Image Map Image Map
Page 6 of 28 FirstFirst ... 234567891016 ... LastLast
Results 51 to 60 of 276

Thread: Pet 2001 chiclet display problem

  1. #51


    Quote Originally Posted by PET 2001-8 BS guy View Post
    Thanks, the board is out of the computer, the monitor is only used at the moment to see if anything changes when I change a chip. Otherwise its powered down.

    By the way, this morning I changed C7 for a new chip, figuring that as its neighbour is from the same batch, it might help - no change.
    Please don't replace random chips.
    Earlier, when we were trying to get a clock from C8 pin 3. We ran an experiment to get a clock to the D flipflop. I believe you replaced C8, restoring that clock. That same clock should also clock the two halves of D5. It will toggle at a frequency too high for your scope. You should still see a 15KHz signal on pin 10 and 13 of D5. D5 divides the higher frequency signal by 2 at D5 pin 5 and again by 4 at D5 pin 3.
    D6 section QA should divide it down by 8. The clock pulses are running at 1 MHz and only about 60 ns wide so way beyond your scope. The signal at D6 pin 5 should be 125 KHz that we believe is in the range of your scope but it is being turned of on the horizontal retrace time. This may be causing missed triggers in your scope.
    The fact that you see nothing bothers me. I don't know if it is an artifact of the scope or there is nothing there at all.
    We should do another experiment but first I'd like to try something. Other people have had problem with the inputs to the address muxes shorting. Since we don't need them to get a raster on the screen, lets try removing them to see if they are causing any issues.
    Before removing them, mark them so we can get them back into the same locations. These are chips D3 and D4. They do not need to be in place while we are looking for our vertical signal.
    with them removed, check to see if you get a signals at D7 pin 5 , 9, 2 or 12.
    If not, we will try the experiment.

  2. #52


    I'm going to look at a second hand scope tomorrow afternoon, it's probably better to wait until then - if I buy it, I can retest the pins I've seen nothing on.

  3. #53


    Quote Originally Posted by PET 2001-8 BS guy View Post
    I'm going to look at a second hand scope tomorrow afternoon, it's probably better to wait until then - if I buy it, I can retest the pins I've seen nothing on.
    10Mhz means analog sine wave of 10MHz will be reduced by 3 db. It will have a difficult time looking at a digital pulse signal of 60 ns. 20 or 30 MHz scope is about the minimum. Also, if it is a digital scope, you will have an extra learning curve on how to use it and interpret the readings. There are plenty of used 100MHz analog scope out there. I don't think you want to waste money on a 10 MHz scope when it won't be much good for digital signals above 2 Mhz or so. Digital signals are not analog sine waves.
    In any case, we are getting close to the problem area. The counters D5, D6 and D7 have to run as their result creates the timing for the vertical signal ( note all the gates below these two on the schematic ).

  4. #54


    ok thanks for the advice, I'll keep looking then. When you say "Since we don't need them to get a raster on the screen, lets try removing them to see if they are causing any issues" do you mean remove D3 and D4? So far I have been removing chips by cutting the pins, pulling them through the top and socketing. It's all perfectly clean and tidy. I can probably remove ICs for re-use but I don't have a desoldering station, I'm just using an iron and sucker (and braid if needed).

    I do actually have 3* new 74LS157, I ordered quite a few ICs earlier this week as they're pretty cheap per unit.

  5. #55


    Opps! I thought your stuff was socketed.
    I'd like to do a little more before we unsolder stuff. Unsoldering can damage the board.
    We can run a little experiment.
    I'd like to disable the horizontal pulses for a short time. This will give us a clean shot at looking to see if we can get these counters to work.
    If we short C6 pin 3 to ground and hold at ground. This will cause C5 pin 5 to go high. This should also cause C6 pin 11 to go low causing C7 pin 2 to go high. The counter stages D5, D6 and D7 should now free run. If they are running, you should see a 62.5 KHz at pin 9 of D6.
    If not, D5, D6, D2 or D3 are at fault, most likely in that order.
    It is generally not good to short to ground a TTL but their outputs are current limited and for a few minutes to check things, C6 should be fine.

  6. #56


    Ok I have a few things to do today but I'll run that experiment.

    BTW, what do you think of this - - there doesn't appear to be much interest so I may get it for that price. It's only just over an hour from me so I can collect it.

  7. #57


    Ok so slightly interesting results. Here's C6 p3:

    C6 p3.jpg

    And here's C5 pin 5 before I grounded C6 p3:

    C5 p5 norm.jpg

    Now here's C5 pin 5 while C6 p3 is grounded:

    C5 p5 ground.jpg

    Here's C6 pin 11 before I grounded C6 p3:

    C6 p11 norm.jpg

    And here is C6 pin 11 while C6 p3 is grounded:

    C6 p11 ground.jpg

    Thing is, C7 pin 2 was high at 5.11V when I measured it before I grounded C3 p3. The frequency counter on my scope was flickering with all kinds of random numbers, indicating some kind of activity. But when I grounded C3 p3, it remained 5.1V high, and the frequency indicator showed no activity at all. I didn't get a picture of that because there isn't really anything to see other than a flat line. At no time have I seen anything on D6 p9. There's some indications of activity on that chip, but I can't measure them.

  8. #58


    Ok, this is what I was expecting if the counters are not working. I wanted to remove any interference from the horizontal. Normally the horizontal reloads the counters on retrace for the start of the lines of the same characters. After the entire characters on a line are drawn, it allows the counter to continue to the next lines of characters. By stopping the horizontal this way, I'm allowing the counter to free run without the periodic reload. It would seem that the counter is not running even with the horizontal stopped. So the problem is not related to the E6 not allowing the counter to run. Even when we remove all the things that could stop it, it is just not running!

    That looks like a nice scope, dual trace is really handy for this stuff. It even has delayed trace. That looks to be a fare price as well.
    Last edited by Dwight Elvey; December 15th, 2019 at 08:09 AM.

  9. #59


    On the 6502 forum we'd noted that D7 wasn't counting (while D6 had activity) and had replaced it, to no avail. Then I'd replaced E6, again to no avail. Once I replaced E6 I noticed that A1, which had been working when I first got the computer, was no longer doing anything. A1 has been replaced too, although I think I would have done that anyway as its legs were in poor condition.

    I have new replacements for all of them, so I will start with D5.

    Many of the scopes sold in the UK seem to be the old circular CRT variety, like my dad's old Serviscan unit (which I played on as a kid!). The choice in the US is much better.

  10. #60


    Most of those older scopes are 10MHz or less. The one your looking at looks to be a good scope.

    That is curious as E6 should have no effect on A1 running. A1 is the row counter for the characters. It should be stopped while the horizontal is stopped but otherwise it should run. It is clocked by the horizontal and reset by the vertical retrace as well as the beginning of the horizontal. This is partly why I don't like replacing things that that are not failing. Once we get this counter to run, we'll need to take a look at A1. The count value decoded at D8 pin 6 is what holds A1 during vertical retrace. It works through D8 pin 6 to C6 pin 11 to C7 pin 3, then to A1 pin 2.
    E8 pin 8 is to ensure A1 stays in sync with the counters D5 and D6. The path E8 pin 6 to E5 pin 10 then D8 pin3 to D8 pin 6 create the vertical retrace ( what we don't have ). This is the purpose of the counters D5, D6 and D7 as well as scanning the video RAM for the desired characters at that count. If you look at the count decoded by the D8 pin 3, you'll see that it is the last row of characters or 25. So the counters normally count to 25 * 16 = 400 then do a vertical retrace. This is why you have no vertical signal. No counters getting to 400 no vertical!
    This is why we are spending so much time trying to get something in the counters to run.

    Start with D5. I'm mixed as to replacing D2 or D6 next but tend to favor replacing D6 first. Keep with the D5, D6, D2 then D3 order I guess.


Posting Permissions

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