Image Map Image Map
Page 2 of 2 FirstFirst 12
Results 11 to 20 of 20

Thread: PET 3008. Father/Son repair project. Display looks like a maze.

  1. #11

    Default

    The video rams can be removed for testing.

  2. #12
    Join Date
    Feb 2009
    Location
    Southern California, USA
    Posts
    2,690

    Default

    With the two video RAM chips removed, you will get all ones going to the character generator. This should generate a checkerboard pattern on the screen if all is well except for video RAM.

  3. #13
    Join Date
    Feb 2013
    Location
    Chesterfield, UK
    Posts
    9

    Default

    So i checked chip G5-pin3 and was getting 8MHz. I was surprised at the voltage range though, it looks like approximately +- 1v, and I was expecting +-5v
    Here is the trace, in case I'm reading it wrong
    74191-pin3.jpg

    I also checked pin 7 and got 1MHz

  4. #14
    Join Date
    Feb 2013
    Location
    Chesterfield, UK
    Posts
    9

    Default

    The video ram chips (F7 & F8?) aren't in sockets. I worry about de-soldering them. if it's an essential next step I'll do it, but is there something else I can try first?

  5. #15
    Join Date
    Feb 2009
    Location
    Southern California, USA
    Posts
    2,690

    Default

    Quote Originally Posted by Eightbysix View Post
    The video ram chips (F7 & F8?) aren't in sockets.
    OK, leave them alone for the present. What ROMs are on sockets? What is the part number (with dash number) on F9 the kernal ROM?

  6. #16
    Join Date
    Jun 2012
    Location
    UK - Worcester
    Posts
    3,153

    Default

    In post #13 you say you had an oscilloscope trace that was indicating +/- 1V and was expecting +/- 5V.

    I am expecting a value from 0V to +5V...

    I would also like to see the timebase on your oscilloscope sped up a bit so that we can see a few cycles only on the screen if that is possible.

    How have you set-up the oscilloscope to read the 8 MHz signal on G5/3? e.g. how many Volts/division (I would suggest 1 V/div), AC or DC coupling (you want DC). And where is 0V referenced to the scale?

    Dave

  7. #17

    Default

    The signal looks like it is through a capacitor. Notice it is going above and below ground.
    Dwight

  8. #18

    Default

    It looks like the scope was on AC coupled. Without a pullup resistor, TTL isn't likely to go from 0.2V to about 4V. You may, also, be hitting the bandwidth of the input of the scope. I only see one sample notch on the rise time.
    +/- 5V would be unlikely from a TTL output. +/- 2V would be more likely through AC coupling. If you have the scope on DC coupling, it is likely that you have a bad ground connection.
    Dwight

  9. #19
    Join Date
    Feb 2009
    Location
    Southern California, USA
    Posts
    2,690

    Default

    Yes, he mentioned something called a Pico scope that has a stated bandwidth of only 10 MHz. I don't thing he can trust it for seeing the 8 MHz clock. I hope we do not have too problems in the video area as the PET dot clock for the data shift register is 8 MHz.

  10. #20

    Default

    Ah, I missed that. The spec the 10Mhz as an analog signal and that would be at -3db level. The frequency content of a 8Mhz square wave is a lot greater. I'd suspect a 8MHz square wave is about 0.5 the level or less. The voltage swing is about 1.3V in the picture. It is likely in the 0 to 3.5V range in real life.
    Working with these low cost DSO's is often at margins when using them as digital scopes. Still, in some ways it is better than some of the cheap logic analyzers. They may show a perfectly shaped square wave when the signal only has a 1 volt swing. The one thing I don't like about the cheap ones is that they have a capture buffer that is only as large as the screen. One of the most valued features of a more expensive DSO.
    Dwight

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
  •