@wottle
Did you get all the bits of tiny dirty particles out prior to deyellowing? May need to use 409 or 100% IPA (something strong, but not brutal) to get the teensy tiny grime out, as that can cause lackluster deyellowing results. You may need to use the rough side of a dish washing sponge too. Something hard but not so rough it would scratch the plastics in any perceivable way.

Also, I have found submerging with the Retrobright mix to be more effective than using a gel form. But in my case, I use 3% H2O2 mixed with OxiClean. Other people simple use strong H2O2 hair cream (like 8-bit Guy) and no OxiClean. OxiClean enhances the H2O2. And in addition to direct sunlight, having your deyellowing session done on a warm or hot day helps a lot. Seems like heat is very important to get the best results, not just UV alone.

>> JDW said:

@wottle Did you get all the bits of tiny dirty particles out prior to deyellowing? May need to use 409 or 100% IPA (something strong, but not brutal) to get the teensy tiny grime out, as that can cause lackluster deyellowing results. You may need to use the rough side of a dish washing sponge too. Something hard but not so rough it would scratch the plastics in any perceivable way. Also, I have found submerging with the Retrobright mix to be more effective than using a gel form. But in my case, I use 3% H2O2 mixed with OxiClean. Other people simple use strong H2O2 hair cream (like 8-bit Guy) and no OxiClean. OxiClean enhances the H2O2. And in addition to direct sunlight, having your deyellowing session done on a warm or hot day helps a lot. Seems like heat is very important to get the best results, not just UV alone. Click to expand...

I started with a spray down with a hose, and a mild soap. it was still very dirty in some areas (mostly the front and the top near the front. Also, the case had some sharpie that came off with a bit of IPA and soft rubbing with a magic eraser like product. Weirdly enough, there were numbers on the side in yellow that didn't look like sharpie / ink, and as I cleaned, nothing would take it off. I ended up having to use a razor blade scraper, as it appears it was a paint that was actually raised on the surface. It came off, but the removal did leave some scratches where I did the cleanup. I tried using a rough side of a kitchen sponge on the really dirty bits, and it simply was not coming clean. So I decided the case looked bad enough that going aggressive with the magic eraser would likely not make it worse. So I put some elbow grease and really went at it hard. It definitely got Mose of the areas clean of the surface coating. I suspect it was actually a mixture of dirt and probably paint, since these were in a shop that did auto restoration. I'm going to get some baking soda and see if a solution of that and maybe some type of detergent will do a better job. I may also try soaking the case for a bit to see if that aids with the cleaning.

These are a whole different level of dirty than I've dealt with, and I've had some other "barn finds" that have required a good amount of cleaning.

As for the retro-brighting, I definitely use a slower, less aggressive approach. It's winter here, so daylight and heat are not on my side, and I've had bad results with outdoor retrobrighting with the peroxide cream. Short of me ordering an industrial size 50 gallon drum of liguid peroxide, I'm going to stick with cream, wrapped in plastic, in my UV light box. It's certainly possible that the lower temp in my workshop may have caused the redrobright to not be as effective. I originally planned to put a little heater attached to the UV light box to warm it up, but decided against it because I was a bit worried about the risk of a fire. I'll likely do another cleaning round on the darker areas of the case, then another round in the UV box. I'll also be grabbing some of the other top cases and testing other cleaning and retrobrighting methods as I work through the pile. I'll definitely have more cases than working machines when this is all done.

All my MAXELL pulls are in this cork stoppered glass bottle. Been waiting well over five years for the first one to 'splode. Hoping it'll set off a chain reaction in the containment. [fireworks]



BTW, LOVE what you did with that fx board, fabulous!

>> 3lectr1c said:

I definitely agree that the Maxells leak the worst of any of the brand Apple used (I have a page here going over them all: https://macdat.net/repair/kb/batteries_macintosh.html ). Maxell and Tadiran are the most common in vintage Macs, followed probably by Sonnenschein (the blue ones) and SAFT (the white/black or white/teal ones). Saft appears to be the next worst after Maxell. I just think it's doing a lot of speculation to suggest that Maxell deliberately cost-cut something when we don't know what went on in their R&D building back in the day - it could have very well been a design element that Maxell thought would extend the life of the battery that ended up making them leak in the long run. Or another problem in the supply chain that didn't have anything to do with them. Or it could be that they cheaped out on something - we don't know. In any case, just get 'em out Click to expand...

If you need more data for your spiciness index, here's a nice lineup of the batteries I've pulled out of these machines, by type. My experience in this batch matches yours. Every single Maxell was in terrible shape. The Safts also did not fare well (at least one survived). The Tadiran batteries did well except for one, and the blue Sonnenschein actually did the best - no leakage in any. (I think there was some rust transfer from the holder on a couple) There was one that was in such bad shape I couldn't even tell what it had been. It may be black, or maybe just turned black...




I wanted to get the Quadra 650 opened since it could be a fill in for my Quadra 800, so I opened it up, removed the battery, and all looked good. It powered on and I got video out, so hopeful I can clean it up and use the logic board in the Q800 case if needed. The awesome surprise when I opened it up was yet another PC Card. This one was a Reply board with an extension to work in the Q800/Q650 case, and it has a Cyrix 586 CPU.



I didn't get a chance to test it, but it looked to be in decent shape. I think it would be very cool to have a Quadra 800 with a 586 in it. So that will be my goal for that machine.

I'd love to use that photo on the page - thanks for posting! I think I'm going to adjust the Tadiran to a 2/5 and the Sonnenscheins to a 1.5/5. I can't ID that completely destroyed battery from the photo either. Could you post a closer up photo (or photos?) - I'm just curious.

>> wottle said:

If you need more data for your spiciness index, here's a nice lineup of the batteries I've pulled out of these machines, by type. My experience in this batch matches yours. Every single Maxell was in terrible shape. The Safts also did not fare well (at least one survived). The Tadiran batteries did well except for one, and the blue Sonnenschein actually did the best - no leakage in any. (I think there was some rust transfer from the holder on a couple) There was one that was in such bad shape I couldn't even tell what it had been. It may be black, or maybe just turned black... View attachment 18998 I wanted to get the Quadra 650 opened since it could be a fill in for my Quadra 800, so I opened it up, removed the battery, and all looked good. It powered on and I got video out, so hopeful I can clean it up and use the logic board in the Q800 case if needed. The awesome surprise when I opened it up was yet another PC Card. This one was a Reply board with an extension to work in the Q800/Q650 case, and it has a Cyrix 586 CPU. View attachment 18999 I didn't get a chance to test it, but it looked to be in decent shape. I think it would be very cool to have a Quadra 800 with a 586 in it. So that will be my goal for that machine. Click to expand...

Cool that it works... That should be a nice system once you're done. Re: the shielding, I think some people have done that. I'm not sure if that'll cause your monitor to start looking like it has an internal earthquake going on, or if it'll cause radio interference, but, worth a shot.

Battery that's too destroyed almost looks like another Maxell from what I can see written on it. So far, I've yet to see a Varta CR14250 in sad shape (in fact, my IIgs still has its original battery in good working order). Probably because they're a different lithium composition, hence the CR instead of the LS identifier used on newer lithium batteries.

>> This Does Not Compute said:

Your IIfx has the same weirdness as mine: All of the SMD caps are tantalum except for those two electrolytics over by the RAM slots. Click to expand...

That's the way mine was when I got it.

>> ScutBoy said:

That's the way mine was when I got it. Click to expand...

So, I decided to do some more testing with the board and tried to power it on. A component started glowing, then popped. Turns out it was one of those tantalums (the one at C1). It popped good, and ripped loose a pad when doing so.



Fortunately, it appears they designed the board to accept tantalums or electrolytic for many of the caps. So, I removed the fragments of the tantalum, cleaned it up, tried to glue down the original cap's pad, and then replaced it with an organic polymer cap on the other sets of pads. After replacing, the PSU seemed to power on fine, but I hadn't plugged in anything speaker wise. So I plugged in a speaker I had lying around and I got a chime! I will definitely be recapping the PSU, but it seems, at least for now, that the logic board is booting. Grabbed a video card and it booted up just fine. Now I need to recap the PSUs.

Really need to figure out how to bulk order my caps so I can try to keep the costs down. I have a bunch of 47uF 16v caps so many of the logic boards I can start tackling, but the PSUs are such a hassle to determine all the caps because there's such a variety of brands for many models.

>> 3lectr1c said:

I'd love to use that photo on the page - thanks for posting! I think I'm going to adjust the Tadiran to a 2/5 and the Sonnenscheins to a 1.5/5. I can't ID that completely destroyed battery from the photo either. Could you post a closer up photo (or photos?) - I'm just curious. Click to expand...

Feel free to use it. Or I can try to take a better photo once I've gotten through all the machines.

I'll also try to grab some better photos of the black one. There's actually a hole in the side because it was fused to the battery holder and the flathead punctured it as I was prying it loose.

>> ScutBoy said:

That's the way mine was when I got it. Click to expand...

And both of the ones I've had.

I do not own a IIsi, but it would be interesting to know more info about C1, such as:

1. What is its capacitance and VOLTAGE RATING.
2. Does C1 see +12V across it at any time? (Is it on the 5v or 12v rail?)



I've seen so many bad designs, even from Apple, using under-rated SOLID tantalum. And of course, because many people in our vintage Mac community merely "follow the leader," it is all too often the case that replacements are SOLID tantalum 16V. A rating of 16V is only fine for SOLID tantalum when used on a 5v rail. It is not OK to use it on a 12V rail, not only because 12V rails tend to be more than 12V nominally, but because it ignores the 50% rule. That rule only applies to "SOLID" tantalum capacitors, but it says that your cap's voltage rating should be TWICE the working voltage. Meaning, if C1 sees 12V normally, you need a 25V SOLID Tantalum.

Alternatively, you could use a 16V Polymer Tantalum because they have a different voltage derating. You get lower ESR too. Fewer worries about voltage spikes, especially because these caps fail open. (SOLID types fail shorted.) But the down side is these caps are outrageously prices most of the time.

Another option is to use Panasonic OS-CON at 16V or higher, and you get the added benefit of the lowest ESR next to ceramic. No worries about voltage spikes killing the cap either.

Don't use Ceramic caps even if you hear about other people using them as replacements, primarily because of the DC Bias effect that drops capacitance dramatically, in accordance with applied voltage. Even if you use a 50V rated ceramic, it's capacitance would likely be 20% lower than nominal, with 12V across the cap. Many other disadvantages to ceramics too when it comes to swapped out a stock bypass cap.

But without knowing the stock ratings and without knowing the voltage across C1, it's hard to make a sound judgment about a "PROPER" replacement.

All this talk of batteries got me remembering, at least Maxell apologized...


I use normal Nichicon electrolytics in all my recaps. If not them, then it's Panasonic. If not them, someone else who is from Japan with some kind of reputation.

JDW -- IIsi has a C1? https://recapamac.com.au/macintosh-iisi/

If you're referring to the IIfx, it should be 16v 47uF on a electrolytic board. Weird, I couldn't find a tantalum board image...

>> Mk.558 said:

JDW -- IIsi has a C1? https://recapamac.com.au/macintosh-iisi/ If you're referring to the IIfx, it should be 16v 47uF on a electrolytic board. Weird, I couldn't find a tantalum board image... Click to expand...

Oh, that photo was of a IIfx? I don't have a IIfx either. But what I wrote about tantalum capacitors still is correct info. If that C1 is across a 12V rail, it explains the burn.

>> JDW said:

Oh, that photo was of a IIfx? I don't have a IIfx either. But what I wrote about tantalum capacitors still is correct info. If that C1 is across a 12V rail, it explains the burn. Click to expand...

Yeah, that was from the IIfx. I actually buttoned it up after getting it cleaned, working, the case retrobrighted, and the battery holders replaced with a 3xAAA holder I can access without removing the drive bracket. Also, it allows me to put in rechargeable batteries which I've found much less likely to leak. Also, quick connects to allow me to remove them from the machine easily when not planning on using. If I pull it apart again, I will try to check the voltage on that cap. I did replace the tantalum with a OSCON cap.

>> Mk.558 said:

I use normal Nichicon electrolytics in all my recaps. If not them, then it's Panasonic. If not them, someone else who is from Japan with some kind of reputation. If you're referring to the IIfx, it should be 16v 47uF on a electrolytic board. Weird, I couldn't find a tantalum board image... Click to expand...

So was the tantalum board a different board? Mine is mostly tantalums, but right beside almost all of the big tantalums are pads and markings for an equivalent electrolytic SMD. I included a picture earlier in this thread. I assumed it was the same board, just populated with 2 electrolytics with the rest tantalums, or a bunch of electrolytics. Mine is now a bit of an oddball with 3 OSCON caps, although I did consider switching the rest to the solid OSCON caps...

On the plus side, other than what's looking like a failed floppy, I have the IIfx back to fully working order, and looking good.

Only the top cleaned:



And the finished product, running of a RaSCSI for now.



Like the PowerMac 6100 I retrobrighted, it didn't quite get it back to the original color (the color you can see that was under a sticker). I will likely do another round of retrobrighting on this when the weather warms in the spring. Hoping that it's lack of heat causing my retrobrighting to be less effective. We'll see. Also, I plan to take one of the PM6100 cases that is in very bad shape and make it black, like I have for the Performa 5400, PowerMac G3 All-in-one, and my Mac TV's mouse. If anyone has a inkjet printer with white ink who's be willing to help me print a waterslide decal, I wanted to try to test out using that method for white labels for the case.

Apparently, I'm all out of internal BlueSCSIs and I think I may need to manufacture a bunch of BlueSCSI v1s. the v2s are nice, but I don't think I can justify the cost. I thought I was able to mass build a batch of 10 BlueSCSI v1s for around $10 each. I'd be willing to make 20 or so and that should put a working HD in most of the machines I get working. Unless @eric wants to work out a sponsorship deal! :)

>> wottle said:

I may need to manufacture a bunch of BlueSCSI v1s. the v2s are nice, but I don't think I can justify the cost. I thought I was able to mass build a batch of 10 BlueSCSI v1s for around $10 each. I'd be willing to make 20 or so and that should put a working HD in most of the machines I get working. Unless @eric wants to work out a sponsorship deal! Click to expand...

Cost is a valid consideration, and bulk buys would be nice if they generated discounts, but I must say this. I've tested v2 with a wide range of Macs, and even on a slow, stock 8MHz Mac SE, benchmark apps across the board show the v2 to be faster. Noticeably? Well, that would depend on you. I believe I can feel the speed difference if I am looking for it.

I'm more worried about the habitual self-nuking that has happened on two 68040 machines, sometimes within single digit minutes of it being booted. It's made me reconsider BlueSCSI for internal applications.

IDK what a tantalum board looks like besides your image. A good image for an electrolyctic is here: https://recapamac.com.au/macintosh-iifx/

That IIfx looks sharp now. It's easy to forget how vast the II/IIx/IIfx is. It is enormous. It's why I like my IIci.

>> Mk.558 said:

I'm more worried about the habitual self-nuking that has happened on two 68040 machines, sometimes within single digit minutes of it being booted. Click to expand...

I had the same problem on my Centris 650 - it was infuriating! I had numerous times where I'd install 7.6, reboot, install the 7.6.1 update, reboot and... corrupted. Or I'd get it installed and then start un-stuffing a bunch of files (games installers), and then... corrupted again.

The recent rewrite of the SDIO code in the BlueSCSI v2 firmware seems to have resolved the issue for me <knock on wood>. I'd recommend installing the most recent firmware revision if you haven't already and give it another shot.

>> Mk.558 said:

I'm more worried about the habitual self-nuking that has happened on two 68040 machines, sometimes within single digit minutes of it being booted. It's made me reconsider BlueSCSI for internal applications. Click to expand...

I hadn't heard of that issue. I have probably 25 BlueSCSIs (almost all v1) and haven't had any issues, although most of my machines on 030 and slower machines. I'll have to keep an eye on it, as I also have BlueSCSI in my PowerMac 6115CD that I've had for a while. I don't boot the machine too often, but it's survived a handful of boots over the last few months.

>> Mk.558 said:

IDK what a tantalum board looks like besides your image. A good image for an electrolyctic is here: https://recapamac.com.au/macintosh-iifx/ Click to expand...

That looks to be the same board as mine. Note the pads for tantalum caps right beside the electrolytic caps.

>> Mk.558 said:

That IIfx looks sharp now. It's easy to forget how vast the II/IIx/IIfx is. It is enormous. It's why I like my IIci. Click to expand...

Yeah, the IIfx is massive. Sadly none of the IIci/IIcx machines I picked up are working. One was battery bombed, and 2 had significant corrosion / rust. I will likely tackle those after the 61xx machines.

>> Mk.558 said:

I'm more worried about the habitual self-nuking that has happened on two 68040 machines, sometimes within single digit minutes of it being booted. It's made me reconsider BlueSCSI for internal applications. IDK what a tantalum board looks like besides your image. A good image for an electrolyctic is here: https://recapamac.com.au/macintosh-iifx/ That IIfx looks sharp now. It's easy to forget how vast the II/IIx/IIfx is. It is enormous. It's why I like my IIci. Click to expand...

Only took about 14 months to forget, right? ;-) They're about the size of a home theatre receiver...

Machine number 2 fully working. I took one of the dirtiest, but otherwise in good shape 6100s, recapped the logic board, cleaned it up and put in one of the DOS Compatible cards.

Since no amount of cleaning and scrubbing seemed to get these cases clean, I decided to turn it into a black Mac, matching my Macintosh TV, Performa 5400, and black PowerMac G3 All-in-One. Pained the case, the feet, the back plate, and the CD ROM drive cover. Printed some replacement labels for the front and reinstalled it with a working floppy, working Apple CD 300i, and a newly created BlueSCSI v1. Working pretty well so far. And I think it looks pretty good. Certainly better than the original, even after cleaning.