Mouse Wheel 'Shaft' Repair Question

boddole

Distinguished
Mar 25, 2012
50
1
18,535
Hello everyone, based on the results of this thread ( Previous Thread ), I've decided to make a more technical post about a problem I'm having with a broken mouse wheel shaft, what I've tried to repair it, and a new ideas about how to fix it. Hopefully a few of you have tried something like this before and have some insight, or just have some general feedback on my ideas to make the fix.

A Few Details:
*This is by far my favorite mouse ever.
*Mouse Model: Cooler Master Storm Recon [it has officially been discontinued]
*Unlike most wheels I've seen, this wheel is only 'flat' on one side (the 'click switch' side). The rotary switch side is actually concave because they use a circuit board extension with the small type LED light to get said light into the wheel. The design is quite clean, but it does make 'working' inside the wheel considerably difficult (and tolerances quite small).
*I personally don't exactly have a lot of tools/machinery available to me aside from scissors and hammers, so all my ideas are either what I could do personally or have 'exported' for some business to complete. That said, if you have some good idea that uses a reasonably easy to get tool, I'm all ears.

The Problem:
The shaft (on the rotary wheel side) cracked completely though, rendering the wheel unusable. See the damage here:
10xu595.jpg


The fix that didn't really work out:
Basically I...
-tightly wrapped a half inch or so of paper around the broken shaft to use as a guide
-tried to place it flatly against the 'stub' of plastic inside the wheel (hard to tell by 'feel') since there isn't much there (the paper is left on since it is quite thin and sticks to the glue)
-take a bit of two part Gorilla glue and apply it to the guide shaft
-press the broken shaft down into the guide (taking care to align the 'jagged' bottoms as best as possible to minimize any length difference after the glue was added
-manually 'eyeball' whether or not the shaft was straight or not by looking at it and adjusting it with my finger, rotating it 90 degrees and doing the same
-let glue set for 24 hours

Result-> while this did appear to work, when the entire mouse shell was put back on the slight amount of wobble caused the wheel to grind against the sides of the mouse shell at certain positions, requiring a very hard 'click'. This hard click eventually caused the glue to give and I'm back where I started)

New Ideas:
1) Try the same 'fix' again
Pros: easy to do by myself, maybe I'll 'be better' at it this time?...
Cons: tolerance probably still won't be tight enough.
-probably my least favorite option.

2) Make a 3D model of the entire 'shaft part' and 3D print it using a clear-ish plastic or resin (and 'scuff it up' if the light is too bright).
Pros: should be accurate enough to fit more/less perfectly (not entirely sure what the tolerance on the 'rotary switch side' would need to be since it may require 6-16 slots to interface properly).
Cons: I'm really not sure how to remove the shaft 'nub' and other side of this thing...I'm not sure if it is pressure fit or glued into the larger wheel (or is actually one single piece and includes the wheel itself), but I don't think bashing it with a hammer would work.
-a great option -if- I could figure out how to remove the current shaft (otherwise I would probably need to make the entire wheel (not impossible, just more work).
2exyb80.jpg


3) Instead of using a floppy paper 'guide', make the guide out of 3D printed material.
Pros: simple to make.
Cons: even if it works, the glue may not be strong enough to hold. Not quite sure how to use the guide without it becoming stuck to the glue (or how to remove it without 'tilting' the shaft)...
-perhaps the most practical.

If anyone has tried something like this before / or just has some feedback on my ideas (perhaps you've done something generally related to them before) would be greatly appreciated.
 
glue is always going to be brittle and break especially on small surface areas exposed to high pressure. if the piece was larger diameter and not a stress point it may have held up. also, certain types of plastics may not bond as well as others.

a better idea would be to either replace the whole shaft or to pin the connection with a small diameter metal rod. its hard to tell from those photos, but the shaft should have some groves or flats for use with the mouse wheel sensor (at least on most mice i've seen) which makes straight up replacement not so fun. if it didnt have any such alignment indicators . pinning may be a bit tricky to get straight but will be stronger and resist shearing forces. before gluing the pins in make sure to rough it up a bit and dont make the holes too loose (it should be a snug press fit).

the best fix would of course be to find a replacement wheel or parts mouse but i'm assuming you want something cheap.
 

boddole

Distinguished
Mar 25, 2012
50
1
18,535


First off, thank you very much for the response and the ideas. Based on your feedback, here is what I'm thinking:

"replace the whole shaft"
-after looking quite closely at the part, it certainly does appear that the 'shaft' is actually one solid piece of plastic which includes the entire wheel body, so aside from trying to 'drill it out' just making a new wheel would probably be easier (going the 3D printing route).

"find a replacement wheel or parts mouse"
-Well, the real problem is because it was discontinued and somewhat obscure (unfortunately). I've only found two on the web: one listed from some retailer for $130 (cost me ~$35 new), and one in Quebec with "in-store pickup only" and I live in Los Angeles so I don't see that one happening...

"pinning"
-If possible, could you elaborate on 'pinning' a bit more (I'm not quite sure what you are getting at / just don't understand). The end of the shaft head is hexagonal (looked at it under magnification), if that is important to your idea.

 
shaft
what i was worried about is how the mouse wheel switches (the switch which handles rotational motion) often is hexagonal or splined like http://cdn.instructables.com/FO4/WZEC/FNI410XJ/FO4WZECFNI410XJ.MEDIUM.jpg and replacing the shaft would require this to be the same as well. not sure how your mouse looks though. 3d printing is only so accurate for small objects (based on tip size of the printer) so you may not get exactly what you want without some manual filing. it might work for getting the general shape though.

pinning
basically the same concept of doweling http://www.craftsmanspace.com/sites/default/files/free-knowledge-articles/various_dowel_joints.jpg where a hole is drilled on each side and a pin with glue used to hold the joint together better than glue alone. a rough pin will hold better than a smooth one. this would prevent the joint from snapping like a pencil when you push down more than just a glue joint

that hexagonal head is exactly the problem with replacing the shaft. to get it to work you need that portion to work with the switch. it could be recreated by hand but its hard to get it perfectly centered and accurate. it can be machined but you have no tools for that likely. something roughly correct may certainly work well enough but if it slips around in the hole you will feel it.

 

boddole

Distinguished
Mar 25, 2012
50
1
18,535
I see, thank you for the clarification (especially on Pinning).

As far as making a new shaft/wheel, it seems that most 3D printing companies list their 'minimum detail / tolerance' (for plastics anyway) at around 0.2-0.5 mm, which I think should be good enough. As long as I can get the 'max extents' on the hex shape I should be able to work out all the distances mathematically from there.

As far as pinning goes, what 'machine / tool' would you ideally use for this? I don't imagine trying to drill through a very narrow plastic tube is an easy job (especially for someone that hasn't done it before).
 
honestly i havent had to pin anything that small so i cant say for sure how it will work out. normally i use a drill press and vise. i suppose a dremel can be used with a vise if you're very careful. the most important thing would be a straight hole in both in the same location which isnt always easy. much of what i've had to pin was more in the size range of quarter inch or bigger materials.

a resolution of .2 or .5 mm should be good enough. honestly better than i thought for 3d printing. the trouble is getting an accurate measurement on your plans. also i'd be wary of cost.. 3d printing can be expensive in some cases. if you decide to go this route i'd make sure to not use a ruler as its not accurate enough but instead use a micrometer and/or inside/outside dial calipers to measure distances for accuracy. keep in mind to leave a little extra meat for some light sanding where the shaft sits and rotates on the one sideif the texture is going to be a bit rough (most 3d printing i've seen has a semi rough finish).

quite frankly if it was me i'd just buy a new mouse but this may be a good learning experience.
 

boddole

Distinguished
Mar 25, 2012
50
1
18,535
I admit, the fun of 'making something' is probably half of the motivation. That said, the list of mice that are: ambidextrous / programmable / at least two buttons on each side is fairly small, and usually no less that ~$60 (mine being the exception). I made a quick 3D mock-up of a 'guide form' that I think might be the best way to go about this:
*the green part is the 'nub' left from the break
*the 'wings' are just for grip (twisting and pulling to remove is needed)

http://i66.tinypic.com/2vaia1j.jpg

http://i65.tinypic.com/tzejo.jpg

http://i65.tinypic.com/i1jxiq.jpg

http://tinypic.com/m/jihvm9/1
 

boddole

Distinguished
Mar 25, 2012
50
1
18,535
As long as the guide shaft is tall enough, the glue shouldn't end up touching anything besides the 'nub' and the base of the wheel. I could even line the inside of the guide with paper (or whatever) as a buffer between the glue and the material just to be cautious.

As far as strength goes, even with Gorilla glue from last time, it only failed after applying quite a lot of pressure when the wheel was grinding against the side of the mouse casing, I don't anticipate any trouble from 'regular clicks' (could always try a stronger glue as well now that I have a guide form).