iiinova — where your digital self keeps evolving

Haha, that video is a nice consolation prize at least! I hope you're able to fix it! If it helps at all, the 2SK1402As I got from activeparts on eBay were legit. Looks like they've more than doubled the price since I got mine though.

The bad news: I don't get to share video footage of my IIsi PSU frying. The good news: it's fixed! Even though I wasn't feeling optimistic after the first Q101 and D104 replacement failed, I tried again and it worked. The difference: this time, I didn't try to power it up with the dim bulb tester.

241

For what it's worth, replacing Q101 and D104 fixed my PSU. This time I only used the dim bulb tester to make sure it wouldn't blow the fuse initially, and for testing the 5VSB circuit. After that, I just went straight to the wall outlet. I think the dim bulb tester is what fried my first replacement

000

Ahh, gotcha! That makes sense. I'm considering trying to come up with some kind of daughterboard tester module that I can use to test all the circuits on it, because I learned the hard way that these things can fry the PSU!

Interesting, my megaohm-level readings are quite a bit different, but I'm guessing those don't matter much. My readings do differ quite a bit on the M51977FP though. Are you sure about the high resistance measurement on pin 3 of the IC? It's tied directly to GND on the bare PCB.

100

Ah yeah, the variable resistor being bad could cause a problem! Nice reference info! I'll be interested in hearing if replacing the transistor and diode fixes it for you. I'm slowly trying to gain the courage to try it again. I'm realizing I haven't really tested my replica in stuff like protection

000

Doug Brown @downtowndougbrown.com · 04/10/2025

Dang! I would definitely check D104 as well because it has shorted out both times I’ve seen the K1402A go boom. I am not sure if the K1402A going bad kills the daughterboard. The gate goes through T102, so maybe not. A faulty shorted D103 originally caused my PSU to go into protection, btw.

100

Oh no! Welcome to the club I guess, lol. I’ve been spending a lot of time studying the circuit. I wonder if yours blew the same transistor and diode that mine did…

100

Doug Brown @downtowndougbrown.com · 04/10/2025

I decided to take a break after the initial repair attempt seemed to re-blow the replacement MOSFET, but it’s possible the dim bulb tester was the cause of that. Each time it fails, it destroys the IC on one of my known-good riser boards, so I have to be careful. I’m going to look at it again soon.

100

Yep, the MOSFET and Zener are shorted again. Fun fun fun!

100

The main fuse that blows in this failure state is a slow blow fuse that doesn’t seem to protect the components either, though :-( I made my own little dummy load box I’ve been using (2 ohms on 5V, 24 ohms on 12V), but maybe I should be diagnosing this problem with a much lighter load.

I did use a 60W dim bulb tester, so a fuse didn’t blow, but I’m back to square one. I’m guessing the diode and/or MOSFET is shorted again. I’ve heard dim bulb testers are bad for SMPS though, so maybe I’m testing this the wrong way.

110

Doug Brown @downtowndougbrown.com · 07/09/2025

My IIsi power supply is proving to be tricky. I put in a new MOSFET and Zener, and this at least restored it to a state where it doesn’t blow the fuse when I plug it in. Good +5VSB too. Tried to turn it on, and it failed again and something’s shorted again. :-(

220

Answering my own question, I think the transformer might actually be okay? I was worried because of low inductance on the damaged side, but it’s literally just a wire going through the core once. Matching the PCB silkscreen. And it’s on a current sense part of the circuit. Whew…I think…

Doug Brown @downtowndougbrown.com · 07/09/2025

It also took out the 18V Zener diode on the gate, and I think it may have damaged this transformer too, which is a major bummer. It’s a custom part, no off the shelf replacement. Why does the PCB silkscreen show an inductor on one side and a straight line on the other?

110

Doug Brown @downtowndougbrown.com · 07/09/2025

I’m not sure if it’s _the_ problem, but this is definitely a problem! This is the main transistor that the riser board controls too, so maybe that’s what was damaged by the bad board.

140

Doug Brown @downtowndougbrown.com · 06/09/2025

Welp, my IIsi PSU is dead. Replacement fuses arrived today, and this time the fuse blew as soon as I plugged the power cord in. I guess the one faulty riser board really badly screwed something up. Damn.

320

Doug Brown @downtowndougbrown.com · 01/09/2025

This is not a time machine. Blowing the fuse in my IIsi PSU gave me an excuse to finish putting my IIci/IIsi PSU tester into a nice enclosure. Pretty proud of this! First time I’ve made a case like this that wasn’t 3D printed. Although I did need to design and 3D print some internal brackets.

040

Doug Brown @downtowndougbrown.com · 30/08/2025

Absolutely! I need to come up with a good set of sanity tests for the board before I try powering it. Hopefully the fuse is the only thing that was damaged.

Doug Brown @downtowndougbrown.com · 30/08/2025

Got out the toaster oven and built up some IIsi PSU daughterboards today. First two test out perfect. The third…I turn it on and kablooey! Visible spark and F101 blows on my power supply. Dang, this is what I was worried about. Hard to test these without risking my PSU. Hope it’s not shot.

220

Doug Brown @downtowndougbrown.com · 28/08/2025

Here's my blog post with the latest info about the recovery partition, plus a few other tidbits discovered since the first article came out: www.downtowndougbrown.com/2025/08/an-u...

downtowndougbrown.com

Downtown Doug Brown » An update about the hidden Performa 550 recovery partition

031

Doug Brown @downtowndougbrown.com · 27/08/2025

If you're getting tired of hearing about my IIsi riser board replica, never fear! It's coming to a close! My design is now live on GitHub. Built with all new components from Digi-Key except the two obsolete ICs. Still deciding how to build/sell them to recoup some costs. github.com/dougg3/iisi-...

github.com

GitHub - dougg3/iisi-riser-board: Replica of Macintosh IIsi power supply riser board model AP41S

Replica of Macintosh IIsi power supply riser board model AP41S - dougg3/iisi-riser-board

063

Doug Brown @downtowndougbrown.com · 26/08/2025

Ahh, got it. Well, if you ever think of someone who might need one, I’d be happy to donate the prize to them instead!

100

Doug Brown @downtowndougbrown.com · 26/08/2025

Haha, absolutely! I soldered it with hot air and didn’t do the greatest job. I bet my toaster oven with controller will do better! As the winner you are entitled to a free board when I have some built up and ready to go! I’m not sure how to get in contact with you for that, but can you email me?

100

It was definitely a tricky one. A phone camera doesn't fully do it justice, but pins 9 and 10 definitely had a unique appearance compared to the others.

000

Oh wow, you designed a replica daughterboard PCB as well? We could have joined forces! If it's useful at all, I have a full BOM all figured out with the help of Bomarc's schematics and a bunch of testing/research. I will upload my design, schematics, etc. to GitHub soon.

100

Just posted it here: bsky.app/profile/down...

000

The chips were floating around in a bag so the pin might have been bent. Pin 9 also looked iffy, but it was connected. I resoldered them both and that fixed it. BTW, today the M51944BML chips arrived and I confirmed them. So I can build these with all new parts from Digi-Key except the two chips.

Okay, my IIsi riser board replica is looking good. Everything will be uploaded to GitHub shortly, and I'll figure out a way to offer them for sale for convenience. As for the failure: it was a bad solder joint on pin 10 of the chip. @48kram.bsky.social had the closest guess and is the winner!

430

It’s a part Sony/Apple left out, it’s missing in your pic too. Yes, I figured out all the cap values with an LCR meter. All the components except the 2 chips are brand new. I will share the KiCad files, schematic, BOM, Gerbers, etc. on GitHub very soon as soon as I’m totally confident in the design.

200

Ewww! Pretty nasty. Thanks for the tip! You might not have seen it, but I did get my replica working. It was a dumb mistake I made. I have a little contest going to see if anyone can identify my mistake in the picture.

120

I borrowed it from my working board. Interestingly, identifying that component was by far the most difficult part of this project. I'm pretty sure that Bomarc was wrong when they called it a DTA124. I'm like 99% sure that it's really an M51944BML. I have some on order and will confirm very soon!

200

I want to leave guessing open a little bit longer, nobody has answered it yet. But don't worry, I won't leave you hanging forever. One person was close when they mentioned how it looked like the board hadn't fully reflowed.

110

The unpopulated component is intentional and the connected R and C in the top right weren't the issue, but you are definitely on the right track with this comment!

100

Those two upper pads are connected by the copper anyway, so it wouldn't hurt if they were shorted together. In terms of board location though, you're getting warmer!

000

That was intentional by Sony/Apple, unfortunately. I think it might be a capacitor they decided they didn't need

100

That component is not populated on Apple/Sony's original board too :-)

It is. It looks kind of funky in that picture though, I will admit!

000

Yes, good eyes, but it’s harmless. The middle pin is connected to the tab on the other side. Apple/Sony’s original design also has it missing, and I accurately reproduced their error. I suspect they originally designed it for another part or something like that.

First successful guess wins a free one (once I’ve fully tested the design and trust it!)

300

Ha! Actually, I found the problem! It's working! You can definitely see the issue in the picture. Anybody want to take a guess? My replica seems to be working now! The only component I reused from the original board is the upper left chip, because my order of those old chips hasn't arrived yet.

100

Doug Brown @downtowndougbrown.com · 23/08/2025

Built up my first IIsi power daughterboard replica. It looks nice, but it doesn’t work! The power supply immediately shuts off like it’s going into protection. Dang. I did end up with random tiny solder balls here and there from the paste. Could just be a soldering mistake.

5110

I'll write a detailed follow-up blog, but I'm now thinking the partition was added in the factory. One computer's partition was created on March 4th, 1994 and the other was created on March 5th. The contents appear to be identical, right down to the icon positions. No special Apple Backup version.