[sansbury]

This is one of the trickier parts someone has asked me to make, and the trickiest I've said "I'll try!" to. It's a screw for a (I assume) plastic extrusion or injection process, made from 4140. .500" x 12" long makes for a not-at-all rigid setup, and it took a few experiments to figure out what would work. I started with a piece of 6061, just to test the code, which was written by hand as I haven't done enough 4th axis work to justify the cost of upgrading my CAM. The aluminum test piece went well, so I switched to the 4140 and the chatter pretty quickly got fatal, so I knew I would need some serious support.

In the end, I took a piece of aluminum and milled out a .500" diameter channel that would sit directly in line with the workpiece. That, plus a little judicious use of a manually-applied semi-rigid vibration damper (aka rubber mallet) kept things in the happy. I used a .375" Maritool 4-flute carbide EM with .015" radius, running around 1800RPM and 2.46IPM at full-width passes .02" deep. The speed was limited by the rotation speed of my rotary table, and DOC was limited by rigidity concerns. This was a 1-piece job, so once I had a formula that seemed to be working, I decided to stick with it. Machining time was about an hour, setup and figure-it-out time was a lot more than that. Customer was happy with the photos, so it's a good start, but the proof will be in the (plastic) pudding it hopefully makes.

[AUSTINMACHINING]

Very cool ! Nice work.

[WOTDesigns]

That's impressive any day of the week. Nice job.

[stargeezer]

That's some awesome coding! I'm a "student" (and not a good one) blacksmith, as such I might have used flatstock, heated it to working range, and twisted it to get the number of turns and length. Follow that by welding on the machined arbor ends and follow by heat treating for hardness. Isn't "making" things a wonderful pastime. We get to explore our limits of creativity, learn new skills (FYI I'm positively horrible at coding) and in the end have some jerk say, "you know, you could have done such and such", BUT, it's all good. And this is a CNC forum after all.

So again I say wonderful job and cudos for the coding efforts. I am greatly humbled by your skill..

[CNC-Dude]

Man, you had me at "This is one of the trickiest..."! I am blown away by this project, not to mention the fact you programmed this by hand. Not that I need such a screw but now that I see it has been done, at least I know it can be done in the event that I ever feel like making one. But to be honest, I am still having a hard time believing it can be done!

Fantastic job! Thanks for sharing! Definitely helps in bringing incredulous fools like me into appreciating the capabilities of this equipment.

[CadRhino]

Very impressive. Thanks for posting.
Just out of curiosity, any idea how much you would charge a client for such a complex part? I am always trying to figure that part out.
Thanks
Nathan

This is one of the trickier parts someone has asked me to make, and the trickiest I've said "I'll try!" to. It's a screw for a (I assume) plastic extrusion or injection process, made from 4140. .500" x 12" long makes for a not-at-all rigid setup, and it took a few experiments to figure out what would work. I started with a piece of 6061, just to test the code, which was written by hand as I haven't done enough 4th axis work to justify the cost of upgrading my CAM. The aluminum test piece went well, so I switched to the 4140 and the chatter pretty quickly got fatal, so I knew I would need some serious support.

In the end, I took a piece of aluminum and milled out a .500" diameter channel that would sit directly in line with the workpiece. That, plus a little judicious use of a manually-applied semi-rigid vibration damper (aka rubber mallet) kept things in the happy. I used a .375" Maritool 4-flute carbide EM with .015" radius, running around 1800RPM and 2.46IPM at full-width passes .02" deep. The speed was limited by the rotation speed of my rotary table, and DOC was limited by rigidity concerns. This was a 1-piece job, so once I had a formula that seemed to be working, I decided to stick with it. Machining time was about an hour, setup and figure-it-out time was a lot more than that. Customer was happy with the photos, so it's a good start, but the proof will be in the (plastic) pudding it hopefully makes.

[sansbury]

I quoted $250 on the basis that I don't like charging customers for my lack of experience or equipment. The secret to this is that it's actually not a very complex part to program--just a coordinated AXZ movement. The real work is three lines of code, and once you see them, it's fairly simple. The workholding was, as usual, the real hard part and where I spent most of the time, and most unproductively. Trimming the fat I probably would have 4-5 hours in this, of which maybe 1-1/2 is machining time. I could probably do a repeat in an hour, two if I had to set up the 4th axis from scratch. I'd probably quote $400 for a new part on the same lines, or $100-$150 each for additional parts. Boston rates are $100+/hr for "real" shops which I don't consider myself to be just yet.

When the part is "educational" as this one was, I mostly try to make sure I'm not getting paid McDonald's wages. If the part is a simple one I know how to do, I might quote it cheap if it's a fast easy job I can throw on the machine and do, if it's going to be ten hours of real work, then I want to get ~$50/hr at least. I operate out of a co-working space so I pick up little odds and ends from other tenants somewhat regularly.

One thing I'm learning from doing more prototypes is that the not-so-clever-but-really-obvious way is often the best one. Spending ten dollars of material on a single-use fixture makes more sense than spending an hour on something "better," unless (a) you know it will work as well or better, and (b) you know you'll use it again. Ditto the programming. The slow-to-run-but-sure-to-work way will probably win the race to make one or three parts. If the customer comes back and wants ten or twenty more, then go back and optimize.

[CadRhino]

Great advice. Thanks for sharing

I quoted $250 on the basis that I don't like charging customers for my lack of experience or equipment. The secret to this is that it's actually not a very complex part to program--just a coordinated AXZ movement. The real work is three lines of code, and once you see them, it's fairly simple. The workholding was, as usual, the real hard part and where I spent most of the time, and most unproductively. Trimming the fat I probably would have 4-5 hours in this, of which maybe 1-1/2 is machining time. I could probably do a repeat in an hour, two if I had to set up the 4th axis from scratch. I'd probably quote $400 for a new part on the same lines, or $100-$150 each for additional parts. Boston rates are $100+/hr for "real" shops which I don't consider myself to be just yet.

When the part is "educational" as this one was, I mostly try to make sure I'm not getting paid McDonald's wages. If the part is a simple one I know how to do, I might quote it cheap if it's a fast easy job I can throw on the machine and do, if it's going to be ten hours of real work, then I want to get ~$50/hr at least. I operate out of a co-working space so I pick up little odds and ends from other tenants somewhat regularly.

One thing I'm learning from doing more prototypes is that the not-so-clever-but-really-obvious way is often the best one. Spending ten dollars of material on a single-use fixture makes more sense than spending an hour on something "better," unless (a) you know it will work as well or better, and (b) you know you'll use it again. Ditto the programming. The slow-to-run-but-sure-to-work way will probably win the race to make one or three parts. If the customer comes back and wants ten or twenty more, then go back and optimize.

[brian257]

The "price it low to get the job because it is a really cool part that I want to make" method of quoting. I do that a lot.

[pickled]

Wow excellent work! I hope that you got paid well for that job because the OEM versions of those types of screws are extremely pricey! Pretty impressive that you pulled that off so quickly. I made a few similar screws for a powder filler (cheese powder) 2 years ago and had nearly an identical set-up to yours to deal with the flex. It was a total fingers crossed type job for me because I really didn't know how well I could hold the tolerances in the unsupported areas.

[Concepts_Design]

Excellent Work!! by the looks of the size of it it's for a macro or even micro injection molding machine, 15 tons or less, and $250 is extremely cheap for that. Off hand I would say a stock screw would be about $1500-2500 and they are usually a long lead time item (3-4 weeks) so if you popped that out quickly for them, that's an added expediting fee. Definitely impressive work, next time don't sell yourself short, not many people would take that job on even for full price. And the ones that would take it would charge double what a stock one would cost as and expediting fee because you can run a molding machine without the screw, so time is money and molders cant afford to be down for days much less weeks.

Awesome job again! I really want to give this a try now since all my customers are injection molders, I apologize in advance if I start asking you a million questions

[LeeWay]

While that is some excellent work and good advice, I wonder about the machine the guy is making. I know some larger IM machines do use a similar screw. I think those are high volume production machines. Smaller machines tend to use a simple hydraulic or even a manual ram to push the plastic through the heat chamber. With such a small screw, he will have to have some very small plastic pellets. I wonder if it will be a working machine or just a scale model. I don't know a lot about them, but have researched them. Small bench top machines I mean. I have a 20 ton air hydraulic press that just sits here. I think I could put it to better use.

[Concepts_Design]

While that is some excellent work and good advice, I wonder about the machine the guy is making. I know some larger IM machines do use a similar screw. I think those are high volume production machines. Smaller machines tend to use a simple hydraulic or even a manual ram to push the plastic through the heat chamber. With such a small screw, he will have to have some very small plastic pellets. I wonder if it will be a working machine or just a scale model. I don't know a lot about them, but have researched them. Small bench top machines I mean. I have a 20 ton air hydraulic press that just sits here. I think I could put it to better use.

just about every injection molding machine out there has a reciprocating screw in it, in fact I don't know of any that don't as they are required to melt the resin pellets. they are not just heated, but are also run thru this screw as the shear heat breaks down the resin back into it's original molecular structure in order to correctly mold parts. What you might be thinking of is a "ram" type which actually has both a reciprocating screw and a hydraulic ram to make the final push into the mold. These are used for machines that require higher injection pressure , 60k PSI, such as what is needed in micro molding and thin walled molding.

[LeeWay]

Right. I understand that the screws do the heating due to pressure. I have not seen but a couple of large machines in person and they were not in use at the time. We were building the crates to ship them in.
I was talking mainly about small scale machines that this size screw might be for. A screw like this is the proper way to design such a machine for sure. There are small machines out there that are sub $2000.00 that just use a manual or hydraulic ram though and do not use a screw. Simply said those are hobby machines for very low volume molds. They heat exclusively in the heat chamber below the ram. The mold is clamped under that. They are gravity fed and are usually in a vertical orientation. The screw types are always horizontal and are continuous feed. High volume. No need to back out the ram to fill the heat chamber. I would just love to get a peek at this screw in place on the machine.

[WOTDesigns]

Right. I understand that the screws do the heating due to pressure. I have not seen but a couple of large machines in person and they were not in use at the time. We were building the crates to ship them in.
I was talking mainly about small scale machines that this size screw might be for. A screw like this is the proper way to design such a machine for sure. There are small machines out there that are sub $2000.00 that just use a manual or hydraulic ram though and do not use a screw. Simply said those are hobby machines for very low volume molds. They heat exclusively in the heat chamber below the ram. The mold is clamped under that. They are gravity fed and are usually in a vertical orientation. The screw types are always horizontal and are continuous feed. High volume. No need to back out the ram to fill the heat chamber. I would just love to get a peek at this screw in place on the machine.

Like the Tormach injection molding attachment For example...

http://www.tormach.com/store/index.php?app=ecom&ns=prodshow&ref=32079

[victorofga]

great post...
I knew best cam program is notepad for windows :-) true multiaxis

seriously, many folks could read your post as you determining price.. many shop loosing works due they stick rigidly what they spending hours.. by equipment or by missing knowledge..


just out of curiosity..
would be realistic to set a high rpm spindle on the Z and grind a temperd one of this screw?
and that can be polished I think better than an untemperd...

[LeeWay]

Well now, that is one I have not seen. Very cool. My first mill is a home made 80/20 mill. It has a pretty strong Z axis with ground ball screws and a 1200 oz nema 34 direct drive motor. I have not really used it as a press before, but do know it is capable of some pretty high pressure on a downward stroke. Do anyone of you guys have the Tormach version?
Here is a pretty good video explaining the screw design for an IM machine and the reason for it's shape. It's about 5 minutes long.
https://www.youtube.com/watch?v=eUthHS3MTdA

[CNC-Dude]

That video has tons of juicy stuff. Here is a paper with some of the equations and the distances in case anybody wants to create their own table top injector.

The Evolution of screw design technology for the Injection Molding Process - Part 1

Sounds like a cool project which I will add to my infinite list of projects which most likely will never get done ;-)

[LeeWay]

https://www.youtube.com/watch?v=YJVwfNNJwsc

[sansbury]

As far as I know, the customer in this case is an engineer working on her own time on a side project (I'll find out more when she comes to pick up the part). So for all I know, if the bids to make this were $1k+, then it might never have been made. Certainly not a rush item to keep a line running. The customers I make parts for tend to be inventors working on prototypes, and the shops around here typically don't really want that business anyway. But I couldn't live off this kind of work alone--I have the shop because I'm working on my own prototype, and these projects help offset some of that cost, teaches me new things, and helps to ensure the IRS sees this as a business and not just a hobby.

I did consider grinding--I thought about rigging up my Dremel on a spindle mount to do a finish pass after milling, but the finish from milling alone was decent. If you wanted a hardened screw, I'd think the way to do it would be to mill close to net shape in the annealed state, heat treat, and then grind to final dimensions. But what the heck do I know?