Image Map Image Map
Page 6 of 6 FirstFirst ... 23456
Results 51 to 56 of 56

Thread: Quest COSMAC ELF

  1. #51

    Default

    Fairchild has always had a bad name in my book.
    When working for Intel, more years ago than I'd like to think about, we had a significant dropout of 7474s. This was traced to Fairchild parts. Our test showed that their output pull up was less than 5 mico-amperes. After testing them, we sent them back to Fairchild and they told us that the parts tested good. We put a stop on ordering Fairchild parts.
    Dwight

  2. #52
    Join Date
    Jan 2011
    Location
    Vancouver, BC
    Posts
    2,005
    Blog Entries
    1

    Default

    Didn't Apple 1s use a lot of Fairchild ICs?

  3. #53
    Join Date
    Jan 2007
    Location
    Pacific Northwest, USA
    Posts
    24,554
    Blog Entries
    20

    Default

    Fairchild TTL, IIRC. Those and other bipolar logic were fine. CMOS logic in the first half of the 1970s was fairly new--I still have my SSS (Solid State Systems) 4000-series databook. IIRC, CMOS was rated for optimum operation at Vdd = 15V. It was considerably slower at 5V.

    The RCA 1802, as well as the InMOS 6100 were among the first CMOS CPUs.

  4. #54
    Join Date
    Jun 2012
    Location
    UK - Worcester
    Posts
    1,309

    Default

    U8 pin 10 would appear the logical choice. Unfortunately, U8 is not present on Falter's the board...

    U8 pin 9 appears the next logic choice - and I would concur that this is the correct location.

    The TIL311 displays are transparent when the signal is low, and latched when the signal is high.

    The MC4508 latches are transparent when the STROBE signal is high and latched when the STROBE signal is low (i.e. exactly the opposite of the TIL311 displays).

    An inverter is therefore necessary for correct operation (which is present at pins 14 and 15 of the 'upper' 4049).

    So, I think the constructor has wired this part of the circuitry up correctly.

    The purple wire (on your drawing) to the 4508 OUTPUT DISABLE pins 3 and (presumably) 15 - would case the discrete display LEDs to be blanked out when the CPU was (I guess) running.

    If you are convinced you have a couple of bad 4016s - and you have some on order - then I would go ahead and replace those before carrying on. This will save confusing us by ensuring that we only make one change at once.

    Dave

  5. #55

    Default

    Quote Originally Posted by daver2 View Post
    U8 pin 10 would appear the logical choice. Unfortunately, U8 is not present on Falter's the board...

    ----snip----

    If you are convinced you have a couple of bad 4016s - and you have some on order - then I would go ahead and replace those before carrying on. This will save confusing us by ensuring that we only make one change at once.

    Dave
    Opps, didn't look at the board and notice U8 was not there.
    I agree, waiting for the new 4016's is a good idea. I still feel some protection
    is desired for the inputs.
    If me I'd solder some diodes on the switches now while waiting for the IC's.
    Dwight

  6. #56
    Join Date
    Jan 2011
    Location
    Vancouver, BC
    Posts
    2,005
    Blog Entries
    1

    Default

    Quote Originally Posted by Dwight Elvey View Post
    Opps, didn't look at the board and notice U8 was not there.
    I agree, waiting for the new 4016's is a good idea. I still feel some protection
    is desired for the inputs.
    If me I'd solder some diodes on the switches now while waiting for the IC's.
    Dwight
    I'm going to do that this week. I think it's a good idea given how apparently sensitive the 4016s are.

    The only reason I'm confident they are a main part of the problem is that the LEDs they affect are constant. If 4016 "a" and "b" are in their a and b positions, certain leds will always behave a certain way. If I put a into b and b into a's socket, certain different LEDs are affected. And then if I switch back again, we're back to the exact same LEDs being affected that were having problems the first time. Just seems too coincidental, too predictable. And I don't think it's a display problem.. the machine clearly is entering the data wrong and displaying it consistently wrong, as opposed to randomly wrong. So I don't think RAM is at fault either.

    I've got 20 4016s headed my way, not to mention another ELF, so we'll get to the bottom of this right quick. I really appreciate all the help!

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
  •