Installed 74F08 in 800XL, machine no longer boots

[+kheller2]

Hmm tin whiskers?  Are you using leaded or unleaded solder?

[flashjazzcat]

I remember when I was starting out, I got trolled by tiny wisps of cheap desoldering braid under the POKEY socket I'd just installed on a 130XE. Symptoms were the keyboard not working properly owing to invisible bridges which only showed up when I studied the underside of the board for the umpteenth time under magnification.

[+x=usr(1536)]

1 hour ago, kheller2 said:

Hmm tin whiskers?

Possibly, but these seemed almost more like contamination.  Think iron filings, but in splinter form.  They did not melt back into the solder when it was heated, and could be extracted individually.  Not ruling the idea out; just describing what I saw.

1 hour ago, kheller2 said:

Are you using leaded or unleaded solder?

It's Radio Shack part number 64-035, which gives its composition as 62% tin, 36% lead, and 2% silver.

[Dysphoria]

19 hours ago, x=usr(1536) said:

I did have to change the OS ROM at one point as I discovered that the one that was in there had lost a leg.  I have absolutely no idea how that would have happened as I did not remove it as part of this up to the point where it was found, but it was replaced.

When I read this, I instantly envisioned the lost part of the IC leg wedged under another IC, causing bridges in the curcuit.

 

12 hours ago, x=usr(1536) said:

What I found was what can only be described as a number of very fine metal slivers - not component legs, they were smaller in diameter than that - that had caused tiny bridges between a number of joints.  A trace had also lifted itself from the board at the pad, but it wasn't visible without removing the entire socket.

However, this discovery sounds like the likey culprit.

[Keatah]

6 hours ago, kheller2 said:

Hmm tin whiskers?

Could also be splatter from a manual desoldering pump.

[+mytek]

I just recently experienced a similar mystery on an old Colecovision game console I was restoring and modifying for S-Video output.

 

I had gotten it to work in a stock condition, and it was stable. Then I started to socket many of the chips (none are originally socketed). I fired it back up and the system would work for a split second and then loose video entirely. If I pressed on the motherboard or slightly flexed it, momentarily the video would come back. So first thing I did was do a visual inspection of every new socket that was installed, then I did continuity checks of all the pins and verified that all connections were still intact on both sides of the board, and there were no shorts. Next I got out the scope and verified that all the system clocks feeding the CPU and the VDP chips were good and didn't change when flexing the board. Also checked activity on the address and data lines, and all looked good. Finally I swapped out the VDP chip which is the heart of the video (kinda like the Antic and GTIA are in the Atari). No dice.Frankly I was miffed and wasn't sure what to do next, so I set it aside for a couple of days.

 

When I came back I once again started with a visual inspection, but this time concentrated my efforts around the VDP chip. I decided to try flexing the motherboard while doing so. Low and behold I saw something!!! A resistor associated with the reset line on the VDP chip was cracked down the middle and I could see it separate when I flexed the board. After I replaced the resistor the console now worked once again and was stable 👍

 

Apparently what had happened, was that when I had removed the soldered in VDP chip to get it ready to be socketed, as part of the final removal I had been lightly prying up on the chip with a screwdriver. Well that resistor was right next to the chip, and I had rested the screwdriver against it when applying pressure. The resistor body was ceramic and fragile, so it didn't take much force to crack it in half. Unfortunately at the time I never noticed that had happened, since it appeared whole and undamaged at the time. Anyway long story short, is that the system had worked reliably before I started changing things, so if it no longer worked after the change it almost certainly had to be associated with what I had done. The moral of the story is that going back to the scene of the crime can often times help to solve a mystery

 

@x=usr(1536) I'm glad the mystery has begun to unfold for you as well

[+x=usr(1536)]

21 hours ago, Dysphoria said:

When I read this, I instantly envisioned the lost part of the IC leg wedged under another IC, causing bridges in the curcuit.

One weird thing I'll mention in relation to this: when I pulled the OS ROM, the broken leg appeared to have cracked across its width about 2-3mm South of where it bends downwards to continue on to the socket.  There was nothing to suggest bending stress: it really did look like the sort of crack you might see in, say, concrete.

 

Related to that, the rest of the leg was never found.  It wasn't in the socket, nor was it anywhere on the board that I ever found.  That socket has also been removed, inspected, and replaced, so it didn't end up being left behind and/or bridging anything on it.

 

For a time, that chip had a field-expedient repair (read: total bodge) involving part of a cut-off component leg soldered to the stub of the broken leg left behind on the ROM.  It worked, but was swapped out once a new ROM arrived.

[+x=usr(1536)]

17 hours ago, Keatah said:

Could also be splatter from a manual desoldering pump.

Not saying it couldn't happen, but I'm using the squeezy bulb type.  Frankly, I don't care for the spring-loaded solder suckers because of exactly the scenario you're describing.

 

It really is time that I update a number of the tools that I've been using, however.  With the sheer amount of soldering and type of work that I'm doing, having basic stuff is great but not really suited to working effectively, efficiently, or at the scale I'm doing things at.

[+x=usr(1536)]

48 minutes ago, mytek said:

Apparently what had happened, was that when I had removed the soldered in VDP chip to get it ready to be socketed, as part of the final removal I had been lightly prying up on the chip with a screwdriver. Well that resistor was right next to the chip, and I had rested the screwdriver against it when applying pressure. The resistor body was ceramic and fragile, so it didn't take much force to crack it in half. Unfortunately at the time I never noticed that had happened, since it appeared whole and undamaged at the time.

This is exactly what happened to me after the first 64K internal upgrade I ever did on a 600XL.  The machine went from 16K and working to 64K and broken in about 30 minutes.  The pit-of-the-stomach oh-crap feeling when it happened was palpable   It was exactly the same 'broken resistor hiding in plain sight' scenario.  Found it by poking at things and noticed that it made a very soft but distinctly crunchy sound when I did so.

52 minutes ago, mytek said:

@x=usr(1536) I'm glad the mystery has begun to unfold for you as well

Thanks!

[_The Doctor__]

oh by the way the chip with the broken leg can have a new leg soldered to it normally, then you can reclaim that chip

[reifsnyderb]

3 hours ago, mytek said:

I just recently experienced a similar mystery on an old Colecovision game console I was restoring and modifying for S-Video output.

 

I had gotten it to work in a stock condition, and it was stable. Then I started to socket many of the chips (none are originally socketed). I fired it back up and the system would work for a split second and then loose video entirely. If I pressed on the motherboard or slightly flexed it, momentarily the video would come back. So first thing I did was do a visual inspection of every new socket that was installed, then I did continuity checks of all the pins and verified that all connections were still intact on both sides of the board, and there were no shorts. Next I got out the scope and verified that all the system clocks feeding the CPU and the VDP chips were good and didn't change when flexing the board. Also checked activity on the address and data lines, and all looked good. Finally I swapped out the VDP chip which is the heart of the video (kinda like the Antic and GTIA are in the Atari). No dice.Frankly I was miffed and wasn't sure what to do next, so I set it aside for a couple of days.

 

When I came back I once again started with a visual inspection, but this time concentrated my efforts around the VDP chip. I decided to try flexing the motherboard while doing so. Low and behold I saw something!!! A resistor associated with the reset line on the VDP chip was cracked down the middle and I could see it separate when I flexed the board. After I replaced the resistor the console now worked once again and was stable 👍

 

Apparently what had happened, was that when I had removed the soldered in VDP chip to get it ready to be socketed, as part of the final removal I had been lightly prying up on the chip with a screwdriver. Well that resistor was right next to the chip, and I had rested the screwdriver against it when applying pressure. The resistor body was ceramic and fragile, so it didn't take much force to crack it in half. Unfortunately at the time I never noticed that had happened, since it appeared whole and undamaged at the time. Anyway long story short, is that the system had worked reliably before I started changing things, so if it no longer worked after the change it almost certainly had to be associated with what I had done. The moral of the story is that going back to the scene of the crime can often times help to solve a mystery

 

@x=usr(1536) I'm glad the mystery has begun to unfold for you as well

That recently happened to me with a capacitor.  But it really shattered and broke.  Fortunately, it was just a decoupling capacitor for a DRAM chip.

[+x=usr(1536)]

21 hours ago, _The Doctor__ said:

oh by the way the chip with the broken leg can have a new leg soldered to it normally, then you can reclaim that chip

Yep, that's exactly what I did with it.  It's kicking around as a spare / test chip.

[magnusfalkirk]

So are you closer to figuring out the problem an managing to get it fixed?

[+x=usr(1536)]

18 hours ago, magnusfalkirk said:

So are you closer to figuring out the problem an managing to get it fixed?

Sort of.  The resistors around U18 need to be replaced (found a couple that have potential cracks in the carbon film), but that's about as far as I got this weekend with testing.  Good weather meant that working on the cars and taking care of some deferred maintenance had priority.

[magnusfalkirk]

Glad to hear you're making progress. Yeah taking care of vehicles and other maintenance has a big priority over hobbies.

[+x=usr(1536)]

At this point, I've decided to back-burner this until I can upgrade some of my tools.  It's been painfully obvious for some time that I'm not using ones that are completely appropriate for many of the jobs I'm trying to undertake, limit my ability to analyse problems, and generally put hardware at risk because of it.

 

Ultimately, I'm the person using them so am responsible for the outcome, but there's a lot to be said for not hamstringing oneself either.

[c1ph3rpunk]

Came across this thread as a result of searching for any info on the 74F08 chip replacement. Not to do the proverbial deceased equine thing, but given it’s like 18 months later, inquiring minds would love to know, did you get back to this?

 

I would also like to add how on today’s Internet these threads remind me of how many clouds I have not yelled at recently, I do miss the old Internet. People can in fact disagree, have a discussion, and do so in a decently polite manner, it IS possible.

 

Also, hope the arthritis gets under control, I spent a dozen years in severe pain with back problems. I ended up having 5 major spine surgeries, 2 of them multilevel fusions, but it’s been 9 years since the last one and, knock on wood, it’s held. Constant pain does weird things to your head, combine that with not sleeping, and the financial challenges as a result of SO many sick days and you end up generally pretty disagreeable. Good on you for recognizing this and saying what you did, wish I was that mature 20 years ago, I have a couple ex wives that will certainly attest to my disagreeableness then.

 

Related, I’ve found the Zoyi 2 channel o-scopes to be decent, and at $100 or so they’re generally obtainable. I’ve recommended them to a few students, they’ve had good results with them despite being on a student budget.