Image Map Image Map
Page 3 of 3 FirstFirst 123
Results 21 to 29 of 29

Thread: Looking for insight into video design

  1. #21
    Join Date
    Jan 2011
    Location
    Vancouver, BC
    Posts
    4,572
    Blog Entries
    3

    Default

    I'm reasonably sure it is multilayer. When I hold it up to light I can see what look like tracks buried in the epoxy. Seems like the 5v rails are in that layer, I can see two traces attaching to the 5v pin at 90 degree angles from each other.

    I just can't find where 4c5 ends up. I'm concerned there is an error in the schematic and that 4c5 might be something else.

  2. #22
    Join Date
    Jan 2011
    Location
    Vancouver, BC
    Posts
    4,572
    Blog Entries
    3

    Default

    I'm reasonably sure it is multilayer. When I hold it up to light I can see what look like tracks buried in the epoxy. Seems like the 5v rails are in that layer, I can see two traces attaching to the 5v pin at 90 degree angles from each other.

    I just can't find where 4c5 ends up. I'm concerned there is an error in the schematic and that 4c5 might be something else.

  3. #23
    Join Date
    Jan 2011
    Location
    Vancouver, BC
    Posts
    4,572
    Blog Entries
    3

    Default

    This is the trace I'm having trouble finding:

    4c5.jpg

    I'm wondering if this is even something more simple.. like maybe that hidden 5V rail being broken and not getting power to one of the other chips on it.

  4. #24
    Join Date
    Jan 2007
    Location
    Pacific Northwest, USA
    Posts
    32,222
    Blog Entries
    18

    Default

    Color me dense, but where might I find the schematic for the whole board, not just little snippets?

  5. #25
    Join Date
    Jan 2011
    Location
    Vancouver, BC
    Posts
    4,572
    Blog Entries
    3

    Default

    Here's the whole one. There are a couple different revisions of the video board but at least from what I can tell mine is a Rev 3.

    https://drive.google.com/file/d/12Cg...ew?usp=sharing

    So far where a trace/track ends with a number like '4c5', I can find 4c5 on another page and follow it to its destination. Haven't found it yet though. I'm probably not understanding the significance of those 3 digit numbers properly.
    Last edited by falter; September 15th, 2019 at 11:24 AM.

  6. #26
    Join Date
    Jan 2007
    Location
    Pacific Northwest, USA
    Posts
    32,222
    Blog Entries
    18

    Default

    Tags like 2D5 or 4C5 are a coordinate system. So 4D3 means page 4, row D, column 3--and then you start looking for a matching tag.

    In your case page 4, coordinate C5 label VSYNC would be the line connected to pin 23 of U21 and pin 12 of U22.

    Coordinate systems were very much in use back when, even to the labeling of PCB components. This latter convention helps locate a part on a large PCB very quickly. So U5G would have you looking at the 5G area of the PCB. This went out of fashion as IC packages got smaller.

  7. #27
    Join Date
    Jan 2011
    Location
    Vancouver, BC
    Posts
    4,572
    Blog Entries
    3

    Default

    Well, I got most of the way there. Basically I removed the 74LS08 I had in place. I then carefully cleaned out the holes. Next, I used those old style pin strips:

    20190915_214245.jpg

    That was on the theory that they'd make better contact with any buried traces. I *really* worked that solder down into the holes just to make sure. And this is the result (and yeah, it's upside down, I don't know wth is wrong with the forum software): (!)

    20190915_213144.jpg

    So yeah, that socket got messed up somehow. I did find a couple of pin pieces from the previous 74ls08 that were stuck in the vias/socket holes, perhaps that was enough to not allow the chip to make proper contact. Not sure.

    So to break it down, originally we had no cursor. I got a non-blinking cursor by disconnecting and then resoldering the VID pin of the original CRT8021-003 chip. Anyway, I had traced signal to its pin 1 originally and found no pulsing. I *think* my soldering iron heat when I reconnected it may have partially woken that chip up. I did try reinstalling the original 8021 chip briefly today and it did not work at all, whereas the new chips did.

    I'm not sure if the 74LS08 was ever a problem. I saw that it was not ANDing two signals together to one particular pin (11), but I did not realize that the device's truth table could create situations where that might occur. So that chip may not have been at fault originally. However, while I was repairing the 74LS08 situation and trying to figure out what I'd done, I noticed a cut pin at U47. I had cut it during my very early forays in trying to diagnose the circuit, and had never put it back. I have a feeling had I seen that and reconnected it, I would have had things working again with just replacement of the CRT8021-003.

    So the only problem I have now is that the cursor, while present and blinking, is not moving along with the text. It's not affecting my ability to type, but it's annoying having it staying blinking up in the upper left. I'm going to reread the circuit (while I savor this victory) and see if I can figure out how that works. I'm assuming it's done by some form of counters.

    Anyway, yay, and thank you for your advice thus far, and if anyone *does* have any lightning bolt thoughts on the cursor, I'm all ears.

  8. #28
    Join Date
    Jan 2011
    Location
    Vancouver, BC
    Posts
    4,572
    Blog Entries
    3

    Default

    So I read the SMC 9007 datasheet, and it says that chip has registers that track cursor location. I am guessing that's updated by the 'cursor' circuits on page 4 of the schematic.

    Now, I know the cursor was moving before, minus text. So I am thinking something happened in my 'repairs' that disrupted that circuitry somehow. I'm just not sure where.

  9. #29
    Join Date
    Jan 2011
    Location
    Vancouver, BC
    Posts
    4,572
    Blog Entries
    3

    Default

    Fixed.

    https://youtu.be/32LqmGW4uhE

    Not sure which fixed it - I found the leg of the 74LS08 at U52 to not be fully in the socket, but it looked like it was making contact. However, remembering that I had had problems with swapping in non-LS versions of chips in my Apple II clone keyboard, I swapped out a 7474 at U1 that I had been using for testing purposes because it was the only one immediately at hand. I dug around and found an LS version and voila.. one of those two things cured it!

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •