Help repairing a hand-fitted plastic injection mold?

The hell where I work (for now) has really thrown me a curve this time. They had a very, very cheap 16 cavity mold built in China. Thanks to the knuckle-draggers who set it up here, they managed to damage all 16 cavities pretty badly. They had the damage welded up, and handed it off to me.

Here's the problem- there are no precision surfaces to work with. Everything was hand fitted with a hand grinder and stones, I assume. The damaged parts are side actions that telescope into the main cavity area. Surfaces that are supposed to be flat aren't. They're wavy and shut-off angles are all over the map. Each piece is going to have to be fitted to it's mating cavity. I honestly don't know where to start. I've never seen such a crudely machined mold in my 40+ years of experience. I don't have the skills to fit it by hand. I kinda have to admire the guy who fit it- it actually made a good-looking part until our guys got a hold of it. The only thing I can think of is to get as close as I dare, then EDM it fit, using the side action as an electrode, burning both parts until it fits.

It gets better- The mold has a mirror polish! Polishing welded and re-machined areas will look like crap; the weld will telegraph through.

Any advice? i seriously considered just flat refusing to work on it, but that would cost me my job.

Are the only damaged portions the slides? Or are the moulding surfaces of the cavities damaged as well? If the latter, did the operators at your work close the press with the mould mis-aligned? I'm having a difficult time seeing how they managed to damage the mould that thoroughly.

If you use the slides as electrodes, it's doubtful that they will be aligned with their current path after.

I'm not sure how the damage was done. The part is circular, with the slides forming part of the circle. They damaged the sharp corners of the slides only, plus some other areas, probably from digging parts out. The mold is supposedly 420 SS, but it is very soft. The parts cut pretty easy with a file. The weld, however, is hard, so I can't just blend it in with a file. That's part of the problem, with the weld being hard, and the parent metal being soft. That's part of why the polishing will be a nightmare.

Anneal the parts so you can hand file them?

I found out how they crashed it. It is stripper ejection, and the stripper plate was forward when they tried to close the mold. Yes, there's a proximity switch safety, but it wasn't attached to anything. Brand new mold.

I got most of the shut-off surfaces close to size today on the surface grinder, then started EDM on the cavity area. Once I get the shut-off's dialed in, I'll just polish the heck out of it with the slides in place in the cavity block. They get what they get.

What kind of part is it ?

And what kind of plastic ?

MushCreek said:

I found out how they crashed it. It is stripper ejection, and the stripper plate was forward when they tried to close the mold. Yes, there's a proximity switch safety, but it wasn't attached to anything. Brand new mold.

I got most of the shut-off surfaces close to size today on the surface grinder, then started EDM on the cavity area. Once I get the shut-off's dialed in, I'll just polish the heck out of it with the slides in place in the cavity block. They get what they get.

Click to expand...

Incredible. So the moulding surface is relatively unscathed, other than the surfaces on the ejection slides? I'm assuming you've removed the slides and are die-sinking each one individually. Was going to suggest putting the whole base on a VMC and hard-milling. Been there done that. Obviously wouldn't have worked if "all" you're cleaning up is the movable slides.

Nice work; should work!

What a horror story

MushCreek said:

The hell where I work (for now) has really thrown me a curve this tiME

Any advice? i seriously considered just flat refusing to work on it, but that would cost me my job.

Click to expand...

Sorry for you ,I have been there.Of course non of the knukle dragers will be fired because NOBODY DID IT.These guys should be beat.Good luck Edwin Dirnbeck

MushCreek said:

I found out how they crashed it. It is stripper ejection, and the stripper plate was forward when they tried to close the mold. Yes, there's a proximity switch safety, but it wasn't attached to anything. Brand new mold.

I got most of the shut-off surfaces close to size today on the surface grinder, then started EDM on the cavity area. Once I get the shut-off's dialed in, I'll just polish the heck out of it with the slides in place in the cavity block. They get what they get.

Click to expand...

Unfortunately, you've succeeded in pulling their ass out of the fire.

Also, you've proven to be good at it, so I suspect more of this work
will fall on your shoulders.

Make sure you document EVERYTHING about the mold. The hardness, the welding, the hand fitting, the lack of the safety switch, EVERYTHING. Being a brand new mold and not having any reference surfaces is amazing. I have a feeling you will called in to explain such shitty workmanship.

Tom

Hi MushCreek:
Tom's got it right on this one; probably THE most important thing you can do is document the hell out of it, because it just smells like a blame game growing.
If you can't restore it to perfect in under an hour for 50 bucks; someone in the executive suite is going to get their panties dirty.

It'll all be your fault of course if you can't blend it back in perfectly, so find yourself a hardness tester and hit the welds in a couple of spots, then hit the inserts and take little videos with your smartphone.
You're going to catch some shit anyway, so take whatever time it takes and just bank the hours with a smile.

Fun Fun Fun!!!

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining

Can you add reference points after done then use them to document and dimension the unit?

Can it be scanned by something to document size or original drawings?

Sent from my SAMSUNG-SGH-I337Z using Tapatalk

I have a 'print' to go by; it's all of the hand work that makes it tricky. The easy fix would be to requalify the shut-offs that the slides fit into to be consistent and accurate, then make a new set of slides. They're not going to do all that, of course. They only paid 10 grand for the mold. The part is a sprayer for a can of bug spray. Not sure of the material- maybe LDPE? The worst part of the story is that they've ordered 5 more molds from the same source before seeing how the first one works out. These molds are run hard- 24/7, 10-15 second cycles. Millions of shots a year. The moving parts on this one are already starting to gall, and hasn't really had a good production run yet.

MushCreek said:

The moving parts on this one are already starting to gall, and hasn't really had a good production run yet.

Click to expand...

Any grease grooves on the working surfaces?

Some, but not on the slide inserts that I'm working on. It's really something- the telescoping shut-off is a compound angle, with 3 degrees one way, and 5 the other. They EDM'd the pockets reasonably well, but rather then grind the mating part, it looks like they completely hand worked the compound angles. Needless to say, most of it is hacked away for clearance, with a small land that actually shuts off. Why the heck not just set it up and grind it, like we used to do back in the 20th century?

MushCreek said:

Some, but not on the slide inserts that I'm working on.

Click to expand...

We have had luck keeping the Chinese molds from galling by hand grinding overlapping diamond shaped grooves into moving parts. I don't think separate grease grooves work as good as ones that connect and allow grease to flow from one area to another.
It would be nice if they don't run the mold till it breaks, instead allowing you to see the shiny spots that are hitting too hard, and grinding them down until the press tonnage is spread out over enough area to prevent excess coining.

Hi again MushCreek:
Seems to me you have two options here:
First is to sort out all that is substandard on this mold, document it and then knock up a quick plan and cost estimate to restore it to decent condition.
Quote order of magnitude numbers only and don't be shy.

Second is to bodge it back to a state where it will run and then throw up your hands.

If you choose option two; it is important that you first pointed out option one to management; otherwise when the mold fails it's gonna be your fault whether it truly is or not.

A note on hand fitting from my own experience:
I got myself a laser welder 10 years ago, and the reason it's relevant is that it made me MUCH bolder than I used to be about hand fitting because a little over-enthusiasm with the pencil grinder didn't have the same consequences anymore...I now have a great "Putting-Back-On" tool.
What it taught me was that I had been far too hesitant bluing things in and I took a lot longer than I needed to; pansying about so as not to take off too much.
You may find, if you're not accustomed to hand fitting, that it's all taking forever, and almost always, that's the reason.

Modern toolbreakers tend to work to the numbers and rely on the precision of the toys to get them there, but us old farts grew up with a Bridgeport as the most accurate machine in the shop and could still build good molds too, (just not with the same techniques), so we tended to get a lot more comfortable with milling it close then hand working it in than most guys do these days.

Also if you do a lot of mold repair on old beater pieces of shit molds as I've done, you get good at seeing just what you can get away with and still get it to work, at least for a short time.
CYA tactics like pre-attack photographs are an important part of this skill set!!

From what you've described to us, this is never going to be a Cadillac; it's a new "beater piece of shit mold", so I would embrace it accordingly if you can find it in your nature to do so.
Don't make much of an effort to improve it unless you have a specific mandate to do so from your management.
Whatever rude and dirty you can get away with...don't be shy.

On a last note; your laser welder buddy is your friend...make use of him...frequently if necessary.
I've had great success welding up boogered parting surfaces, dressing them back aggressively to where I think they should be and then welding them up again in spots and dabs where my guess was not perfect.
It's WAY WAY faster than meticulously stoning off tenths at a time and blueing with great care to get it perfect on the first go.

This all sounds like the work of a complete hack, and it is, but it's so much less painful than doing it the proper toolmaker way.
A POS like this sounds to be, is not worthy of the stress of trying to do it perfectly, so BE BOLD AND DON'T SWEAT IT.

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining

Can you post any pics? I have nothing to add to the gent's posts above but I'm curious to see the issues.

implmex said:

CYA tactics like pre-attack photographs

Click to expand...

^^^What a perfect phrase

Adding to Marcus's point, put the photos in a PowerPoint presentation. Use it like a slide show so you have one picture and one or two points on each slide, and then make lots of slides. Management loves PowerPoint. Or even save it as a PDF.

"Dear XXXX, please find attached a summary of issues and errors in the mold and the steps I undertook to repair them."