Image Map Image Map
Page 7 of 11 FirstFirst ... 34567891011 LastLast
Results 61 to 70 of 101

Thread: PET 3032 garbage screen problem

  1. #61

    Default

    Ok, I went back to the first post. On the first post, it was stated that all the RAM was tested good. I'm not that sure that the video RAM was included on that test. Looking at the web page referenced by Fran, I don't see anything to indicate that there was a test of the video RAM. I think I've been barking at the wrong tree.
    We have looked at just about every thing related to the video except the video RAM itself. I see that we get reversed video and graphics characters. I believe that is only related to the video RAM chip F8. I think it is about time we pull F8 and replace it. 6114s are hard to find but you should be able to get a 2114 a bit easier. They are still both hard to come by. I see Jameco list 2114s at $2.49 but the shipping would be a killer. One might do better from ebay or a Euroean parts house.
    I see a number of different brands listed from Bulgaria on ebay for under $8 USD shipped to the USA. I suspect these might be a good shot. Order several different brands. These parts are NOS and have most likely been sitting on shelves for 15+ years. They age just like the ones in your machine. These sellers don't test them, these are "you-test-ems".
    Dwight

  2. #62
    Join Date
    Feb 2009
    Location
    Southern California, USA
    Posts
    2,668

    Default

    Certainly worth a shot. Before unsoldering, perhaps a piggyback test on F8 with a spare 2114 to see if anything changes?

  3. #63

    Default

    Hi Dwight / Dave_m

    I apologize, I did not explain myself well, I only tried the system ram, the video ram was not tested since I have no spare parts. Last week I ordered 10 units https://es.utsource.net/itm/p/642397.html and 10 units https://es.utsource.net/itm/p/1711317.html.

    They are shipped from China, so they will take some time to arrive.
    Fran

  4. #64

    Default

    Quote Originally Posted by Fran View Post
    I have checked the ram and roms with the http://blog.tynemouthsoftware.co.uk/...nt-boards.html and they are ok.
    Whit the replacement boards you cannot test the video ram/rom.

    Years ago I bought spare parts in jammeco, but the shipping costs are very expensive, from China the shipping costs are free.

    One clarification, the video ram of my pet is 2114

    Fran

  5. #65

    Default

    If I'd realized that the video RAM wasn't tested, I'd have looked at the screen a little closer and how it failed. We'd still have scoped the SA4 some but not spent as much time looking at good signals. It was good to do though. Did you understand What we were looking for. Since the problem seemed to be related to SA4, we were looking for a fault of the SA4 getting to the RAM. The mux would switch back and forth between the CPU's address and the video counters address at the rate of the main clock. We finally saw solid highs and lows from both sources. Unless the values reaching the inputs to the mux were completely wrong, the mux was working as expected.
    On one of the earlier pictures you showed that the video's address counter looked to be working.
    Do you think you picked up any of the thinking that went into solving the diagnosis? Is there any question about why we looked at what we did?

    The schematic shows 6114s but the more generic 2114 parts were often used. On my KIM-1, the schematic shows 6102 for the RAM but my board was originally populated with 2102. I think this was related to the desire to be separate from the Intel side of things. Both the 2114 and the 2102 were originally Intel designs that other manufactures second sourced as Intel found the processors to be a more lucrative business. The Motorola side of the business never liked the Intel side of things. The MOS group were an offshoot of Motorola.
    Dwight

  6. #66
    Join Date
    Feb 2009
    Location
    Southern California, USA
    Posts
    2,668

    Default

    Dwight,
    The symptom of inverse video on every odd screen address is trying to tell us something, but what? The Latched Screen Bit 7 must he high on very odd address. I could go with a possible bad F7 RAM, but there must be something else wrong also with addressing circuit. And then the SA4 groups of video. Multiple faults?

  7. #67

    Default

    Quote Originally Posted by dave_m View Post
    Dwight,
    The symptom of inverse video on every odd screen address is trying to tell us something, but what? The Latched Screen Bit 7 must he high on very odd address. I could go with a possible bad F7 RAM, but there must be something else wrong also with addressing circuit. And then the SA4 groups of video. Multiple faults?
    I suspect the problem is inside of the RAM for the addressing as well and data. We can see that the counter is getting through the mux for both high and low for SA4. Although, we've not tested for the entire video RAM address space, we can see that the RAM addressing is unique for several lines. For part of the addressing of SA4, it looks to be working for the SA4 being high but fails for SA4 low.
    I agree that there has to be multiple failures but all still points to something inside the RAM chip. On ICs I've looked at in the past, a power to ground short would blow a hole in the silicon of a sizable diameter. Several circuits in the RAM could easily be damaged in the proximity of the failure. It doesn't look like the typical open bond wire type failure.
    Dwight

  8. #68
    Join Date
    Jun 2012
    Location
    UK - Worcester
    Posts
    3,010

    Default

    Bit of a dumb thought - but if you used my PETTEST and it passes the first (VDU) test, then the CPU can correctly read and write to/from the VDU memory (or what it thinks is the VDU memory at any rate). This 'may' narrow the problem down to an addressing/data or VDU logic area at least.

    Dave

  9. #69
    Join Date
    Dec 2005
    Location
    Toronto ON Canada
    Posts
    7,184

    Default

    Quote Originally Posted by daver2 View Post
    Bit of a dumb thought - but if you used my PETTEST and it passes the first (VDU) test, then the CPU can correctly read and write to/from the VDU memory (or what it thinks is the VDU memory at any rate). This 'may' narrow the problem down to an addressing/data or VDU logic area at least.

    Dave
    That's what I've been wondering throughout this most informative technical thread: does the OP have a diagnostic ROM or the means to make one, and if so did we see the output anywhere along the way?

  10. #70

    Default

    Quote Originally Posted by daver2 View Post
    Bit of a dumb thought - but if you used my PETTEST and it passes the first (VDU) test, then the CPU can correctly read and write to/from the VDU memory (or what it thinks is the VDU memory at any rate). This 'may' narrow the problem down to an addressing/data or VDU logic area at least.

    Dave
    Yes, I used your pettest

    20190917_224238.jpg


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
  •