold in the Catalan market were different from the ones of the rest of Spain (which have that stylish Triumph-Adler badge).

As seen, the case is the standard one for the P2, but with the distributor badge (Guillamet) under the Alphatronic Sticker. My unit doesn't has the badge, apparently it was lost long ago but the glue is still there as a reminder of what once held.

On the technical side, the computer is also a little different. Up to that point all known P2 units were 64K units (even mine, which is not a "U" model). This one is the unupgraded 48K unit.

Now, the tests... the power supply is defective. It was shorting 5v, 12v and ground. All the boards were fine, there was no dead capacitor, even in the 48K DRAM card. With the boards in good shape I considered the computer to be ready for a test so I explained his owner how to wire an ATX into the backplane. After he checked the supply voltages from the backplane we considered it to be secure for the power up so he connected the three boards required for minimal operation: CPU, Keyboard and Video. The system resets correctly when powered on (also beeps as it should) and reacts to commands from keyboard. But the video display seems to be faulty.

The monitor is known to be working fine, so the fault is up to the video interface. I think his problem is CSync but I would like to know what do you think the problem would be. VTAC?

Note: this unit is not mine and I haven't even seen it other than i pictures and a couple of videos, the images posted here were uploaded with the consent of his owner. I'll try to convince him to come here. When he rescued it he said there was some possibility he had the Spanish manuals but after a search he found none. Still, the simple fact to appear was enough to give me one of the best days of the year.

Regards,

Jaume

Hello Helmut,

That's a relief. I told him the failure was most likely caused by the VTAC, but I wasn't completely sure. If that's the cause, the computer may be working very soon.

I haven't only guided him to find what was preventing the computer to start, but I supplied him my remaining ATX adapter and 48K replacement kits.

gpospi I was writing this when you replied. Thank you!

Regards,

Jaume

Greetings,

It's been some time. Last week the VTACs arrived and the one in the card was replaced by one of the Polish units. The image was stable, cursor moved but there was no character at all! The invert signal was also activated, so the screen was green with the exception of the cursor... This means, that the data path from the DIN to the VRAM and from the VRAM to the shift register should be checked. Being completely blank with the invert I imagine:

A) some 2114 may be faulty.

B) some 74 series IC in the bus may be faulty.

He has no scope to check the signals in the card... so he was stuck with it. For this reason, we agreed I would be in charge of the board for some time as I received one of such instruments not long ago. I don't expect it to be a long repair: as the VTAC retains most responsiblity for the generation of the signals, it shouldn't be as complex as a non-6545 PET.

Regards,

Jaume

Hello helwie44 †,

After the replacement, image seems stable, at least from what I've seen. It is a perfect blank, but inverted screen.

Picture courtesy of Shiri

The replacement is a legit VTAC, otherwise we wouldn't see the cursor (or anything at all). This seems a fault from the video RAM, latches or maybe some other common component coming before the shift register because (as seen on SKS documents and later with the EPROM backups) MSB is the invert signal. As you said, the TA board is a modified SKS but this remains unchanged.

Yes, Shiri is a bit far, but even if he was near I couldn't meet him due to restrictions on mobility... Still, he sent me the board through mail. When it arrives, I'll give it priority over the PETs.

Kind regards,

Jaume

Greetings,

The board is in transit and may arrive as soon as tomorrow. Meanwhile I have checked again the SKS schematics. From the description and pictures, and if there isn't another problem with power supply (such as failing -12V on Shiri 's side or broken zener) I have reduced the options to four possible causes, involving up to eight components:

A) VRAM failure (D5, D6, E5, E6). If the memory was wrong and somehow it always outputs 80h the result would be a screen as the one shown. Still, I thing such a case is so remote that I'm sarting to rule it out.

B) 74LS175 failure (B4, B5). If the memories were fine but the EPROM doesn't receive a correct value, culprits would be this pair.

C) 74LS166 shift register failure (B6). If its output maintains a high or undefined value (not low) next component that feeds its inputs with this signal may pull it up and generate an inverted blank screen.

D) TMS2716 failure (A6). This is the worst possible case to be expected and also the most probable of all scenarios. Was the EPROM the cause, its data output could go high or undefined, making the shift inputs to pull them high internally and generating the inverted blank screen.

As soon as the board arrives and I have some time, I'll report what I find.

Regards,

Jaume

Quote from jlopez

As soon as the board arrives and I have some time, I'll report what I find.

Two days ago the board arrived as expected and, when I had some time just before lunch I made some tests.

(Top board, my Italian video system; Bottom board, Shiri 's Spanish video card)

First I checked (again) if there was no shortcircuit in any of the power supply lines. There was no problem.

I plugged the board at slor nº4 of the Italian unit and turned the system on, being greeted with the RESET message. So, with the new VTAC the board works, and so does with the second VTAC of the lot he bought. Then I proceeded to test the original 5027 only to find it is not broken. Both facts come as a relief as I was expecting a worst-case scenario. Still, even being great news they are confusing.

So...

The video card works as expected. No VRAM or EPROM fault.

The monitor was tested with a game console and found to be working.

His CPU initialized correctly the VTAC on both cases he tested it, judging by the cursor images he shared. So system ROMs seem to be fine.

Other than the power supply hack (which may prove to be unstable), could the wiring between computer and monitor be the cause of such distortion? Any other board interfering someway?

Regards,

Jaume