Odd EGA Feature Connector Voltages...

[T-Squared]

Yes, this is a continuation of my little experiments with my Sanyo. This time, I'm trying to feed an EGA signal into the monitor's cables (since basic EGA is at the same frequency as CGA)

I haven't blown up my monitor yet, because putting the video cable back where it was still produces a CGA picture, so that's a good sign. The other good news is that the computer is still starting and executing the operating system properly as well. I actually had to go through a few trials, because it seemed that my XTIDE was in addressing conflict with any of the EGA cards, and my DIP switches weren't set properly to use their video BIOS.

However, I can't get an EGA picture.

I have two cards, one of them (A Paradise PEGA1A Autoswitch) has a feature connector, the other one (A CHIPS CT-8090) doesn't because its connector socket was weak (I could not get inserted connector pins to reliably stay in the socket), so I removed it, and I'm planning to get a new one that's newer and stronger.

Anyways, to the point: I've found that the connector on the Paradise card is not outputting the proper H&V sync signals. No voltage at all. However, the RGBI signals work (albeit a bit weak too, but I think it's more that each particular signal is not switched on at that time. I've been testing it in pure DOS, and of course, it's all-black with a gray text prompt, so maybe only the Intensity bit is being used then.)

I don't know what's wrong, or if this feature was removed in the Paradise card. I'll talk about the other CHIPS card when I have a new socket header for it.

 

[kdr]

Why don't you just connect to the normal 9-pin video output of the EGA card, since you have a CGA compatible monitor?

If you measure the hsync/vsync signals with a standard multimeter, you might not get an accurate reading -- especially on the hsync signal, which oscillates at 15.7khz

Also the RGBVH logic signals on the feature connector may not be strong enough to drive the monitor directly. On the EGA card, there will be a 74LS244 (or equivalent) connected to the 9-pin vieo output connector, and this chip provides sufficient output drive strength to get the signal through a video cable and into the monitor's input circuitry.

 

[T-Squared]

Why don't you just connect to the normal 9-pin video output of the EGA card, since you have a CGA compatible monitor?

Also the RGBVH logic signals on the feature connector may not be strong enough to drive the monitor directly. On the EGA card, there will be a 74LS244 (or equivalent) connected to the 9-pin vieo output connector, and this chip provides sufficient output drive strength to get the signal through a video cable and into the monitor's input circuitry.

Well, the Sanyo's monitor is internal, and the board is modular. One board contains the CPU, RAM, and Parallel Port, the other one contains the CGA circuitry and power conversion (from the PSU) circuitry. The CGA board has 3 outputs: Composite, DB-9, and a special 9-pin header inside (With GND, RGBI, and HV signals). When using a video card, the CGA circuitry is disabled, and thus, no image appears on the monitor. I have tried to connect the feature connector directly to the signal cable with patch wires (i.e. wires with Dupont pins on the ends), but I think the signals are too weak.

I wonder if I could just use a 244, like you said, to boost the signal. There was another circuit I found that uses that to produce a composite signal, and the chip might just be used as a booster.

 

[kdr]

Well, the Sanyo's monitor is internal, and the board is modular. One board contains the CPU, RAM, and Parallel Port, the other one contains the CGA circuitry and power conversion (from the PSU) circuitry. The CGA board has 3 outputs: Composite, DB-9, and a special 9-pin header inside (With GND, RGBI, and HV signals). When using a video card, the CGA circuitry is disabled, and thus, no image appears on the monitor. I have tried to connect the feature connector directly to the signal cable with patch wires (i.e. wires with Dupont pins on the ends), but I think the signals are too weak.

Ah, gotcha.

Looking more closely at the IBM EGA schematics, there are actually two sets of signals on the feature connector: the R/G/B/HIN/VIN inputs, and the R/G/B/H/V OUT 'outputs'. Have you tried connecting the monitor to the 'output' signals? They're meant to let the feature card take over the video output by having software set a bit in one of the EGA control registers. But that bit is usually not set, and in that case those lines are connected to the outputs of the '244 and (except for lacking a series resistor) are equivalent to the external 9-pin video connector's signals.

 

[vol.2]

Have you measured the signal levels on the wires between the feature connector and the internal monitor? What do they look like on a scope? (assuming you are using an isolation transformer)

Presumably, you just need to make your EGA levels like those, right?

 

[T-Squared]

Have you measured the signal levels on the wires between the feature connector and the internal monitor? What do they look like on a scope? (assuming you are using an isolation transformer)

Presumably, you just need to make your EGA levels like those, right?

That's what I'm thinking too. And yes, I have measured them. That's why I'm saying that they seem odd.