Improving signal strength on Superboard II

[Sideburn]

No change with he 8mhz crystal.

I’m reading 15.62 kHz on pin 13 and 60.98hz on pin 12.

Was - 15.43khz and 60.24hz

[Sideburn]

Found something interesting.

If I do a hello world loop and run it then the video stays locked into sync.

It’s Like if the screen has enough content we are good.

But the power on random characters filling screen has trouble locking on…

[Sideburn]

Here’s a video clip demonstrating how it locks once I run a print loop: https://youtube.com/shorts/w8v1h_aI4nM? ... 2rB572CLaF

[bxdanny]

Oh, well. I thought it was worth a try. I don't suppose adjusting R58 one way or the other does any good? (I think it alters the relative levels of the syn and ideo data, rather than just boosting the whole signal.)

It looks like the signal is just on the hairy edge of being good enough for the LCD. Has anyone else here had problems getting LCD monitors to display video from OSI systems (540 or 600 board)? It might or might not be the case that the mod done on your board somehow puts it's signal further "out of spec" than the output of unmodified OSI boards. Possibly if you post oscilloscope images of the video waveform, someone might notice a problem with it. Or possibly not.

Are the RC circuits at U65 still as shown in the schematic (capacitors of 0.1 uF and 1000 pF, with 10k resistors)?

You mentioned that you had two different LCD monitors you had tried this with. Are they different brands? Have you tried it with the other monitor since changing the crystal?

[I thought I had posted essentially this message this morning, but when I looked later it wasn't there. So I'm trying again.]

[Sideburn]

adjusting R58 doesn't help it just makes the image brighter and i see the signal size in increase on the scope.

I will check the the caps and resistors at U65. They are two different brand LCD's one is a small 5". I will get it out and try it again.

in one of my video clips i have the signal in the background:

ntsc.jpg (437.83 KiB) Viewed 167 times

I can take some photos of a working NTSC signal and then compare against the Superboard's and see if we can tell a difference.

[jbtech]

OSI video signals don't adhere that closely to the NTSC standard in the first place, having a horizontal-to-vertical frequency ratio of just 256.

Hi, that's a good point and that alone could be enough to stump a lot of LCD monitors, not to mention that TV video signals are interlaced, alternate 60Hz fields beginning or ending on a 1/2 line while the Superboard II and most computer video outputs of the era are non interlaced.

Having said that some LCD monitors can cope with the non standard timing but I've also experienced issues with poor contrast and marginal sync due to the composite video output levels from the Superboard II, using both LCD and colour CRT monitors.

Taking some measurements around the video and sync output section I'm guessing that the composite video output was designed around black and white monitors or video converted televisions with relatively high impedance inputs and not terminated with 75 Ohms.

A lot of monitors at the time had loop through inputs and used separate terminating plugs or switchable 75 Ohm terminations while most modern video monitors, apart from professional or broadcast models, have permanent 75 Ohm terminations on their composite inputs.

On the Superboard II, video and sync signals are driven by TTL gates with open collector outputs and pullup resistors combined by potentiometer R58.

With no external load the proportion of video and sync can be adjusted well however placing a 75 Ohm load across the composite video output affects the level and balance of video and sync.

In fact when R58 is turned fully clockwise the video level at the output pin 6 of IC70 actually decreases under the additional load.

Rather than attempting to modify the monitors in question I decided to try adding a simple buffer to drive the external load without loading the composite video output from the Superboard II.

I've included a 75 Ohm output resistor to improve the matching between source and load as well as reducing the composite video to a more standard level, the unloaded output having over 2 Volts of signal.

buffer.jpg (24.77 KiB) Viewed 156 times

To me, that seemed to greatly improve the contrast of the image on both CRT and LCD monitors. It also appears to have solved a marginal sync issue with the LCD monitor pictured here.

osi600.jpg (112.92 KiB) Viewed 156 times

The buffer could be inserted between the output side of R58 and pin 12 of J2 but I've left that connection intact and added it to an otherwise unused section on the PCB, relocating the composite video cable to J2 pins 1 and 2.

buffer2.jpg (74.3 KiB) Viewed 156 times

On the Rev. D PCB U67, U68 and associated components are not populated and J2 pins 1 to 6 unused so the existing traces can be utilised as shown here.

buffer2b.jpg (81.53 KiB) Viewed 156 times

Although the original Superboard II composite video output would be perfectly fine for use with black and white CRT monitors of the era the signal isn't well suited to the standard 1V, 75 Ohm terminated inputs of colour CRT or LCD video monitors. Adding a buffer and source resistance seems to improve the image quality with some modern monitors however some LCD displays may still be unable to sync to the non-standard signal timing.

Hope that is of some help, regards John.

[ronin47]

OSI video signals don't adhere that closely to the NTSC standard in the first place, having a horizontal-to-vertical frequency ratio of just 256.

Hi, that's a good point and that alone could be enough to stump a lot of LCD monitors, not to mention that TV video signals are interlaced, alternate 60Hz fields beginning or ending on a 1/2 line while the Superboard II and most computer video outputs of the era are non interlaced.

Having said that some LCD monitors can cope with the non standard timing but I've also experienced issues with poor contrast and marginal sync due to the composite video output levels from the Superboard II, using both LCD and colour CRT monitors.

Taking some measurements around the video and sync output section I'm guessing that the composite video output was designed around black and white monitors or video converted televisions with relatively high impedance inputs and not terminated with 75 Ohms.

A lot of monitors at the time had loop through inputs and used separate terminating plugs or switchable 75 Ohm terminations while most modern video monitors, apart from professional or broadcast models, have permanent 75 Ohm terminations on their composite inputs.

On the Superboard II, video and sync signals are driven by TTL gates with open collector outputs and pullup resistors combined by potentiometer R58.

With no external load the proportion of video and sync can be adjusted well however placing a 75 Ohm load across the composite video output affects the level and balance of video and sync.

In fact when R58 is turned fully clockwise the video level at the output pin 6 of IC70 actually decreases under the additional load.

Rather than attempting to modify the monitors in question I decided to try adding a simple buffer to drive the external load without loading the composite video output from the Superboard II.

I've included a 75 Ohm output resistor to improve the matching between source and load as well as reducing the composite video to a more standard level, the unloaded output having over 2 Volts of signal.

buffer.jpg

To me, that seemed to greatly improve the contrast of the image on both CRT and LCD monitors. It also appears to have solved a marginal sync issue with the LCD monitor pictured here.

osi600.jpg

The buffer could be inserted between the output side of R58 and pin 12 of J2 but I've left that connection intact and added it to an otherwise unused section on the PCB, relocating the composite video cable to J2 pins 1 and 2.

buffer2.jpg

On the Rev. D PCB U67, U68 and associated components are not populated and J2 pins 1 to 6 unused so the existing traces can be utilised as shown here.

buffer2b.jpg

Although the original Superboard II composite video output would be perfectly fine for use with black and white CRT monitors of the era the signal isn't well suited to the standard 1V, 75 Ohm terminated inputs of colour CRT or LCD video monitors. Adding a buffer and source resistance seems to improve the image quality with some modern monitors however some LCD displays may still be unable to sync to the non-standard signal timing.

Hope that is of some help, regards John.

This is interesting, it's pretty much identical to the comp mod I did on my Coco1, and I wondered if it might work for my Klyball 600D.

I think I'll give it a whirl and see what happens.

[ronin47]

OK, so I tried a slightly different circuit to you, but only because I had it ready made in a drawer!! It's just a composite mod I made for an Atari 2600 (I think) that I abandoned and bought a ready made one for that included a nice cable and audio built in too.

IMG20240329145534.jpg (2.4 MiB) Viewed 149 times

IMG20240329150554.jpg (1.59 MiB) Viewed 149 times

And there is a definite improvement with the LCD.

This is before:

IMG20240329145249.jpg (5.11 MiB) Viewed 149 times

This is after:

IMG20240329145340.jpg (4.37 MiB) Viewed 149 times

However this is the CRT with it:

IMG20240329145453.jpg (5.01 MiB) Viewed 149 times

And this is the CRT without:

IMG20240329145649.jpg (4.58 MiB) Viewed 149 times

Personally I prefer the CRT without.

I'd be interested in what you think.

[Sideburn]

I’m going to try this. All I have laying around here that is close to that transistor is a BC338 and I’ve got a 100 ohm pot. Hopefully this is good enough.

[Sideburn]

I tried a quick bread board of that circuit but with a bc338 and could not get a signal on the monitor but could be my error and quick and dirty breadboarding.

Anyway, out of the three LCD displays I have, the large one is working but it is too large. It’s 19”.

I took some photos of a known working composite signal vs the superboard and I can’t see what the difference is.

Would that circuit you guys have built possibly fix my synch issue or is it just to address the contrast? My contrast is good.

IMG_8079.jpeg (2.27 MiB) Viewed 139 times

This is the superboard:

IMG_8075.jpeg (2.98 MiB) Viewed 139 times

This is a Nintendo clone ntsc:

IMG_8073.jpeg (3.06 MiB) Viewed 139 times

Here’s my video section:

IMG_8080.jpeg (2.22 MiB) Viewed 139 times