Mac68k Forums

Home


Welcome, Guest
Guest Settings
Help

Mac68k Forums » Development » Hardware Hacking

Thread: IIci Failure (and recovery?) [success!]


Reply to this Thread Reply to this Thread Search Forum Search Forum Back to Thread List Back to Thread List

Permlink Replies: 14 - Pages: 1 - Last Post: Aug 20, 2012 6:31 PM Last Post By: landonf
landonf


Posts: 86
Registered: 7/23/12
IIci Failure (and recovery?) [success!]
Posted: Aug 1, 2012 1:39 AM
Click to report abuse...   Click to reply to this thread Reply
Attachment IMG_0482.jpg (109.6 K)
Attachment IMG_0491.jpg (186.3 K)
My new IIci failed today, just when I opened it up to install my new RAM.

The previous tenant had an apparently ill-trained electrician install new electrical wiring in our office; depending on how I lean on the floor boards, ground appears to be carrying current. When plugging in the monitor I zapped myself, and seemingly the IIci, as it would no longer boot.

I can boot the machine if I short pins 9 and 10 on the PSU (see IIci Startup Circuit Schematic); it appears that the machine is mostly still working, so I brought it home to work on it where I have proper tools. Only thing I had at work to test this was a twist tie:



First thing I noticed was the completely corroded battery holder (and battery). The thing is a disaster-zone:



Unlike the IIfx, the battery isn't required to boot, but I figured I might as well rectify this as long as I'm in there, especially since the corroded metal from the battery holder was getting all over the place. It's a quick job with the desoldering iron -- the battery holder is only soldered on via its two leads.

I also ordered an identical replacement -- Keystone 108 -- from Digikey, along with a replacement 3.6V 1/2AA battery: http://www.digikey.com/product-detail/en/108/108KK-ND/275347

I'll have to install those once they arrive. Next step for tomorrow is tracing out the startup circuit to see what is failing. I expect that the original caps are bad, but seems a bit funny that they'd decide to fail completely coincidentally at the same time I managed to zap things.
landonf


Posts: 86
Registered: 7/23/12
Re: IIci Failure (and recovery?)
Posted: Aug 1, 2012 11:03 PM   in response to: landonf in response to: landonf
Click to report abuse...   Click to reply to this thread Reply
I pulled the motherboard out and starting probing out the power circuit while trying to turn it on. I was seeing sporadic readings floating between ~1.25 and 5V across P9/P10, and then ... boom ... the machine started turning on/off reliably (<1v across P9/P10). At least until it stopped doing so a few minutes later.

Despite the coincidence I'm going to just assume the capacitors are spent. Even if that's not the source of the problem, can't hurt to replace them.

Some candidate SMD capacitors that fit (found via 68kmla):

7.30mm length (size D) TAJD476K016RNJ: http://www.digikey.com/product-detail/en/TAJD476K016RNJ/478-1739-1-ND/564771

3.50mm length (size B) TAJB106M020RNJ: http://www.digikey.com/product-detail/en/TAJB106M020RNJ/478-3876-1-ND/1126922

I verified the lengths with my calipers; the 7.30mm is a smidge larger than the existing capacitors, and the 3.55mm is a smidge smaller, but it looks like they should still be adequate.

Now to wait for another Digikey order. Someday I'll have a sufficient stock of parts ..
landonf


Posts: 86
Registered: 7/23/12
Re: IIci Failure (and recovery?)
Posted: Aug 3, 2012 9:16 PM   in response to: landonf in response to: landonf
Click to report abuse...   Click to reply to this thread Reply
Still waiting for replacement parts to arrive, but in the meantime I've been reading up on the power circuitry. There's an in-depth discussion in Apple Guide to the Macintosh Family Hardware 2nd Edition in Chapter 6 (Power Supplies - Page 239).

No schematic is provided, however; it's a basic block diagram of the power-control circuit (attached).
landonf


Posts: 86
Registered: 7/23/12
Re: IIci Failure (and recovery?)
Posted: Aug 8, 2012 10:03 PM   in response to: landonf in response to: landonf
Click to report abuse...   Click to reply to this thread Reply
Attachment IMG_0517.jpg (1.8 MB)
Attachment IMG_0516.jpg (2.7 MB)
Attachment IMG_0515.JPG (2.5 MB)
It lives! I swapped out all the SMD electrolytic capacitors for the tantalum replacements I listed above, and it booted right up.

I'm still not getting any audio on the built-in speakers, but I do get audio via headphones. Also, the machine will continually restart if I shut it down, as if I had the power button in 'server' mode. More to poke at later, especially since the battery + battery case haven't arrived yet, and I'll need to open it back up again when they do.

Update #1

Turns out the audio problem was just due to my poor form with the soldering iron. I wanted to try using the new hot air gun on this job, but my solder paste is backordered for three weeks, I had a hard time getting in there with the hand soldering iron, and two of the audio capacitors had particularly bad (loose!) joints. Whoops :(

I may have had better luck with smaller size capacitors (size C, 6mm), as the size I got (D, 7.3mm) almost entirely covers the pads.

I reworked those joints, and the machine is much happier now -- boot chimes! However, there's still an issue with the power circuit that I'm trying to track down.

In the meantime, the replacement battery and battery holder arrived from Digikey:



I had trouble soldering the ground side -- there's a large ground plane, and it just sucked up the heat. Once I brought out a big chisel tip, the solder flowed just fine. Here's the tip I used, along with the soldered battery holder:



I also made sure to save the original battery holder cover. The replacement holders I ordered did not include covers -- the old cover dropped right on to the new battery holder:



Now I just need to figure out exactly what's going wrong in the power circuit.
landonf


Posts: 86
Registered: 7/23/12
Re: IIci Failure (and recovery?)
Posted: Aug 9, 2012 12:20 AM   in response to: landonf in response to: landonf
Click to report abuse...   Click to reply to this thread Reply
Attachment HC132.png (16.7 K)
Attachment macIIcisch.gif (16.9 K)
I believe I've isolated the power issue to the top half of this circuit:



If I manually bridge the 5V trickle to the emitter of Q4, the mac will boot reliably. Likewise, it shuts down cleanly once booted.

By my (currently, possible demented) understanding, pins 2->3->4->6 on UE13 should be going low when the power button is pressed, driving the base on Q4 low and thus allowing the 5V trickle current to flow across base->collector and trigger PS power-on.

UE13 is a HC132: http://www.ti.com/lit/ds/symlink/sn74hc132.pdf
Q4 is a MMBT3906 PNP transistor: http://www.nxp.com/documents/data_sheet/MMBT3906.pdf

update

Even when both 1A and 1B are high, there's 0.7-1V across 1Y and ground:



On the other pins and HC132 ICs, if the two inputs are high, the output is less than 5mV.

When the power button is pressed, voltage on 1B drops, but 1Y still doesn't go high -- voltage on 1Y actually drops by about .100V.

If I apply 5V to pin 4 (2A) myself (connected to pin 3/1Y), the machine boots as it should. So, I believe the question is -- why isn't 1Y going high? Why is there 0.7-1V on 1Y (pin 3) when it's supposed to be low? I think this isolates the issue down to the single (replaceable) UE13 IC, but it's getting pretty late ...
landonf


Posts: 86
Registered: 7/23/12
Re: IIci Failure (and recovery?)
Posted: Aug 10, 2012 1:19 AM   in response to: landonf in response to: landonf
Click to report abuse...   Click to reply to this thread Reply
Attachment fan.JPG (2.6 MB)
Attachment fan_holder.JPG (2.1 MB)
Time for another Digikey order. I'm going to order a TI SN74HC132DR SOIC (14-pin) to replace what appears to be the failing UE13 NAND gates:

http://www.digikey.com/product-detail/en/SN74HC132DR/296-14832-1-ND/562693

While I'm at it, I thought I might look into fixing the IIci's fan noise. The fan appears to be a relatively standard square tubeaxial fan, roughly 80mm x 80mm x 25.4 mm. It's placed in a custom fan carrier that then slots into the PSU:




If I can replace the fan on this system, I might also give it a go on my other Macs -- they're all too loud.

While the specific Panasonic fan no longer appears to available, I found a similarly rated fan from Sunon Fans (P/N ME80251V2-000U-A99) that should fit just fine. It's rated at 30dB, so it should be considerably quieter than the failing fan:

http://www.digikey.com/product-detail/en/ME80251V2-000U-A99/259-1459-ND/2021087

The fan power connector on the Apple PSU (699-0392) looks like a standard 0.1" pitch Molex KK-style male locking connector. To mate the new fan with that, I found the Molex 2 conductor KK 2695-series connector (P/N 0022013027) and corresponding KK 2759-series crimp terminals (P/N 08-55-0102):


Of course, I don't actually have the crimp terminal. Found a cheap one on Digikey that should work -- GC Waldom W-HT-1921:

http://www.digikey.com/scripts/DkSearch/dksus.dll?vendor=0&keywords=W-HT-1921

Now to re-assemble the IIci and wait for the Digikey order ...
ajacocks

Posts: 4
Registered: 8/13/12
Re: IIci Failure (and recovery?)
Posted: Aug 13, 2012 11:18 PM   in response to: landonf in response to: landonf
Click to report abuse...   Click to reply to this thread Reply
I'm new here, but I have a similarly complex IIci power issue to work out, on my IIci. My machine was mute, until I recapped it, and the sound came back. However, the power issue remains.

My machine's issue is that, after running fine for approximately 30 minutes, it powers off. It will power back on for a very short time, before powering back offi, if the power key is pressed. If I wait a while, the machine will run for another 30 minutes, before powering off.

As mentioned, I have replaced every capacitor, both SMT and electrolytic, on the board, to no avail. I will admit that my SMT soldering isn't the best, but I was able to fix the sound circuitry. I'm assuming, at this time, that some IC on the board is overheating, after about 30 minutes of use, and returns to functionality after it cools down. However, I've tried to find any hot ICs, several times, and have failed.

I'm just about to the point of junking the board, and sourcing a replacement, but I hate to throw away all the work (and caps/recapping time!) that I have invested, especially given that any other IIci board that I get will also likely need to be recapped.

Any thoughts are welcome.
  • Alex
landonf


Posts: 86
Registered: 7/23/12
Re: IIci Failure (and recovery?)
Posted: Aug 14, 2012 1:15 AM   in response to: ajacocks in response to: ajacocks
Click to report abuse...   Click to reply to this thread Reply
ajacocks wrote:
I'm new here, but I have a similarly complex IIci power issue to work out, on my IIci. My machine was mute, until I recapped it, and the sound came back. However, the power issue remains.

My machine's issue is that, after running fine for approximately 30 minutes, it powers off. It will power back on for a very short time, before powering back offi, if the power key is pressed. If I wait a while, the machine will run for another 30 minutes, before powering off.

As mentioned, I have replaced every capacitor, both SMT and electrolytic, on the board, to no avail. I will admit that my SMT soldering isn't the best, but I was able to fix the sound circuitry. I'm assuming, at this time, that some IC on the board is overheating, after about 30 minutes of use, and returns to functionality after it cools down. However, I've tried to find any hot ICs, several times, and have failed.

I'm just about to the point of junking the board, and sourcing a replacement, but I hate to throw away all the work (and caps/recapping time!) that I have invested, especially given that any other IIci board that I get will also likely need to be recapped.

Any thoughts are welcome.
  • Alex

Obviously there's a lot that could potentially be going wrong, and I'm no expert, but in addition to motherboard failure, it's also possible that the PSU is overheating or otherwise failing. I don't believe that the IIci has a thermal power cut-off on the motherboard (someone correct me if I'm wrong), but according to the Apple Guide to the Macintosh Family Hardware, the PSU does (page 265):

  • If the heat sink of the primary switching devices in a Macintosh II-family power supply exceeds a temperature of 125 C, the power supply shuts down. Before shutting down, the power supply pulls the / PFW signal low for at least 2.2 ms.

Seems worth starting with trying to isolate the issue to either the PSU or the motherboard?
ajacocks

Posts: 4
Registered: 8/13/12
Re: IIci Failure (and recovery?)
Posted: Aug 14, 2012 3:20 PM   in response to: landonf in response to: landonf
Click to report abuse...   Click to reply to this thread Reply
Yeah, unfortunately, I had first thought it was the PSU, and so I sourced an up-rated Quadra 700 PSU, that was tested good by the seller, whom I know and trust. That didn't help, though.

If it were easier to cable the PSU to the IIci when it was not installed in the case, I would have done more exploration, but I suppose that I can attempt to remove the PSU's outer case, and use an IR thermometer on it, to see if I can find any particular hotspots. I'm not inclined to use my finger, there, though. :-)

I have a bunch of investment into the IIci, so I'd really like to get it working. I have 128mb of RAM, a Turbo '040, an FWB JackHammer card, running a 4.3gb UW SCSI disk, a Radius PrecisionColor Pro 24XK, and an Asante 10/100 NIC. The IIci is my favorite of all Macs, as may be obvious. When working, this IIci is seriously fast.

  • Alex
landonf


Posts: 86
Registered: 7/23/12
Re: IIci Failure (and recovery?)
Posted: Aug 14, 2012 10:39 PM   in response to: ajacocks in response to: ajacocks
Click to report abuse...   Click to reply to this thread Reply
Attachment psu_bench.JPG (1.9 MB)
ajacocks wrote:
Yeah, unfortunately, I had first thought it was the PSU, and so I sourced an up-rated Quadra 700 PSU, that was tested good by the seller, whom I know and trust. That didn't help, though.

If it were easier to cable the PSU to the IIci when it was not installed in the case, I would have done more exploration, but I suppose that I can attempt to remove the PSU's outer case, and use an IR thermometer on it, to see if I can find any particular hotspots. I'm not inclined to use my finger, there, though. :-)

I have a bunch of investment into the IIci, so I'd really like to get it working. I have 128mb of RAM, a Turbo '040, an FWB JackHammer card, running a 4.3gb UW SCSI disk, a Radius PrecisionColor Pro 24XK, and an Asante 10/100 NIC. The IIci is my favorite of all Macs, as may be obvious. When working, this IIci is seriously fast.

Heh, I'm a bit envious.

When did the power problem crop up?

The only other applicable reason I can find for the PSU itself to shut down without pin 9 being pulled low:

  • If any output of the Macintosh SE, Macintosh SE/30, or Macintosh II-familty computer is short circuited to ground (that is, connected to ground through approximately 100 milliohm or less), the power supply shuts down within 25 ms.

I suppose it's possible that a failing component could be shorting one of the PSU outputs? Perhaps you could attach a scope to the outputs and look for a short that occurs ~25ms before the PSU output drops to 0? Not sure what to use as a reference, ie, whether +5V.TRKL will stay active.

If that's not the failure cause, then I guess it's a question of determining what is triggering the NAND gates to drop the +5V applied to pin 9 via Q3 in the above schematic, since I believe that's the only way for the machine to trigger a PSU shutdown.

It'd be nice to get a 10 pin cable that fit the power supply. My solution for running this on the bench is pretty ugly (and involves opening the PSU):



I was looking at some molex connectors that seemed like they could work, but
ajacocks

Posts: 4
Registered: 8/13/12
Re: IIci Failure (and recovery?)
Posted: Aug 15, 2012 12:11 AM   in response to: landonf in response to: landonf
Click to report abuse...   Click to reply to this thread Reply
Thanks...I've spent some time collecting parts for this machine. None of the stuff I got was at all expensive, but it took some serious eBay/mailing list watching.

Unfortunately, the power problem has been with me, as far as I know, since I acquired the machine. I likely didn't notice it immediately, since I was constantly installing and removing hardware and software. I also don't own an oscilloscope, nor do I have the knowledge to use one. That's something that I've been meaning to learn, but I haven't managed, so far.

I really wish that I had a known-good IIci board to test with, but that's something I'm short.

It's such an odd thing for the machine to shut down after working for a significant amount of time, and then after a "cool-down" period, start working again. That really says heat issue, to me. Since you tell me that the PSU is the only component with a temperature sensor in it, I really do wonder what is up.

  • Alex
landonf


Posts: 86
Registered: 7/23/12
Re: IIci Failure (and recovery?)
Posted: Aug 15, 2012 1:47 AM   in response to: ajacocks in response to: ajacocks
Click to report abuse...   Click to reply to this thread Reply
ajacocks wrote:
Thanks...I've spent some time collecting parts for this machine. None of the stuff I got was at all expensive, but it took some serious eBay/mailing list watching.

Unfortunately, the power problem has been with me, as far as I know, since I acquired the machine. I likely didn't notice it immediately, since I was constantly installing and removing hardware and software. I also don't own an oscilloscope, nor do I have the knowledge to use one. That's something that I've been meaning to learn, but I haven't managed, so far.

I really wish that I had a known-good IIci board to test with, but that's something I'm short.

It's such an odd thing for the machine to shut down after working for a significant amount of time, and then after a "cool-down" period, start working again. That really says heat issue, to me. Since you tell me that the PSU is the only component with a temperature sensor in it, I really do wonder what is up.

The PSU is the only thing with a thermal switch, at least according to the the Apple Guide to the Macintosh Family Hardware. However, that doesn't mean that it's not possible for other components to fail in ways (due to heat) that would trigger a power-off.

As far as an oscilloscope goes, I'd suggest picking up a Rigol (eg, DS1052E) -- they're comparatively cheap, and I'm pretty happy with mine. It's pretty straight-forward to figure out, once you actually have one on-hand to play with.

I imagine "go buy an oscilloscope" is not the most immediately helpful advice, but that's the general direction I'd personally start poking. I'm also trying to order some IIci schematics from http://www.bomarc.org/, so I may have a better general notion of where to poke about once/if those arrive.
landonf


Posts: 86
Registered: 7/23/12
Re: IIci Failure (and recovery?)
Posted: Aug 18, 2012 2:37 AM   in response to: landonf in response to: landonf
Click to report abuse...   Click to reply to this thread Reply
Attachment IMG_0627.jpg (2.4 MB)
Attachment IMG_0628.jpg (1.6 MB)
SUCCESS!

The machine now boots 100% reliably after replacing the SN74HC132 at UE13. Here's the new one:



Here's the devil that failed:



Now I just hope that was the only zapped IC :)
ajacocks

Posts: 4
Registered: 8/13/12
Re: IIci Failure (and recovery?)
Posted: Aug 19, 2012 9:52 AM   in response to: landonf in response to: landonf
Click to report abuse...   Click to reply to this thread Reply
Excellent, congratulations!

  • Alex
landonf


Posts: 86
Registered: 7/23/12
Re: IIci Failure (and recovery?)
Posted: Aug 20, 2012 6:31 PM   in response to: ajacocks in response to: ajacocks
Click to report abuse...   Click to reply to this thread Reply
ajacocks wrote:
Excellent, congratulations!

  • Alex

Thanks! If you hit a total brick wall on your IIci motherboard, I'd be willing to at least try diagnosing the issue locally (I'm in Brooklyn, 11201).

Point your RSS reader here for a feed of the latest messages in all forums