Plastic molding or 3D printing question

[csapp05]

Hello, I’m hoping I can tap into the collective minds to help get a project started!

1976 Porsche 912e throttle cable - NLA - I finally located probably the last NOS one, in Budapest!?!

I need to make a small approx 1” round x 4.75” long plastic “funnel” piece for a throttle cable. The funnel as far as I can tell makes it unique to the car, and the cable has a 923 part number, ie not shared with 911 or 914. The rest is pretty standard throttle cable (eyelet, threaded connector and solid rod end)

Pic of piece:



The actual function is to allow a stepped cable connector room to attach to a eyelet connector. The funnel is press fit into the cars bulkhead.

The left one is a NOS cable, right is my original. I’ve struck out locating a cable manufacturer to reproduce, so I’m thinking getting the funnels made and buying bulk cable and connectors and assembling them in house (garage!). I don’t have a huge demand (low production car, my cable lasted 40+ yrs, etc) but think a run of 50 funnels/cables to start. So this is a labor of love for helping keep the cars on the road.

So, obviously 3d printing seems to be a logical choice, but initial quotes seem like if I wanted 1-2 made then it works out, ie 3d printing takes too long. Injection molding is probably how it was made in the first place but I’m not sure how much a mold would cost? Finally I’ve searched for an off the shelf component to either modify or use. Maybe machining a plastic rod? I’d like for it to at least appear 1976 factory made, ie not having aluminum piece made.

Any ideas? Thoughts?

Thanks!

Additional pics:

[island911]

A 3D printed version ... is that part loaded in compression only? --most 3D printing is 'grainy' (anisotropic) like wood.

Because of that o-ring groove an injection mold I would expect to be 4-5k, in aluminum; which would be good for a couple thousand parts.

[csapp05]

It might have a slight compression but really it’s main purpose is to shield and allow free movement of the cable. The o-ring seems to be an overkill sealing element, the tube is snug in the bulkhead with out.
I was afraid of molding costs, might have to reconsider duplicating the funnel and fabricating a metal one.

Thanks Island!

Originally Posted by island911

A 3D printed version ... is that part loaded in compression only? --most 3D printing is 'grainy' (anisotropic) like wood.

Because of that o-ring groove an injection mold I would expect to be 4-5k, in aluminum; which would be good for a couple thousand parts.

[lendaddy]

Appears that it could be machined on a lathe from nylon.

[aschen]

Looks like an easy part to machine. Seems like you could make it out of aluminum for extra durability as well. I usually source parts in the 5-10 range and would expect to pay 150-200 for that type of part in those low quantities. Probably in the 1/2 range for making 50 of them.

I think 3d printing would be fine and easy as well. I recently bought a $200 machine (creality ender 3 pro) and it makes surprisingly decent parts. Even being the lowest grade chinesium printer I think it could make suitable parts.

I think it is a pretty ideal part and quantity for 3 printing. You could fit 10 or so per print on a cheap machine and have them knocked out in a few evenings. Pretty cheap to sublet out the printing to shapeways or similar.

[island911]

Do you have a CAD model of this? Or even a dimensioned sketch?

The shape certainly is easy to 3D print. And if you can lose the o-ring grove the (injection)mold becomes a simple single-pull tool. Likely under 2k at a quick-turn Inj. mold proto house. Or, for an even cheaper mold, the part could be cast.

[Zeke]

Originally Posted by island911

A 3D printed version ... is that part loaded in compression only? --most 3D printing is 'grainy' (anisotropic) like wood.

Because of that o-ring groove an injection mold I would expect to be 4-5k, in aluminum; which would be good for a couple thousand parts.

Originally Posted by island911

Do you have a CAD model of this? Or even a dimensioned sketch?

The shape certainly is easy to 3D print. And if you can lose the o-ring grove the (injection)mold becomes a simple single-pull tool. Likely under 2k at a quick-turn Inj. mold proto house. Or, for an even cheaper mold, the part could be cast.

I think you could look at the design again and see that the flange requires the same 3-piece mold that just the o-ring would require.
For 50 parts initially, machining makes more sense. AL or plastic. I know plastic is more given to be original but AL actually makes sense.

OP, when you find someone to make these for you, ask me for my billet taco plate design that incorporates an oil temp sending unit. Only one was produced and Jake Raby installed it on his wife's car. I have the CAD file.

[93nav]

I am just starting to learn CAD (Solidworks) and 3D printing (Ender 3 also), and that does not look like a difficult part to model or make. The question would be what material to use that will hold up to any heat, grime, vibration, etc. I currently am making stuff in PLA, and it seems kind of brittle with very little flexibility.

Looking at it, how is the 'funnel' attached to the black sleeve? Is there a nut or something where they attach?

[Captain Ahab Jr]

NOS part looks like blow moulded plastic

3D printing could do it, but you may want to explore the original manufacturing route

[island911]

Good on you 93nav. And yes, ABS and/or PETG would be much stronger parts. I like printing the PETG as it's not so stinky.

Originally Posted by Zeke

I think you could look at the design again and see that the flange requires the same 3-piece mold that just the o-ring would require...

nah. The core would wrap up to the flange, parting line on flange.

[csapp05]

I could do a CAD blueprint in a minute (drafting & design degree) and it is as simple as it looks, it’s a combination spacer & clearance apparatus.

The funnel end has approx 1/4” of cable crimped in it to hold it.

You can buy nylon solid rod for pretty cheap, might see about buying and doing a trial run of them.

Thanks Zeke, Raby was going to repop the cable but never followed thru!

[gtc]

I bet you could have it printed in laser sintered nylon for under $10 per piece. I use shapeways.com for printing some of my models, but there are probably local rapid prototyping shops that could do both the modeling and printing for you.

[island911]

Good call Graham!

Laser sintered nylon is amazingly affordable, and tough. Look for Multi Jet Fusion (MJF) process. It's similar to SLS (Selective Laser Sintering) but puts down black ink to aid heating of the nylon. Anyway, this mean you need to be ok with black parts. ..they have a grey, but that's only the surface.

I've been happy with Jawztec for this.

[csapp05]

All great suggestions and I’m very thankful for them! Gives me a new list to work on!

[unclebilly]

Originally Posted by lendaddy

Appears that it could be machined on a lathe from nylon.

I would use PEEK instead of nylon but that’s exactly what I would do for a onesy twosy...

[WolfeMacleod]

Originally Posted by csapp05

So, obviously 3d printing seems to be a logical choice, but initial quotes seem like if I wanted 1-2 made then it works out, ie 3d printing takes too long. Injection molding is probably how it was made in the first place but I’m not sure how much a mold would cost?

Mold costs vary greatly. I own an injection mold, cost me near $17,000 to have made in California. But it's much more complicated than a mold for that would be.

This company could probably do it for about $5,000, based on what I know about the costs my competitors made paid for mold similar to mine.

Here's mine.

[drew1]

I would look at products from Smooth On, I think the company is. This is a company that makes plastics that are 2 part like epoxy. I just gave some left over urethane that I had from probably 15 years ago to a buddy.

Every thing is done at room temperature. A part could be made from aluminum and used to make the mold. Plaster of paris is good for the mold, which could be made in several parts.

Spray the aluminum real good with silicon, and pour some mixed plaster around it. Before the plaster of paris dries completely make some dimples, for alignment. After it dries, respray and pour another part of the mold. A simple 3 part mold would probably do. With a file, openings for pouring the plastic and air release could be easily cut in the mold.

Plaster of paris isn't the most durable stuff, but is easily changed in the prototype stage. Several molds may get you the 50 castings. Except for the original part, $100 is invested. But then like 3D printing, there is the time. From memory it is several hours before a part can be removed and handled.

[rfuerst911sc]

What is/was the issue with the cable that is making you replace it ? If it's that white plastic piece degrading or breaking then I would change to a different material . I'm going to assume the plastic gets brittle from all the heat cycles under the hood . I would think a decent machinist could duplicate that out of aluminum or steel and it would be more durable . Good luck with the project .

[csapp05]

Originally Posted by rfuerst911sc

What is/was the issue with the cable that is making you replace it ? If it's that white plastic piece degrading or breaking then I would change to a different material . I'm going to assume the plastic gets brittle from all the heat cycles under the hood . I would think a decent machinist could duplicate that out of aluminum or steel and it would be more durable . Good luck with the project .

The funnel almost never breaks, it’s not close to any heat or light, semi protected by the transmission mount. The cable itself breaks, and Porsche hasn’t had one available in over a decade. It took me 5+ years to locate a NOS, and it was in Budapest... Seriously!
Porsche made approx 2100 912e’s in 76, approx 1300 are registered on the 912e registry. The cables (and cars!) are 40+ years old and over the years enthusiasts have cobbled 914, vw bus, bicycle etc cables to keep the cars on the road.
I’m hoping to make an Porsche OEM equivalent that is a direct replacement and maintain factory design.

Thanks for responding!

[piscator]

Another possibility is to have a machinist make a male mold in nylon or other plastic and lay-up the part in fiberglass (or, if you like sexy, carbon fiber, or kevlar). If I had to do this completely DIY, I might use the part to make a female two-part mold in fiberglass. Then use the mold to make duplicates in glass. In both cases, you're making a mold for a miniature boat hull and using that to lay up the glass.

I don't know if this is the best process for your application, but it does work for making one-off parts that are difficult to duplicate. Currently, I'm using this process to replicate the plastic hose barb on my headlight washer pump. The shape is vaguely similar to your part. I'm impregnating long fiberglass strands from some woven roving cloth and winding them around a length of polyethelene tubing, which is serving as the male mold. With this method you have to secure the ends of your woven roving with tape or weights to make sure it doesn't 'unwind' off the tube. Using epoxy thickeners or a compound like Marine Tech to impregnate the glass helps mitigate the inherent 'memory' of the fiberglass strands,

This link is to a company that does this kind of work, mostly on contract to the Navy. They also sell materials. I don't know if they'd undertake a small project, but the owner is a good guy and I do know he'd be happy to offer advice.

EDIT: The more I think about this, the more it seems that knocking out fifty of these is probably quite do-able. Maybe even better is that you could lay-up five at a time. Or whatever you might need to fill an order

For that purpose, I'd make both the inner 'male mould' and the outer 'female mould. I'd also integrate pins into the ends of one or the other mould to regulate the thickness of the lamination. I probably wouldn't try to mould the ring that protrudes from your part. I'd attach that separately. Your inner and outer moulds are, therefore, torpedo shape. That makes it relatively easy to lay-up the glass on the machinist made male mould, place the two part female mould around the male mould -- and then vacuum bag the whole affair to get even clamping pressure and a perfect finish.

If you want to get fancier, you can incorporate drains for resin overflow during the vacuum bagging. And there may be some way to integrate the ring-shaped protrusion. The guys at LBI would be better able to assess that.

Good Luck!

https://www.lbifiberglass.com/shop/