ThinkPad 350: hinge and shell repair

[panips]

[warning: text generated by GenAI ]

My collection of five ThinkPad 350-series laptops (including the PS/Note 425 and 350c) is facing a common, critical issue: the poorly designed and prone-to-failure hinges. My initial structural fixes were only temporary, failing quickly upon use. I realized that 3D printing the entire chassis wouldn't solve the problem, as the key weakness is the hinge mechanism itself. I am now actively developing a better, more durable hinge solution for these classic machines.

It's worth noting that, in contrast to the external issues, the electronic stability of the 350 series is remarkably high; all five units were functional after minor repairs. The greatest internal challenge encountered was the prevalent 'vinegar syndrome' affecting the displays. This issue is easily mitigated with the correct polarized film replacement, or by using compatible donor panels such as the Sharp LM64P873 (monochrome) or the Sharp LM64C052 (color DSTN).

Returning to the hinge issue: The core problem is a combination of excessive hinge tension and inadequate structural support from the chassis mounting points, which are simply too weak to handle the twisting force. I've developed and tested three distinct solutions to address this:

• Custom Spring Tensioning: Creating or modifying the hinge spring for lower resistance.

• Adjustable 3D-Printed Spring Substitute: Designing a printed component to replace the spring and allow tension tuning.

• Full 3D-Printed Chassis Support Component: Reinforcing or replacing the entire mounting bracket.

I can confirm all three approaches are viable fixes, though there are subtle performance differences. I'll share my recommendations and a visual comparison of these methods shortly.

The Results: Post-Repair Showcase
Before diving into the comparative analysis, I present the visual outcome of the complete restoration:




Solution Breakdown: Three Approaches to Hinge Tension

Custom Wire Spring: This is the most direct fix, involving crafting a replacement spring. The required diameter is approximately 8.5mm (<9mm), yielding a final component with lower resistance than the original. The main drawback is the tooling requirement and the high likelihood of trial-and-error failures before achieving the perfect tension.


Adjustable 3D-Printed Clip: This solution involves printing an adjustable clip that utilizes the existing hinge 'teeth' for fine-tuning and securing the tension. It offers a cleaner, more controlled result, but successful installation typically requires post-processing (sanding or Dremel work) to ensure a perfect, flush fit.
Model file: hxxps://archive.org/download/ThinkPad350HingeFix/adjustable%20clip.stl


Full 3D-Printed Support Component: This is the 'set-it-and-forget-it' option, replacing the entire weak mounting structure. While requiring no further adjustment, it presents unique challenges: due to 3D printing deviation and filament shrinkage/contraction, you may need to print several copies to find one with the ideal fit and long-term stability.
Model files:
hxxps://archive.org/download/ThinkPad350HingeFix/Left.stl
hxxps://archive.org/download/ThinkPad350HingeFix/Right.stl

[panips]

More About the Shell Repair
Now for a closer look at the shell repair process. For areas with severe cracks or even missing material, the fix can be challenging.


Building Structural Support
We start by building up the initial support structure. This is a rough shape meant to provide a base for the subsequent application of plastic welding and epoxy.


While a more robust solution might involve adding steel wire or mesh reinforcement, I found this method to be sufficiently strong for my needs.


After some careful work with a Dremel and polishing tools, the area is now structurally sound and visually improved.



Fixing Standard Cracks
For simple cracks, instead of just gluing the parts together, I recommend adding internal support for long-term durability.



We use a solder iron to weld the glued sections together, strengthening the bond significantly.



This is then followed up with plastic welding to completely integrate the repair into the shell.



Small Part Repair: The Hinge Cap Clip
Smaller, delicate parts like the hinge cap clips are easily damaged. For these, a metal sheet reinforcement works best.

Glue the broken part first.


Reinforce with a Metal Shim: Use the soldering iron to secure a piece of metal sheet (I cut mine from a Diet Coke can) as a reinforcement.


Final Results on a ThinkPad 425
Here are more pictures of a ThinkPad 425 after these repairs were completed:



The LCD now opens smoothly, and the hinges hold the screen firmly in place. While the exterior currently looks a bit rough, I plan on adding more putty patches and paintwork to finish the cosmetic restoration.

[solidpro]

This is absolutely brilliant. I'm going to run off a few prints myself of "Full 3D-Printed Support Component", but I wonder what is the best material if I use PCBway? I often just guess my way through materials on there, but they do offer a huge amount, including blending...

Also, have you considered applying his method to the 700C? Surely that is much more of an important machine but suffers from the exact same issue? I'd love to be able to do this to restore some of my 700 machines back to working condition as I currently have to store the lids seperately from the bases.

[panips]

I just use the normal PLA filament for such kind of 3d printing. I always want to have a relatively cheap and sustainable solution, so I didn't try pa-cf and some other materials.
700/720 are almost the same, I have fixed 3 of them, but due to its design, it's hard to have some reinforcement based on existing structures. I can try it again later, but before that I'd finish the 350/425 battery part first

BTW:
1) 425 and 350 are reportedly the same, but my 425 has no PCMCIA support, while 350 provides one type II slot.
2) 350 has the Sharp LM64P873 DSTN panel, and IBM Type 2141 PS/note 182 uses Sharp LM64P805. They are exchangeable.

[Voodle]

More About the Shell Repair
Now for a closer look at the shell repair process....

That's an excellent write-up and a great result, thanks for sharing
Very nice crafting of the missing section with tape covered flat panels? - I do the same when repairing areas where the hinge mount is missing. Blutack is pretty good for controlling where the epoxy goes if you're dealing with a rounded shape.
I've done similar repairs and they're holding up fairly well, if it's a big area then now I'd try to use a 3d printed part with epoxy joining it to the plastic, often with some straight bits of metal in the epoxy or plastic to improve the ridigity.
Rebuilding hinge mounts with epoxy and blutack:
https://www.vogons.org/viewtopic.php?p=1363678#p1363678

Using a 3d printed part along with epoxy to make a rigid hinge mount, print orientation matters (sadly the mainboard still doesn't work on this one):
https://www.vogons.org/viewtopic.php?p=1356382#p1356382

This is absolutely brilliant. I'm going to run off a few prints myself of "Full 3D-Printed Support Component", but I wonder what is the best material if I use PCBway? I often just guess my way through materials on there, but they do offer a huge amount, including blending...

Also, have you considered applying his method to the 700C? Surely that is much more of an important machine but suffers from the exact same issue? I'd love to be able to do this to restore some of my 700 machines back to working condition as I currently have to store the lids seperately from the bases.

It's a small solid part, any filament should be fine - PETG has better temperature resistance but is less rigid than PLA. Being in a screen the temperature should not be an issue so PLA should be fine.

[astral]

Watch out for the display shell breaking. The first thing that my 425 did after I fixed its base hinge mounts was to form a big 'ol crack in the display housing around the right hinge, which I had to stick an ugly blob of epoxy over.

Edit: Ah, missed your new second post - I see that did happen to you and your solution looks pretty clever! I did epoxy reinforce inside my display shell, but not with metal bits like you did.

Also curious - how do your 3D prints for the hinges themselves fix the tightness issue? Both of my hinges are still functional, but are quite stiff just due to their terrible springy design.

[panips]

I have lot of failed attempts in 3d printing some support, i.e. for the missing part, better to 3d print a rough piece based on good one and then apply the epoxy or glues, but I switch to an easier solution with epoxy supported by blutack or plate and tapes.

Watch out for the display shell breaking. The first thing that my 425 did after I fixed its base hinge mounts was to form a big 'ol crack in the display housing around the right hinge, which I had to stick an ugly blob of epoxy over.

Edit: Ah, missed your new second post - I see that did happen to you and your solution looks pretty clever! I did epoxy reinforce inside my display shell, but not with metal bits like you did.

Also curious - how do your 3D prints for the hinges themselves fix the tightness issue? Both of my hinges are still functional, but are quite stiff just due to their terrible springy design.

The GenAI missed some details on the tightness adjusting part, so for the printed "clip", it looks like


the blue arrow points to the support of building this part, remove it and you'll get the clip. And then, you need to fit it to other parts:


blue circles indicate the parts you need to polish with sanding papers, for different 3d-printers may not print exactly and accurately. Also make sure four legs work well to secure the clip.


and the last step, use the pliers to adjust it until you feel it can hold the lcd without breaking the shell.


I planned to replace all these springs, as you have also addressed, it's such a terrible springy design.

[astral]

So these fully replace the springs? Wouldn't that cause the plastic part to wear away very quickly, or is the mechanical strain placed somewhere else?

[panips]

Yea, cuz I found it's the easiest way. I guess it can be opened and closed for thousands of times before it's worn. Or, you can easily 3d print another clip when it stops working.
Another work around I tried is to apply one more piece of metal sheet (yes, from Diet Coke can) inside the clip, but I don't think it helps a lot.
And, you may also try the cinch clamps rings if you have concerns about the durability of 3d printing material, it also works but it's ugly .

[astral]

For the low amount of use a computer like this would see these days, and with how easy it would be to print replacements, it should be a pretty decent fix. I may get some of these printed to try on my own PS/Note!

[panips]

IBM Type 2141 PS/note 182 and ThinkPad 350 have very similar design and many of their parts are swappable, so I copied the previous 182 hinge cap and remade it in TinkerCAD.

hxxps://www.tinkercad.com/things/if1Z2T3oAr1-35 ... OhWjXU24jU


and here is the prototype of battery pack for 350, while I am still waiting for the battery contact plate and spring

hxxps://www.tinkercad.com/things/d4YeSnXsMMD-35 ... Zq4IeagYnA

[panips]

I just finished rebuilding a battery pack for the IBM ThinkPad 350 using salvaged contacts and a custom 3D printed frame. There are two ways to go about this depending on your original battery type, but here is the method I used to get a working, rechargeable solution.

1. Identify and Salvage the Contacts

First, check the battery that came with your laptop. If it matches the style in the photo below, you’re in luck! It will save you a lot of time. Carefully salvage the contact assembly and set it aside for reuse.


2. Understanding the Pinout

Here is the breakdown of the contact plate. From left to right, there are four pins:
Pin 1: Positive (+)
Pin 2: Battery ID (Connected to Negative via a 20k resistor)
Pin 3: Thermal Sense (Connected to Negative via a 10k NTC Thermistor)
Pin 4: Negative (-)


You will need to trim the salvaged contact plate to fit the 3D model and solder your components (the resistor and thermistor) accordingly.

3. 3D Model and Assembly

You can download the 3D model I designed here: hxxps://www.tinkercad.com/things/d4YeSnXsMMD-35 ... Zq4IeagYnA


You will also need some generic battery contact springs/plates:


Assembly Tip: There is no undercut (internal groove) in this 3D model, so you will need to manually press the clips on the metal contacts to ensure a tight friction fit. Once all the small parts are inserted and secured, it should look like this:



To finalize the build, I recommend using a plastic welder or a bit of hot glue to seal the assembly and ensure everything stays in place.


4. Testing

Now for the moment of truth! Load in your rechargeable cells and use a multimeter to verify the output is approximately 10V.


Finally, test it in the ThinkPad 350. Verify that the system boots and, most importantly, check that the "charging" indicator comes on when the AC adapter is plugged in.


Happy rebuilding!

[update]
1. Sourcing the NTC Thermistor: The component is likely a 10k B3435 NTC Thermistor. These are still very common and easy to find on eBay or electronics suppliers for a few cents.
2. Fitment Note: Due to the original design of the laptop's battery bay and the tolerances of 3D printing, the pack can sometimes be a bit of a tight fit and may get stuck. If you find it difficult to remove, you can simply use a credit card to help or just leave it in the machine as a semi-permanent AA cell adapter...this makes it much easier to swap batteries without messing with the enclosure itself.
3.Work in Progress: I am currently developing a second method for rebuilding these packs, I will post more updates once that prototype is finalized.

[solidpro]

I have had 4 versions of the "Full 3D-Printed Chassis Support Component: Reinforcing or replacing the entire mounting bracket." from PCBway - 2 sets in nylon and 2 sets with an ABS mix to compare.

Before I dismantle a 425/350, I'm slightly confused how to apply these L/R pieces - is it re-using the centre of the existing metal hinge mechanism?

I'll also have a go at this on a 182.

Cheers

[panips]

I have had 4 versions of the "Full 3D-Printed Chassis Support Component: Reinforcing or replacing the entire mounting bracket." from PCBway - 2 sets in nylon and 2 sets with an ABS mix to compare.

Before I dismantle a 425/350, I'm slightly confused how to apply these L/R pieces - is it re-using the centre of the existing metal hinge mechanism?

I'll also have a go at this on a 182.

Cheers

the original hinge has 3 pieces (be super careful when tear it down, the spring may hurt you)


and the 3d model you printed could be the replacement of left two


so all you need are


just simply insert the right piece to the 3d-printed one, and you are all set


HOWEVER, just like what I have mentioned, due to 3D printing deviation and filament shrinkage/contraction, and the deviation of the original piece, you may need printing several copies and pick up the "right" one, not too loose while not too tight. Or, if it is too tight, you may use sand paper to adjust it a little bit.
What I really recommend is the second solution, a 3d-printed clip. It's a replacement of the spring


insert the clip and make sure it's secured by its four legs


and insert the last piece, so they look like


if it is still too loose, you don't need to 3d-print one again, instead, you can use the pliers to adjust it until you feel it can hold the lcd firmly while not break the shell, like this


or like this


so install them as normal, and then test the lcd without the cable to see if it fits

[solidpro]

Ok, this is all great work. I’ve had a chance to do some of my own ground work using ‘panips’ models and I think I have some things to note – some of which overlaps panips’ response to me today!

The second solution is a 3D printed clip to replace the spring, which uses friction instead of tension to control the stability of the lid.

The third solution is a 3D printed block to replace both the base metal and the spring. At first I thought by calling it 'set and forget' that the laptop would have to be set in an open position and left - which isn't the case, it also uses friction just like the second solution.

These are excellent but they both rely on the standoffs in the base and the lid to be intact, whereas in nearly every late PS/note or 350 I've seen, the brass standoffs have cracked and broken. Panips - I think your design could be evolved slightly...

This is what I'm working with as a base to your design - a base that's cracked and has a previous repair (which I removed):



To line up with the hole for the hinge thing to pass through it needs to be on those terrible standoffs -



On the other side, I tried your spring replacement block - which does work well but with the same issue, needing a standoff to sit on:



So, I think a refinement for the 3rd solution - the 'set and forget' one would be to assume the remnants of any standoff are going to be dremeled down to a flat surface - in other words, the block would be fatter at the base.

Finally, the other end of the issues are with the lid. The same issues with standoffs. I wonder if a block could be easily produced which could be epoxy'd into the position of the standoffs and with a soldering iron, the original brass female bases could be sunken into it (like Polymatt would do to re-use the standoffs).... Here is my lid:





On the lid I have, essentially, nothing to screw the original metal pieces into due to the brittle plastic. I think if I could 3D print a replacement standoff base (firmly epoxied into the lid avoiding overlap with the LCD front bezel, but as big as possible to create the biggest foundation), I'd be able to actually connect back to the lid.

I *really* need to learn how to use a 3D design tool like tinkercad.... what are you using?

[panips]

Your image links seem broken?
I was using SolidWorks before, but I realize that designers might not have such powerful tools and they might use some really primitive ways to build the model back to 30 years ago. So I decided to rebuild everything in TinkerCAD using only basic shapes. It’s been an iterative prototyping process, so these models have gone through quite a few versions already. They’re still a work in progress, and I’d love to get your feedback on how to improve them.
For 350/350c, I have some models based on which you can remix:

the hinge caps:
https://www.tinkercad.com/things/if1Z2T ... OhWjXU24jU

the hinge solutions:
https://www.tinkercad.com/things/3UAcus ... 0lbePVRPAA

the battery:
https://www.tinkercad.com/things/d4YeSn ... Zq4IeagYnA

350c/425c LCD lid: (test and prototype...too big to print on most 3d printer)
https://www.tinkercad.com/things/9s2mke ... 5c-lcd-lid

etc...

[solidpro]

Thanks for the links. In the new year I'm going to spend a week trying to make my own replacement bases for the 350 lid and then we'll have an end-to-end solution. I've seen a few types of software to use, but I forget most of them and tinkercad has the most memorable name, so I think I'll start on that.

What's the difference between the original and the remix?

I think my pictures are now working?

[panips]

Happy new year! Yea, I can see all the pictures now.
Hmm, the standoffs seem easily broken, but it's also easy to fix with some plastic welder, i.e. J-B Weld (plastic bonder high strength structural adhesive 15 min black, or Professional Size Steel Reinforced Epoxy Dark Grey - Hardener and Steel pack)
I always fix the standoffs before any further repairing, and it never be a problem for me (however, if it's so badly damaged, a better idea might be to 3d print the whole piece of the shell). In fact I have no idea on how to improve existing design if the standoffs are dremeled, eventually the 3d-printed piece needs to be glued/secured to both ends using some, for example, plastic welder. How about using more plastic welder to "wrap" everything, including the brass base?

[solidpro]

Ok, so I've attempted to rebuild my lid and for this attempt I'm using Araldite. As you can see I've tried to get the bases back into original positions. I think what I've learned in this test run is to get some kind of malliable putty to 'box in' where the adhesive is going to sit, to contain it to a specific area and....

I've now got the issue where it's hard to get the original inner bits lined up with them as the friction 3D prints are now tightly in place and I am a few mm out. So I've also learned to line everything up before fixing it permanently in place. Still trying.

Here's the lid:





Sorry for the tiny images but this archaic forum with it's bizarrely out-of-date ideas about images force me to make them ludicrously tiny.

[RealBlackStuff]

The Forum rules say:
It would also be greatly appreciated if you can keep the pixel dimensions to something around 800x600...
Yours are quite a bit smaller!

[panips]

Ok, so I've attempted to rebuild my lid and for this attempt I'm using Araldite. As you can see I've tried to get the bases back into original positions. I think what I've learned in this test run is to get some kind of malliable putty to 'box in' where the adhesive is going to sit, to contain it to a specific area and....

I've now got the issue where it's hard to get the original inner bits lined up with them as the friction 3D prints are now tightly in place and I am a few mm out. So I've also learned to line everything up before fixing it permanently in place. Still trying.

Here's the lid:





Sorry for the tiny images but this archaic forum with it's bizarrely out-of-date ideas about images force me to make them ludicrously tiny.

OK, if I understand it correctly, you may 3d print a plate to help you to fix the mounting point:



hxxps://www.tinkercad.com/things/jXfzcfM4lDQ-35 ... chor-plate

[panips]

More about the battery...
If you don't have an original battery on hand, you can easily fabricate a functional replacement using 3D printing and basic electronics. Follow the steps below to build your own.

1. The Foundation: 3D Printing
First, print the battery housing. This model serves as the structural frame for your contacts and internal components.
hxxps://www.tinkercad.com/things/eAx3TRi8VZ7-350-battery-kai

2. Fabricating the Contacts
To ensure a solid electrical connection, you will need to create custom contact plates:

• Use 0.1mm copper foil for the best balance of conductivity and flexibility.

• Cut the foil to fit the slots in your 3D-printed model.

• Insert the contact springs or plates into the designated channels.

and it looks like


3. Securing the Hardware
Before moving on to the wiring, the contacts must be immobilized to prevent them from shifting during use.

Use a plastic welder (or a fine-tip soldering iron at low heat) to lightly melt the surrounding plastic over the edges of the copper.
This "stakes" the contacts permanently in place.



4. Precision Soldering & Circuitry
This is the most critical step for device compatibility and safety. You will need to install two specific components to ensure the device recognizes the battery correctly:

• Temperature Monitoring: Solder a 10k B3435 NTC Thermistor between the Negative terminal and the 3rd pin (the pin immediately next to the negative). This allows the device to monitor battery heat during charging.

• Capacity Identification: Solder a resistor (10k or 20k) between the Negative terminal and the 2nd pin.

Note: The resistor value typically acts as a "ID" for the device to determine the battery's capacity. If you're unsure, 10k is a common standard for these cells.



5. Final Assembly and Sealing
Once the wiring is tested and the connections are solid, you need to seal the internal cavity to protect the components from moisture and vibration.

• Apply hot glue or a non-conductive electronic grade silicone to fill the gaps.

• Ensure the sealant is flush so the battery fits perfectly into the device compartment.

and, it's done!

[solidpro]

Thanks for making the STL for the mounting points. I had them printed in nylon and fitted them into a 2141-182 which has the same design of hinges (well close enough) and lid....

I sanded down the original spaces, cleaned with alcohol and then Araldited the 3D printed bases in place, left it 24 hours and then sunk the brass knurl nuts inside.



In the next day or two, I'm going to try and do the hinges using other 3D prints you did (at the top of this thread) and then I'll see where we are at. This is great work!

[panips]

Restoration Update: 5 ThinkPads Fixed + PS/Note 182
I’ve just finished repairing five of my laptops, plus an additional PS/Note 182. I figured I should share some tips and experiences gained during the process for anyone else working on these vintage machines.

1. Repairing FPC Cables
The Flexible Printed Circuit (FPC) cables for keyboards and mice are notoriously fragile, but they are fixable.

The Fix: Carefully scratch the insulation away with a sharp hobby knife until the copper trace is exposed.
Soldering: Apply flux and a small amount of solder. You can then bridge the break using a tiny strand of copper wire.

Protection: Don’t forget to apply UV-curing solder mask to both sides of the repair to insulate and reinforce the joint.


2. Reinforcing Hinge Mounts (Brass Inserts)
You don't necessarily need a 3D-printed jig to accurately reposition the brass threaded inserts for the hinges.

The Method: Secure the brass insert to the hinge using its screw first,


Apply plastic welder (or epoxy) to the base. Then press the entire hinge assembly into its correct position so the insert seats into the plastic welder.


Curing: Leave it overnight. Once the welder has cured, the insert will be perfectly aligned. You can then loosen the screw, remove the hinge, and apply a bit more plastic welder around the insert to fully secure it.


Note: Some hinges (like those on the Type 182) have a flat "tooth" (circled in red below). Make sure to bend this properly to ensure it securely grips the 3D-printed replacement part.


3. LED Backlight Mod
I had an inverter board where the CCFL driver was dead, but the contrast circuit still worked. I decided to replace the CCFL with an LED strip from AliExpress.

• Voltage: These LEDs typically operate between 7.0V and 9.6V. For STN or DSTN screens, 7.6V–8.0V seems to be the "sweet spot."
  Driver Board: If you can't find a direct voltage match, use the included driver board. As others have noted, these usually require:

VCC: 15V
Enable: 5V (4.5V–15V range, eh, maybe 15v works as other members shared)
Adjust/Dimming: 0–5V
GND: Ground The LED strip comes with adhesive backing, making it very easy to install. The results on the Type 182 are actually quite good!



4. Battery Rebuilding
While rebuilding the battery packs, I added a custom "lid" for the pull-tab (the belt used to extract the battery).


It makes the pack much easier to remove and gives it a cleaner, more finished look.


5. The Collection: Five Working ThinkPad 350 Series
Here is the final lineup of the "survivors":

ThinkPad 350: Original Korean keyboard and OS.
ThinkPad 350C: Fully refurbished.
PS/Note 425: nothing special but badly damaged
ThinkPad 350: Standard US model.
ThinkPad 350C: Working, sort of, but has a temperamental keyboard (ESC and several other keys are unresponsive).




Some of these were in terrible condition when I started, but they’ve all been restored to a functional "daily driver" status.