Improving signal strength on Superboard II

[Sideburn]

Hi all, the ntsc video signal from my superboard II is not strong enough for these lcd displays. I’ve tried two and the picture keeps dropping in and out. It is ok on an analog crt but not these LCDs. Any suggestion on how I get boost the signal?

[Sideburn]

A little more into I’ve tried both a video amplifier and an ntsc to hdmi converter and no luck. The signal just won’t lock onto a sync. It keeps dropping in and only. On both my 10” and my 5” lcd monitors I’ve tried. Hopefully someone can help I was planning on making the superboard two into a luggable laptop. But with no LCD it kind of kills the idea.

[ronin47]

I know what you mean regarding the signal strength, I have a Klyball remake, and the only thing I can drive reliably is a small 5" CRT TV.

I can get a picture on my LCD display that has a direct composite input and works with both PAL and NTSC (I'm in the UK, so PAL, but my replica uses 60Hz) but the picture is very grey and rather fuzzy. On my 5" CRT I get screen wrap which isn't ideal either.

I've tried an NTSC to PAL adapter and a composite to VGA adapter, but they simply don't pic up the signal. If you have any luck with finding a way to boost the signal I'd be interested.

[Sideburn]

Well I tried a booster already. One that’s adjustable and has 8 outputs. I turned all but one down and when it locks onto a signal it looks bright and crisp but then dropped out of sync and back so I’m not sure what the issue is.

The monitor I’m using has ntsc composite, hdmi, vga, and BNC inputs. It uses 12v input. Standard cheap displays you can find on Amazon.

Have you tried adjusting the POT next to pin 12 ofJ2 ? That adjustments brightness.

[bxdanny]

Why do you think the signal strength is the problem? Since the video amplifier doesn't help, I suspect that it isn't. LCD monitors are more picky about what signals they will display than CRT monitors are. If the frequencies or waveform are too far off from what is expected, the screens will blank. I've heard that some early models that didn't have such protection could be permanently damaged if fed an incorrect signal. 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. Possibly, the modifications done to your board make it differ even more from the standard. Do you have, or have access to, an oscilloscope so that you could take pictures of the video waveform(s)? Or a frequency counter, to see what the vertical and horizontal frequencies are? What is the frequency marked on the crystal that provides timing for the board?

[Sideburn]

Right I don’t think it is a signal strength problem either after I hooked the booster up. I should have clarified that in my second post. So I have been trying to figure out if there’s a way to correct the signal / sync.

I found this on marks page just now “ One-shot U65 generates a horizontal sync pulse at the end of each line (C7, at 15.375KHz) and a vertical sync pulse at the end of each frame (C15, at 60.0 Hz)”

Yes I have both analog and digital oscilloscopes. I wil try and take some photos today.

** I don’t know much about NTSC. Time to learn I suppose. From what I found, the horizontal sync is supposed to be 15.734 kHz and apparently the superboard is 15.375 kHz. Could this be the problem and how can I adjust it? Another crystal?

Could it being B&W with no color subcarrier be causing the issue as well?

I will checks all the frequencies next.

[bxdanny]

I think the standard Superboard can display its output on most LCD screens, but I'm not 100% sure. I made up this chart a while ago showing the horizontal and vertical frequencies of a C1P/Superboard II with several choices of crystal frequency, along with the deviations from both the color and black-and-white versions of the NTSC spec, and the serial port baud rate and its errors as well. It relates specifically to the unmodified 600 board, I don't know if yours needs a crystal that is double the frequency of what the unmodified board would use, or if it uses the same clock-doubling trick that the 600D uses, or something else altogether. I won't be of much help in interpreting the actual waveforms, though others here may be. But if your board uses the original 3,932,160 Hz crystal, you might try replacing it with a standard 4.00 MHz crystal, as I think the horizontal frequency may be more critical than the vertical. Likewise, if the crystal frequency is around 7.9 MHz, you could try an even 8.0.

Code: Select all

Crystal     Clock	VSync	  Vsync	    VSync      Hsync    Hsync     Hsync     Bit      Bit rate
frequency   frequency   freq.     error     error      freq.    error     error     rate     error
                                  rel. to   rel. to             rel. to   rel. to
                                  color     B&W                 color     B&W
                                  spec      spec                spec      spec

3,932,160     983,040   60	  +0.1%	    0	       15,360   -2.38%    -2.48%    295.38   -1.54%
3,993,600     998,400   60.9375	  +1.66%    +1.56%     15,600   -0.85%    -0.95%    300	     0
4,000,000   1,000,000   61.0352	  +1.83%    +1.73%     15,625   -0.69%    -0.79%    300.48   +0.16%

[ronin47]

Well I tried a booster already. One that’s adjustable and has 8 outputs. I turned all but one down and when it locks onto a signal it looks bright and crisp but then dropped out of sync and back so I’m not sure what the issue is.

The monitor I’m using has ntsc composite, hdmi, vga, and BNC inputs. It uses 12v input. Standard cheap displays you can find on Amazon.

Have you tried adjusting the POT next to pin 12 ofJ2 ? That adjustments brightness.

Yes, I've adjusted multiple times in both directions, but it just ends up looking like a weak TV signal. I am wondering if the OSI is the full issue since I have a TRS-80 CoCo 1 with a composite mod and the picture on the same display varies from crisp to very fuzzy.

I did try fitting a 100uF capacitor inline in a similar manner to how the simple composite mod is done on the Sinclair spectrum, but it didn't really make much difference.

Looking at the signal on my oscilloscope, it seems almost like the sync drops off occaisonally.

[ronin47]

I think the standard Superboard can display its output on most LCD screens, but I'm not 100% sure. I made up this chart a while ago showing the horizontal and vertical frequencies of a C1P/Superboard II with several choices of crystal frequency, along with the deviations from both the color and black-and-white versions of the NTSC spec, and the serial port baud rate and its errors as well. It relates specifically to the unmodified 600 board, I don't know if yours needs a crystal that is double the frequency of what the unmodified board would use, or if it uses the same clock-doubling trick that the 600D uses, or something else altogether. I won't be of much help in interpreting the actual waveforms, though others here may be. But if your board uses the original 3,932,160 Hz crystal, you might try replacing it with a standard 4.00 MHz crystal, as I think the horizontal frequency may be more critical than the vertical. Likewise, if the crystal frequency is around 7.9 MHz, you could try an even 8.0.

Code: Select all

Crystal     Clock	VSync	  Vsync	    VSync      Hsync    Hsync     Hsync     Bit      Bit rate
frequency   frequency   freq.     error     error      freq.    error     error     rate     error
                                  rel. to   rel. to             rel. to   rel. to
                                  color     B&W                 color     B&W
                                  spec      spec                spec      spec

3,932,160     983,040   60	  +0.1%	    0	       15,360   -2.38%    -2.48%    295.38   -1.54%
3,993,600     998,400   60.9375	  +1.66%    +1.56%     15,600   -0.85%    -0.95%    300	     0
4,000,000   1,000,000   61.0352	  +1.83%    +1.73%     15,625   -0.69%    -0.79%    300.48   +0.16%

I have a 4Mhz crystal in mine and the hsync and vsync values you have there match mine. I haven't tried any other crystals.

[Sideburn]

Is the character generator creating the ntsc timing timings and everything else other than the two sync frequencies?

I wonder if converting to rf and then using an lcd tv on channel 3 would make a difference.