Machinists: how hard to redo fragile case parts in metal?

[gsteemso]

Hi all,

It seems to me that (beyond the stuff common to all vintage machines like leaky electrolytic caps) there are three common points of failure in old PowerBooks -- batteries, 2.5" SCSI drives and shatter-prone case plastics. I've seen options, if occasionally pricey ones, to deal with the first two, but I've never seen anyone step up with a solution for the third.

I know it would not be straightforward to replicate _every_ replacement case part for each of the many, many different PowerBooks that were made, but you'd think some of the more common ones would be doable... at least the parts that almost everyone sees fail, like the stuff around the hinges, for example. I recall when my 2300c's screen frame plastics died, back in '99 or thereabouts, the repair guy said there were several PB models in his boneyard that could contribute the needed bezel... except the ones he had were all brittle and disintegrating as well. It seems somewhat incredible to me that in the 15 years this problem has been rendering otherwise good PowerBooks useless, no one has ever tried to fab a more durable replacement. (I've some experience with fabrication houses, and if I had the CAD skills to draw up proper designs, I'd run a kickstarter campaign to do exactly that, but I just don't have the expertise to follow through.)

My sister is a certified machinist, and what little I picked up during her training tells me that exactly reproducing a complex shape like that would probably be horrifically expensive, whether made directly of metal or by milling a mould and pouring new (stronger) plastic. That said, it seems to me that the really complex fins and stuff around things like binding posts would be largely superfluous if you made the replacement out of a more sturdy material, like metal or modern high-strength plastics... especially if you weren't too bothered about making it as light as possible, like they had to be back in the day. Honestly, at the weight those old beasts mostly were, adding another pound or three is hardly going to be noticeable unless you are very young, very old, or very ill.

Can anyone comment on the feasibility of this idea, or some modification thereof? I can't really ask my sister, as she lives in another country and has essentially zero knowledge of this specific problem domain (computer case design).

 

[Byrd]

It might be more relevant in the future, but when replacement parts machines are usually still available it makes it a lot cheaper, faster and easier using those as oppoised to fabricating new case parts fabricated from a 3D scanner/printer. Why metal?

 

[gsteemso]

Why metal?

Mainly just so I wouldn’t need to repeat the exercise ten or fifteen years later… heck, if I used the same crummy materials Apple did, new parts might well not last the first year if you needed to take your machine apart again for some other problem.

Speaking of, you probably COULD do a successful kickstarter to make more durable screen ribbon cables. That was the failure that ultimately made me give up the 2300c (still regretting that one).

 

[Trash80toHP_Mini]

I've been toying with the notion of making a wood/plastic model of a Compact ProtoHoaxMac™ for plaster casting female molds. Laying up fiberglass for the front bezel and bucket would be the method of production.

It might be interesting to try it with small parts and PowerBook cases.

 

[James1095]

It wouldn't be too hard to replicate certain parts, like hinges and whatnot, but trying to make something like a display bezel or molded case part would be a nightmare. Most machining done these days is subtractive, as in you start with a billet of material big enough to contain the entire part, then you selectively mill away material leaving your part as the result. You also have to consider how you will hold down the material securely enough to mill it while still being able to get a cutter into all the places it needs to get without cutting into your fixture or the attachment points. Many times you need multiple fixtures to mill a complex part in multiple separate operations, and one mistake along the way means you toss it in the scrap bin and start over. Figure several hours to draw up the part in the CAD/CAM software, cycle time anywhere from 10 minutes to over an hour, all of that at a typical shop rate of $85/hr. Then on top of that there's the cost of material, which can be pretty significant if you need to start with a big slab to mill out something like a bezel.

Bezels and plastic case parts is the sort of thing a 3D printer would be well suited for. There is still a very considerable amount of effort in getting the part drawn up, so unless you know someone who does that as a hobby you will spend far more than the current value of these machines.

 

[gsteemso]

Perhaps jumping straight to “machining” was premature. Stamping someting out of sheet aluminum & welding standoffs etc. to it would be a lot cheaper and probably lighter & stronger, too. Machining would look a lot more finished, of course, but honestly I just want the farkin’ thing to stop attempting suicide when opened.

 

[Trash80toHP_Mini]

Let me make a wild guess: you've got a 1400 with stress cracks near the hinges? Is it another PB? If so, which one?

Stamping aluminum isn't much less complicated or expensive and welding aluminum is no joke either.

 

[haplain]

I can whip up whatever you'd need for fun. Got a full machine shop with 2x 3-axis CNC's, 2x vertical mills, 2x lathes. Between those combinations of machines, if you cannot make what you want, you don't know how to use the machines

 

[gsteemso]

Trash80toHP_mini: the only 68K PowerBook I currently own is a 180c with almost no intact case plastics remaining. I’ve not put much effort into fixing it because all the power supplies I’ve found turned out to provide such dirty power that the electronics (CPU board, RAM, etc.) are almost certainly ruined. Danged unrepairable power bricks… I just don’t have the tools to open one up without completely destroying it. Did I mention it came without a hard drive, and whatever incompetent baboon performed the HDectomy lost one of the hinge screws so the other one tore out the first time it was opened, probably before I even bought it?

Byrd: I am envious that you can say parts PBs are plentiful, as they are anything but in these parts. The usual suspects for vintage gear discarded theirs years ago — it costs several times the $20 one will sell for to get one into working order, so they are viewed as a liability. Even scrappers don’t get much metal out of one.

Haplain: I am grateful for your generous offer, but the logistics are daunting. If I mailed you the wad of shattered plastic, would that provide enough data for a reconstruction? How much would you need for materials?

 

[naryasece]

I've been toying with the idea for the last couple of years of using the old case part of my collection to create moulds in which to create new case parts from.

My thought was to go beyond duplicating the parts and explore the possibilities of using metal frames when casting new parts to better disperse the force of opening and closing screens or high stress points (like the Duo Mini dock's spoiler release handle thingy). My plan was to look into it when I had free time and post to 68kmla if I could start making parts.

Using a 3d printer is an intriguing idea. I had some run-ins with a startup company looking to deploy 3d printers into malls so people could print out custom iPhone cases, jewelry, etc at my last job. If design specs for case parts could be open sourced, then replicating broken case parts might be simple as printing it out at the mall or getting really chummy with a 3d printer owner.

3d case scan might make replicating parts faster? http://www.makerbot.com/blog/category/scanner/3d-scanning/

I'll see if I can connect with the 3d printer people and get their input on the idea and I will finally look into DIY injection molding for some simple case parts, like the PB1xx port cover. I'll post if anything works out.

 

[techknight]

makerbots are nice, but they are damn hard to calibrate, and once running, to keep right... eh... your mileage may very.

youll need someone good with CAD to copy the case parts. The damn chinese do it all the time. haha. The other problem with the makerbots are the "print lines" or scanlines if you prefer.

How to hide those? overprint and sand maybe?

 

[James1095]

Stamping has the same issues as molding, the setup costs are obscene since you have to machine the molds/dies which have to be made out of a durable material. Once you have this done, the parts are cheap per piece, but you usually have to make thousands of the part before it becomes economical.