• Updated 2023-07-12: Hello, Guest! Welcome back, and be sure to check out this follow-up post about our outage a week or so ago.

New theory on the cause of the infamous Tunnel Vision problem - Testers needed!

Paralel

Well-known member
You can bake it again, but be forewarned that doing do can stress the liquid crystals to the point where you start to get what's known as "pixel rot" or "liquid crystal rot". What happens is that the helical structure of some of the liquid crystals degrades due to heat stress. It will appear as small black dots after baking, that get bigger after a few hours as the degradation spreads. However, it is not usually terminal. Liquid crystals have self-healing properties to a limited extent. With some time, usually on the order of days if I am remembering correctly, the helix will usually come back together, and the pixel rot will shrink and disappear. Thermal pixel rot is not the same as the temporal pixel rot some people see with certain vintage LCD screens. Temporal pixel rot is a defect in the liquid crystal where the helix is lost due to chemical instability and is non-reversible. What we don't know at this point is whether thermal pixel rot is a risk for future temporal pixel rot, as in, does it possibly lead to temporal pixel rot in the future due to the stress that was imposed on it currently? No one knows. The perfection of the bake method for TV was only determined last year. We won't know for some time if the rate of temporal pixel rot in the future is provoked by thermal pixel rot in the present. It's possible one has nothing to do with the other. We simply don't know. I mention it because I believe one should be aware of all the potential current and future consequences before making a decision on what to do.

We had such a case of thermal pixel rot appear on the Think Different forum after someone did a repeated bake and the screen completely healed from said thermal pixel rot. The tunnel vision on that particular screen had reached a rather terminal state, completely unusable. Multiple bakes were necessary to bring it back to an even reasonably useable state. In the end the bakes worked to reverse the terminal tunnel vision back by about 50% and the thermal pixel rot completely reversed itself.

This brings up the point that baking for 4-5 hours at 100c works, but has its limits. It can only cause so much healing. After a certain point it isn't possible to bring a screen back any further with a single bake. Multiple bakes has worked to cause further reversion of the TV syndrome, but this has only seen success with severe cases. Minor cases that have attempted multiple bakes have not seen perceptible improvement. However, the number of cases to pull data from is quite limited. At this point I wouldn't rule out the possibility of further reversion of TV syndrome with multiple bakes, but limited anecdotal evidence thus far does not support it for minor or limited residual TV.
 

3lectr1cPPC

Well-known member
Just an FYI - that guy with terminal TV that got pixel rot was here, not on Tinker Different, it was earlier in this thread.
The temporal pixel rot described - I’ve seen it the most due to pressure damage, like the PB 100 LCD that the trackball likes to squish.
 

Paralel

Well-known member
Yeah, it's a bit unnerving to remove the LCD panel and seeing those fine contacts, but it does seem to get put back together. There is a little bit of wiggle room so you don't have to align it 100% (that's the part I worried about since the + symbols did not align for me). The one concern I have of repeatedly opening it was fatigue on the little metal tabs. I have some smaller LCD's (Z88) and they came in two varieties: metal tabs and screws. I would be more comfortable with having screws there. My TRS-80 Pocket Computer 1 also has metal tabs, but its bezel seems to be much thicker/stronger metal.

It's hard to know if a second attempt will fix it or put it at risk. I had someone post on my video that 100 celcius won't damage electronic circuits (he was asking why I took the panel apart). But there is concern that you start adding black spots to the middle of the screen, though I think someone here did that when the temp was higher than 100 celcius. For me, I was going to be unhappy using my BP180 with tunneling so I figured I'd look for something else if it didn't fix it (this was not my 90's Mac but one I purchased for $50 last year on eBay, so I had no connection to it). I had looked at the PB150 since it offered IDE and a passive LCD screen. The lack of ADB port was going to be a fun add-on hack that I found online (soldering four traces on the circuit board since Apple had implemented it but decided no to expose it to users...sigh.

If the fix is indeed about re-flowing the seals, I suppose the seals could be too far gone. After posting the videos above on LCDs, I'm even more of the belief that it's not water related since the liquid crystal itself would end up mixing with moisture -- I don't think it's like oil and water...would love to hear from someone that's played with the substance to get more insight.

I still find the seal-reflow theory unconvincing. If it was simply about seal-reflowing then multiple bakes shouldn't provoke further improvement, while longer bakes should provoke an improvement, but what has been seen so far is the inverse of that. Longer bakes beyond 4-5 hours are not useful and multiple bakes, in severe cases, have demonstrated improvement. I don't see how that can be explained with only seal-reflowing. Also, how does one explain why its always as the area of impact approaches the edge of the screen that the compromise is most severe with the seal-reflow theory? The seal should be uniform. One area shouldn't be more likely to be impacted over another, but the degradation always follows a very specific pattern, and the healing also happens in the same pattern but in reverse. Most of this however can be explained by moisture intrusion from the edges inward. I think if both theories are combined all the elements that have been observed can be explained.
 

bwinkel67

Member
I still find the seal-reflow theory unconvincing. If it was simply about seal-reflowing then multiple bakes shouldn't provoke further improvement, while longer bakes should provoke an improvement, but what has been seen so far is the inverse of that. Longer bakes beyond 4-5 hours are not useful and multiple bakes, in severe cases, have demonstrated improvement. I don't see how that can be explained with only seal-reflowing. Also, how does one explain why its always as the area of impact approaches the edge of the screen that the compromise is most severe with the seal-reflow theory? The seal should be uniform. One area shouldn't be more likely to be impacted over another, but the degradation always follows a very specific pattern, and the healing also happens in the same pattern but in reverse. Most of this however can be explained by moisture intrusion from the edges inward. I think if both theories are combined all the elements that have been observed can be explained.

If you watch the second video, the liquid crystal appears very miscible...if moisture is involved it would mix in with that and not stay separate. The seal re-flow in my opinion is about the temperature needed to get the sealant viscous again and it depends on what state it is in. The issue is balance, if you heat it up at a much higher temperature, the sealant would re-flow rather quickly but the LCD would be damaged. So you hit it at that 100 celcius point for some duration of time. But the pattern you observe is definitely not that of water, which would linger for much longer if it intruded into the liquid crystal space. Again, the video shows the kind of pattern you get through panel separation.
 

Powerbook27364

Well-known member
Just bought a duo 280 (active matrix grayscale) so hopefully I get lucky with this one! Friend bought a duo 250 (also active matrix grayscale) as well, so going to be interesting to see what happens with them when they arrive. At the very least, should help give some more information into the baking process on different models of display. Re-celling the battery will be a whole new adventure for me!
 

3lectr1cPPC

Well-known member
Good luck! I wouldn't hold out too much hope that they'll be free of the curse, most aren't. The Duos are all at least 16-color grayscale, so that's an improvement as well. They're quite uncommon systems, and they also need their logic board capacitors changed.
 

Powerbook27364

Well-known member
Got an amazing deal on it so even if it has some tunnel vision, it would have been well worth the price. Cost me just a little more than my 170, being only $62 plus shipping and included a small dock for scsi and adb and its original charger. Already have the equipment for spot welding and cells for my titanium and 145B/170 on the way, so I can just order a couple more cells for it.
Didnt realize they use logic board caps, so that will be another thing I need to order. Thanks for the heads up about that. Do you have any idea which caps they use or where I can find out? It will take about 2-3 weeks to arrive due to my location, so I have plenty of time to order capacitors for it.
 

3lectr1cPPC

Well-known member
Cost me just a little more than my 170, being only $62 plus shipping and included a small dock for scsi and adb and its original charger.
Excuse me??? How??? I've been looking for a Duo w/ a SCSI dock for over a year now but nothing for a price I'm willing to pay. Then you manage to find deals like this. Wow. Congrats on that, that's huge.

I'm unfortunately missing the logic board cap values for the 280/280c as of now, I've got others up on my website, but the 280s use a different logic board than the 68030 based Duos do. I just did some research and cross-referenced between the values I have for the Duo 2300 logic board and an image of the 280c logic board with some values visible (https://68kmla.org/bb/index.php?threads/powerbook-duo-280c-no-chime.37803/) and I've been able to see that the 280 logic board very likely uses the same caps as the 2300 LB does, but I can't confirm this for sure.
Here's the values for the 2300: https://macdat.net/cap_reference/apple/powerbook/duo2300.html
I'm 90% sure they're the same as the 280, but it would be great if you could confirm this once you get your duo in, it would really help with my reference site. Bonus points if you can get me an image of the 280 logic board to use on the reference page.

Hope this helps!
 

Powerbook27364

Well-known member
I was shocked that I got it for so low! I fully expected it to go for well over $150 at least! Its my first duo and I was not willing to pay ebay prices so it took a long while of waiting to find. I will have to look around a bit to see if I can find the values. Hopefully everything works well when I get it. I will probably order those caps just to have them in case they are the same. Ill be sure to send pictures and values of all the caps and other parts of interest. I will make sure my friend does the same with his 250, though he should get that in a couple of days instead of 3 weeks, so that will be much quicker.
I'll be sure to let you know if I ever see a good priced one again, though that might not happen for a while. side note, but your website is really great, thanks for keeping it updated!

I may send a message on that post just to see if they still have the values they used
 

3lectr1cPPC

Well-known member
Thanks a bunch! Images especially are helpful, in order to make them into reference images for the site, such as what you can see on my PB 100 page and others: https://macdat.net/cap_reference/apple/powerbook/100.html
250 logic board is even more urgent in terms of caps, as it's older and so they leak more often. 250 should be the same logic board as the 210 and 230 though, which I have the values for: https://macdat.net/cap_reference/apple/powerbook/duo230.html
LCD will be different though.
side note, but your website is really great, thanks for keeping it updated!
Thanks a bunch! I'm making it for folks like you, I'm just glad to see that people are finding it useful.
 

Powerbook27364

Well-known member
Thanks for the info! I'll be sure to let my friend know about the caps, and about your website too. He will be able to provide images as well, so just in case they are different you can have images for that too. I will make sure to get the values for the LCD as well. I will link him to this thread and hopefully his duo will arrive this week and he can have the duo 250 images to you by the weekend all going well.
 

Powerbook27364

Well-known member
I just remembered, but the 170 I have does uses tantalums for the display, so there is a chance that these duos may as well, but they are grayscale compared to black and white, so I'm not entirely sure on this.

Wow that's a lot of caps on the 100 boards :oops:. My friend bought one recently and it hasn't arrive yet, but he will be in for a shock!
 

3lectr1cPPC

Well-known member
To my knowledge, none of the Hosiden LCD panels use electrolytic capacitors. You both should be fine there. I could use the LCD P/N's for the 250 and 280 LCDs for this list too, if you do take the LCD assembly apart to fix hinge mounts or anything like that: https://macdat.net/extras/powerbooklcdindex.html
Wish your friend good luck fixing his PB 100, he'll need it. I've gone through that mess once, and it's a huge pain, but I managed it. Main issue is that a bunch of them won't come back to life even after new caps are installed, they can be really finicky, and absolutely won't do anything until the caps have been changed.
Thanks so much for helping with the images! An image of the 250 logic board means I'll have an image for the 210, 230 and 250 as they're all the same.

Edit: https://macdat.net/macintosh/powerbook/100/issues.html
This page should also help him out with other issues with the 100.
Just one note though, I'm in the process of re-arranging the directory structure for the PowerBook pages on my website right now, so whenever I push that update live, this link will break. I'll send the new one in a reply when that happens.
 

Powerbook27364

Well-known member
He got it in "working" condition from facebook marketplace for $60 with floppy drive and charger, but there were no pictures of it powering on, so hopefully its working. He has some previous recapping experience so hopefully he should be fine with the process though that looks like a massive pain! Hopefully the images will help people even though duos are relatively uncommon.
 

3lectr1cPPC

Well-known member
Uh huh, "working". Right. I've seen a grand total of one singular 100 that hasn't been recapped show up "working" on eBay in the last two years. So, I highly doubt that, especially if there were no pictures of it on. I'll bet the seller plugged it in, saw the screen light up blank/blue and the hard drive spin up and called it working. Hopefully new caps will get it working either way though.
The PowerBook 100s are particularly sensitive about having working capacitors for them to work, unlike an SE/30 that will probably still run with leaking caps as long as the leakage hasn't caused a terrible amount of damage yet. You'll never see a 100 running in that state.
 

Powerbook27364

Well-known member
Yeah that's what I thought as well. I'm quite surprised it even turned on at all since the power adapters normally fail as well. No idea about those lead acid batteries though. I know there was a place on ebay selling them a while back, but they are pretty expensive.
 

Powerbook27364

Well-known member
My original PSU works fine currently, but I try to avoid using it unless I need both my 170 and 145B running at the same time for something. I plan to recap it but probably only after it stops working entirely. Those are not cheap batteries, but good to see they are still available if he ever has the need for one. I am going to recell my powerbooks (apart from my g4 12" since it works) and hopefully help another person locally with it as well. Unfortunately I havent been able to find anyone offering new batteries for the older laptops apart from the 100, so DIY it is!
 

3lectr1cPPC

Well-known member
The new 100 battery is great because it uses modern lipo cells, so you don't have to worry about lead acid charging characteristics, plus extra capacity. I'd also assume that you could swap the LiPo cells out when they eventually die a lot easier than it would be to rebuilt an old NiMH pack. I've been meaning to learn how to rebuild batteries for a good while now, over 2 years. I'm going to give it a go eventually for my PB 500 series packs.
I've also got a couple original PSUs, although they're later ones. I've got a 3 Amp high-power version for the color 100 series (which I need to run my 180c) and a later "Low-Power" 2A charger that shipped with the 150, circa 1994. Both work fine but I don't use the low-power one as I have aftermarket ones, and I'm going to recap the 3A eventually. I've got to recap my 500 series charger first though, I've got two of them and neither work properly.
 

Powerbook27364

Well-known member
Oh nice, I thought they were lead acid still. I wonder what the battery life would be like with one of those. The NiMH packs arent too bad for the original 100 series, I have a simple wiring diagram and picture here and they can be remade with new NiMHs for much greater capacity.diagram.jpgimage.jpg
 
Top