68kMLA Classic Interface

This is a version of the 68kMLA forums for viewing on your favorite old mac. Visitors on modern platforms may prefer the main site.

Click here to select a new forum.

SE/30 video issue that has ME stumped!

Posted by: NE555 on 2018-11-19 19:51:01 I hear ya. I got a SE/30 that does exactly the same thing.

Posted by: bibilit on 2018-11-20 02:41:13 Have you tried swapping PAL's as i did ? solved the issue for me.

Posted by: NE555 on 2018-11-20 08:37:25 I swapped the PAL (all of them) and cleaned the pins, I swapped the  PAL sockets, I swapped all the 8 bit shift registers, I swapped the ROM SIM socket and I swapped the memory SIM sockets. I replaced the capacitors and some of the resistors underneath the PAL chips.

Posted by: NE555 on 2018-11-26 18:52:27 any more or other suggestions?

Posted by: BadGoldEagle on 2018-11-27 02:00:28 Well, if you've swapped every suspect chip on that board, the only things that could be bad are traces... I know that's kind of a stupid suggestion, but I would go bad trace hunting with a schematic if I were you.  https://www.macintoshrepository.org/875-macintosh-se-30-schematics-and-repair

Posted by: NE555 on 2018-11-27 07:48:22 I noticed that there is a slight difference between the OP's screen output vs mine. My lines seem to go through the mouse icon and the question mark disk.

Posted by: techknight on 2018-11-27 15:01:21 yea thats commonly fixed. If you have already changed out UE8, then check the traces to it.  But more importantly, check the VRAM I have seen those go bad as well.

Posted by: NE555 on 2018-11-27 15:15:14 The LS166 is new. I am not 100% sure anymore if I changed out the VRAM. I'll check the UE8 traces again.

Posted by: NE555 on 2018-11-27 16:11:14 I think I found it Pin 5 on the ls166 (vid3) to UC7 pin 23 is OL. Pin 5 to via on the ls266 is OL. via to pin 23 on UC7 is .3ohm a closer look reveals that the pad is disconnected from the Via thank you for the help  techknight

Posted by: NE555 on 2018-11-27 16:33:24 and here is the result

Posted by: techknight on 2018-11-28 18:05:49 Yeap, I figured as much.  the board I have is way worse. lol.

Posted by: redruM69 on 2019-09-19 15:47:22 Sorry to dig up a dead thread. Techknight, did you ever find anything on this?  I have precisely the same video bug, but cant seem to figure it out.  Both VRAM's and UE8 have been replaced.  Tested all traces to and from VRAM.  I still get vertical stripes with a slightly abnormal disk icon.  Looks identical to your screenshot.

Posted by: redruM69 on 2019-09-19 17:02:14 Also, I have tried swapping all 5 PALs with known good.

Posted by: redruM69 on 2019-09-27 11:33:07 Bump?

Posted by: redruM69 on 2019-10-10 12:27:48 ISSUE RESOLVED!  Techknight, you may want to add this to your notes. It appears that Texas Instruments SN74HC166DRG4 is NOT a valid replacement for the the 166 shift register @ UE8.  It causes the discussed artifacts. I swapped it out again, but this time with a Toshiba 74HC166D, and the issue completely resolved!

Posted by: Bolle on 2019-10-10 20:59:16 I would always use the appropriate 74LS166 instead of a 74HC. It’s supposed to act the same but as you see it might not always do.

Posted by: techknight on 2019-10-16 05:57:29 You cant swap in an HC for an LS, because that is TTL vs CMOS, they have very different logic level High and Low thresholds.  And yes, I checked for that, although its been a very long time, I can check again as if I recall, I dont know if I was the one who changed the IC.

Posted by: techknight on 2019-10-16 18:21:18 So, I pulled out that board, I had to remember where it was as its been a couple years. Sure enough, it has an HC166 in there.  Didnt even think about it honestly. I checked my stashed that I was using and all of mine are LS166s like they are supposed to be.

Posted by: error1 on 2019-11-17 13:10:05 Thanks everyone for sharing your repair experiences, it was an awesome help to finally get my SE/30 recapped and fixed. I was stumped by the black vertical lines but after inspecting the corrosion damage around the old caps under a microscope and thinking that maybe there's something I'm not seeing, I finally desoldered UE8 and found the broken trace. A bodge wire later and my SE/30 is good as new. Now that I have a microscope it's making me really want a ultrasonic cleaner because the corrosion looks really disgusting up close and brushing with IPA apparently just smears it around  😀

Posted by: redruM69 on 2020-02-12 10:54:11 It's worth noting that although the Toshiba 74HC166D is a CMOS part, it appears to be TTL tolerant.  It worked fine as a replacement in my case.

< 4 >