IBM PS/note N45 SL teardown

[solidpro]

Ok, not a thinkpad. More of a silly uncle.

The PS/note N45 SL was priced at $2,045 and contains a 25MHz 386SL. It had 2MB RAM and a 80 or 120MB HDD and was equipped with only 6-row keyboard without dedicated navigation block, the similar layout as a low-end ThinkPad 300 laptop. PC Mag considered the display a disappointment, but noted its good design and performance. It was manufactured by Zenith Data Systems.

I have 3 of these and have been putting off trying to determine 'what can go wrong' due to their age, non-standard design and brittle plastic. The point of this post is to just report on what I found, what other owners should look out for and some questions that some others might know the answer to.

It feels very much like the IBM Thinkpad 300, which is made from the same plastic and indeed made by Zenith. The power supply is totally unique and has a connector which is a bit like an S-Video connector. I hate it.

The plastic on the case also is starting to show loads of 'swooshes' where areas have seen more wear than others.

The LCD

The LCD is a monochrome affair presenting 640x480. My one was really flickery so I decided to see what the capacitor issue was like. Getting to the screen is actually quite simple but you do have to pry the old plastic apart to release clips around the edges. I actually didn't realised there were 6 screws holding the LCD bezel onto the back - there are 4 under covers underneath the screen and 2 screws under rubber pads either side of the top of the LCD. Take these 6 screws off and the back will come off. Note also there are actually 4 sets of hinges in this machine - maybe they started to realise that the hinges were a big problem! One set bolts to the LCD itself from the back and one set bolts from the fascia to the back from the front (under those 2 plastic panels).

There were 6 'through hole' electrolytic caps which were soldered on sideways to the LCD itself. No other caps at all! They were easy to remove and solder on new ones except there are 3 ribbon cables which have to be carefully de-soldered and lifted before you can access the PCB with the 6 capacitors on.

This is the back removed:




Here are the plastic clips along the top of the lid, which were the hardest to loosen:




Here are the 3 ribbon cables you can see which are ensuring you can't get to the 2 caps top and 4 caps bottom:





The caps were 4 x 22uf/16v, 1 x 4.7uf/50v and 1 x 10uf/16v. All have to be really small to not obstruct anything.

Amazingly, first try all my soldering was fine. I spent £400 on a Thermaltronics solder station which doesn't even have a temp control - it is completely self regulating and I have to admit this first go was bliss!

The base

I wanted to know if there were timebomb caps inside, what hard drives was in there and if there was a hibernation battery. There were 4 electrolytic caps inside and they all looked OK. I decided to park them for another day.



I managed to determine that:

1 x 100uf/16V
2 x 1000uf/6.3v
1 (small surface mounted tiny cap, unknown values)

The hibernation battery. I have never seen a battery like this before. It weighs like it's made from lead and consists of 4 x long, rectangular blocks bound together in plastic. At first I wasn't sure what I was looking at because it just says "JAPAN WD" on the outside and was next to the hard drive. But I could tell from the red and black wires coming out of it it was battery related. Anyway, I removed it and the machine powers on and behaves the same without it but it was showing plenty of signs of leakage.





Has anyone seen this battery type before?

The hard drive. The machine wouldn't boot from the disk but now I've removed it, I have a feeling it's just because there is no CMOS, no built in BIOS configuration and definitely no auto configuration. It's a connor CF2084. The PIN connector looks exactly like every other PATA/IDE 2.5" hard drive i've seen. Does this mean it is IDE? I put in a CF card on an adaptor but I haven't yet gone to the trouble of downloading a diagnostics disk to try and configure it. Thats for another day. Is an IDE drive going to work in this machine?

Memory

Does anyone know what this memory is?



I've never seen it before. I don't actually know yet how much is in this machine but it's weird there are 3 of them.

Finally here are some pics of me during the teardown:





Anyway, it seems like these machines aren't that scary and are relatively easy to teardown. I just need to do a little more research before I can truely consider it preseved.

[solidpro]

I did find that there is no configuration disk. You have to run the PREP command using the advanced diagnostics disk in order to get it to find the hard drive and make it ready to be used...

http://ps-2.kev009.com/eprmhtml/eprm2/h3232.htm

[astral]

dead conner drive probbaly has a rubber bumper inside thats gone sticky. tape over it and it may work again. that surface mount cap at the very least needs to be replaced for it to be considered preserved, it looks ok, but its not. trust me. hinges are a bad design again it looks like, those standoffs in the display panel will end up cracking im sure. having to desolder ribbons to get to the lcd caps is a poor design as well hah. doesnt surprise me coming from epson... ive heard bad things about the epson screens in general. maybe it's a rebadged epson laptop? they made their own back then.

how's the keyboard on that one?

battery looks like a nicad probably. lead acid batteries usually dont leak like that, looks like nicad or nimh leakage to me but a laptop that early is probably nicad.

[solidpro]

It's a Zenith laptop. I agree about the surface mount but I ran out of time. I'll be doing that next week.

Re: the connor driver - what do you mean 'tape over it'? Tape over what?

The keyboard is really nice to use. It doesn't feel very 'thinkpad' but it has good feedback and is soft but not 'damp'...

[Franchute13]

Re: the connor driver - what do you mean 'tape over it'? Tape over what?

Quantum has the same problem on some models.
https://www.youtube.com/watch?v=FQrMXOEuSbE

[astral]

Not sure exactly which style of Conner drive that is, but the bumper is usually pretty obvious. Either black rubber or some I’ve seen use a clearish colored one. you can go the proper route and remove the old rubber and replace it, or just tape over it so the heads don’t stick. also some conncer drives didn’t have that rubber, so it’s possible the drive is just cactus.

[RealBlackStuff]

Here is some info about your 80MB Connor CP2084 and some more IDE info.

[solidpro]

Not sure exactly which style of Conner drive that is, but the bumper is usually pretty obvious. Either black rubber or some I’ve seen use a clearish colored one. you can go the proper route and remove the old rubber and replace it, or just tape over it so the heads don’t stick. also some conncer drives didn’t have that rubber, so it’s possible the drive is just cactus.

Ah, the conner drive I pulled out has these two rubbery crusty things sticking out of the side. One snapped off but the bottom half of both are both sitting in the drive. I'll check that video and link later and see what I can do with it....

[solidpro]

Having watched the video I see what you mean about 'bump' stops now. Definitely something I can do but I'm concerned that whatever I replace it with will do the same thing in 30 years...

[solidpro]

Took apart the Conner drive and it doesn't have any rubber bump stops...

[astral]

I think it does. The thing the arm coming off the right of the head stack is touching in photos one and three. I believe you’ll find that the head sticks a bit to it if you carefully try moving it with your hand. Place a tiny piece of tape over it and it may work again.

[solidpro]

I moved the arm with my fingers and it was free.... the little hole through which the arm sits at is just another piece of metal.... I think the drive is spinning and seeking OK, it's just not reading for some other reason.

[astral]

You're probably right. Just making sure though, the stick is just a tiny bit, not a lot. the arm doesn't have much torque so if it's stuck even a little bit, it won't be able to free itself. It should be completely free when you try to move it away from that suspected rubber or metal.

[RealBlackStuff]

If you need them, I still have quite a few IDE drives on my shelf.

[solidpro]

If you need them, I still have quite a few IDE drives on my shelf.

Thanks for the offer. I invariably end up buying them on ebay or Aliexpress from China but it's getting hard to find smaller capacity ones now...

[solidpro]

I decided to bite the bullet and finish the job!

This meant that I needed to remove that DC/DC board inside, remove all the capacitors, clean and replace as well as figure out the hard drive.

First the caps, the surface mount one was completely rotten underneath but it cleaned up well. Also at least 2 others smelt of fish and had dark residue underneath, so the thing was definitely on the way out. For anyone who has an N45 SL and is thinking of doing the same you need the following:

Through Hole:
C602 330u/16v
C303.1 and C303.2 1000u/6.3v (2 of)
C404 100u/16v
C112 100u/6.3v
C108 10u/6.3v

Surface Mount:
C504 47u/25v (easy to replace with a through hole or Tantalum)

A nice surprise after replacing these caps is that the original Connor drive suddenly sprang to life and started working. Maybe the DC/DC board wasn't powering the 2.5 IDE bus properly to give it the power it needed. It booted up into German Windows 3.1 no problem.

So I put it all back together, with only 1 screw left over and called it a day!

[astral]

great work! not surprised at all that the caps were bad as usual. they're the worst.

[solidpro]

Urgh you should have seen the muck and gunk underneath. I've got 2 more machines to do and I'll take some photos then. It was disgusting. There is an art form for getting it all off without damaging the silkscreen, traces or pads!

[astral]

yeah I’ve been there. It takes ages to get pads clean sometimes.

[solidpro]

I did the second and third machines this week. Both of them had not suffered as much. The first machine came up smelling of roses - recapped everything including the LCD and no problems at all. Hard drive was dead though.

Second machine a different story. I think it's been pillaged before because it had no RAM and the ram is really weird. Then I tried to recap the screen but I made the mistake of desoldering the entire LCD PCB which is a total of 3 ribbon connectors and some of the pads came up. Seems one end of one of those connectors is so fragile it happens. Anyway, after doing all the recapping of the LCD, I made it worse and it just displays garbage now. I could give it another go but seeing as I have 2 good ones and one machine which is missing bits I'm just going to keep the third machine for parts.

It does seem quite similar and at the same time a bit different to the 300, which I have a perfect, but dead example. So I'm going to have a go at that in the next month or two.

[turbinee]

Hi solidpro, very nice post!

I recently got the same ps/2 note, but dont have the 4 pin power supply to start it up.

Is there a way to use regular +/- power supply to get this thing running again?

If any of you know a workaround let me know, thank you

[solidpro]

Hey - If I am remembering correctly, as with the 300, you can either try and hunt down the zenith psu or infact I did modify a S-Video connector to use as a power feed. You have to snap off the bezel which leaves the pins very fragile and exposted but I did use that as it's around the right dimensions with 4 of the pins in the right place to (delicately) run power to it...

[turbinee]

Thank you for the reply.
I read the instructions on the other tread about the s-video connector, I will try the same. My problem is understanding what to do with the 4 pins? I see in the other tread, that you've said that top left is ground. I am looking at the manual for this model and ground is top right on the schematic, and also there is a signal pin.

Can you tell me what you think:
https://imgur.com/a/1GeBWGM