What's new
What's new

Recommendations: CNC machine for very precise plastic work

Alex_I

Plastic
Joined
Apr 12, 2019
Hi guys -

I'm looking for recommendations for a CNC machine that can be used to cut very small, very precise parts in plastic.

My typical parts would fit in a 50mm cube. They do have some tolerances down to +-0.05mm and walls < 1mm thick. The material is PC/ABS, ABS, PC/PET or similar. Geometrically, most of the features can be cut with a 1mm end mill, but some might require a smaller tool 0.5mm or even less.

Ideally I'd like to be able to do this both for easy prototyping of one-offs and for small scale production (10-100 parts/day). I think my designs are going to evolve quite a bit before I go to injection molding; 3D printed parts are trouble since they can't usually hit those tolerances and the mechanical properties just aren't the same.

I've done very little CNC milling work before (mostly whatever machines were at hackerspaces - wood or metal, no plastic) so I don't really know what I should be looking for here. A benchtop mill like a Taig or Sherline? Something like a CNC conversion of an X2? A small VMC eg Haas VF2? A mill specifically for micromachining eg Microlution, Datron etc?

What are the important specs? Which of these has the better positional accuracy / tolerance? What kind of tool and collet do I want? (and do I want an automatic tool changer?) etc.

Thanks for your advice! I hope this helps anyone else looking to make micro-machined parts out of soft materials.

Best regards,
Alex
 
Better start with you budget. If you serious about it you should look for a used brother speedio
 
Hi Aejgx6 - I should have mentioned the budget - the truth is I've no idea what is possible though.

If I don't mill these myself, I'd probably need to spend $5-$10k on getting a shop to make prototypes for me over the next year, so anything less than that is a no-brainer - assuming I can actually make good-enough parts, of course.

For production, $10/part is okay. If I can get (eg) 10 parts/day and the effective cost of leasing + operating the machine is significantly under $100/day, that's also pretty darn good.
 
Thanks! Researching the Speedio now - interesting!

Could you tell me why that's the best choice? It looks like a fast machine, but not specifically made for micromachining. What is the typical accuracy and how does it compare to the other options? Can it use 0.5 - 1mm tools?
 
Could you tell me why that's the best choice? It looks like a fast machine, but not specifically made for micromachining. What is the typical accuracy and how does it compare to the other options? Can it use 0.5 - 1mm tools?

Put simply; a Brother Speedio punches well above its price class in regards to machine accuracy, repeatability, speed, and reliability. This is the reason so many of us on here have them.

It is not a "micromachine," which is good because if it was, Brother would probably want to double the price. What it is though, is a *production* machine, built to bang out massive quantities of work with extremely high repeatability and accuracy. And we aren't talking sloppy +/- 0.005" American jobber accuracy... if you've ever seen a Toyota/Apple/GE blueprint, they ask for parts with 0.05mm tolerance features all over them. Increasingly, those guys are buying Brother Speedios to do it.

It is also about the least expensive way to get a crazy high-quality 27k spindle machine that is ideal for your kind of work. This isn't an air speeder, or Dremel bolt on bullshit - the 27k Speedio is probably (owing to Apple/Samsung aluminum chassis phone demand) the first or second most popular spindle they make, and they are built for production. Extremely robust, rated to run at full RPM all day long, and with a high service life.

You also get Yamazen support, which is just about the best in the industry. Andy ("2outof3" on here) is the sales manager for your area; ask him for a quote on an S500/27k. It'll whip any other option out there. Join the Speedio Mafia!
 
Hi guys -

Ideally I'd like to be able to do this both for easy prototyping of one-offs and for small scale production (10-100 parts/day). I think my designs are going to evolve quite a bit before I go to injection molding; 3D printed parts are trouble since they can't usually hit those tolerances and the mechanical properties just aren't the same.

Alex

For those qty's don't waste your time with a Brother or Robodrill, very good small Fadals can be had well within your budget, they're easy to fix and maintain, multiple sources for parts, and if anything goes wrong there's innumerable people who would know how to fix it. If you cam maintain a bicycle you can maintain a Fadal.
 
Since he is doing mostly plastic he could look for even a Fadal "A" model with the linear. Probably get one at an auction for $4k or less and still have money to fix it and be within the $10k. Plus production looks like it will be molded parts anyways.
 
To be efficient he needs spindle speed.

I am guessing he wants a machine not a project.

A used brother would be a better bet than most other things

IMO he needs to study tooling and process to get that kind of accuracy in plastics.
 
To be efficient he needs spindle speed.

I am guessing he wants a machine not a project.

A used brother would be a better bet than most other things

IMO he needs to study tooling and process to get that kind of accuracy in plastics.

Why would a Fadal be a project? A fadal's not going to be a "project" any more than a Brother or a Robodrill. I would think there's not many low hour Brothers/Robodrills out there that haven't run production and are worn out and within the OP's budget.

And if he gets a Fadal with a 7.5k or 10k spindle that's going to be more than adequate for what the OP's doing. You did read the part about 10-100 parts a day right?
 
And if he gets a Fadal with a 7.5k or 10k spindle that's going to be more than adequate for what the OP's doing. You did read the part about 10-100 parts a day right?

Yeah, and I also read about nibbling with 1 and 0.5mm endmills. If it's surfacing with a 1mm endmill, cycle time could easily be 20 or 30 minutes.

Regards.

Mike
 
Hi guys -

I'm looking for recommendations for a CNC machine that can be used to cut very small, very precise parts in plastic.

My typical parts would fit in a 50mm cube. They do have some tolerances down to +-0.05mm and walls < 1mm thick. The material is PC/ABS, ABS, PC/PET or similar. Geometrically, most of the features can be cut with a 1mm end mill, but some might require a smaller tool 0.5mm or even less.

Ideally I'd like to be able to do this both for easy prototyping of one-offs and for small scale production (10-100 parts/day). I think my designs are going to evolve quite a bit before I go to injection molding; 3D printed parts are trouble since they can't usually hit those tolerances and the mechanical properties just aren't the same.

I've done very little CNC milling work before (mostly whatever machines were at hackerspaces - wood or metal, no plastic) so I don't really know what I should be looking for here. A benchtop mill like a Taig or Sherline? Something like a CNC conversion of an X2? A small VMC eg Haas VF2? A mill specifically for micromachining eg Microlution, Datron etc?

What are the important specs? Which of these has the better positional accuracy / tolerance? What kind of tool and collet do I want? (and do I want an automatic tool changer?) etc.

Thanks for your advice! I hope this helps anyone else looking to make micro-machined parts out of soft materials.

Best regards,
Alex

For this I would recommend a Haas CM, it is more or less made for exactly that. Spindele speed is everything with these small tools, especially in plastic and aluminium. This machine has a 30k rpm ISO20 spindle, the tiny toolholders also makes balancing unecessary if you by good holders (Regofix has special made ones for this machine).
I have an older version of this machine (Officemill, same machine) and for what you describe it works very well. Thecompetition is manly Swiss made machines Kern etc. at about 3 times the price. Long time when I bought mine there was also a very nice and very expensive Mori.
I would not consider a Datron, if you want a more versitile machine go for Spedio or Robodrill. I would say 20K rpm is absulute minimum. I always run my spindle at 30K rpm and higher speeds would not hurt at all...

/Staffan
 
Yeah, and I also read about nibbling with 1 and 0.5mm endmills. If it's surfacing with a 1mm endmill, cycle time could easily be 20 or 30 minutes.

Regards.

Mike

20-30 minutes based on what? OP hasn't shown what he's machining, hasn't shown what details he's machining with the small endmills, may not know that you can rough with larger endmills and step down to smaller end mills for the fine detail work etc etc

Also OP may not know about making multiple parts in a vise, or vises, fixturing etc etc. I suspect getting the right machine is just the start of his education

Sure it'd be good to have as much rpm's as possible, but he might have to do with what he can get that works, and is within his budget, not yours.
 
Thanks! Researching the Speedio now - interesting!

Could you tell me why that's the best choice? It looks like a fast machine, but not specifically made for micromachining. What is the typical accuracy and how does it compare to the other options? Can it use 0.5 - 1mm tools?
I agree with what gkoenig said.

For the size of your tools, the correct tooling is more important than the machine by itself. Good tooling is $$$...

I machine plastic every day (mostly PET-P and PEEK) with the Speedio. You can achieve 0.05 mm tolerances with your eyes closed. But again, isn't just only the machine. Workholding and appropriate cutting tools are very important, specially with plastics.

Maybe somebody will throw me a rock, but I think you can go away with the standard 16k spindle. Many people think that you can go nuts with plastic, but it isn't right.
 
I have done what you are looking for with a DynaMyte with stepper motors and a 6k spindle, down to .005" end mills. A .02" end mill is not that small, you can still create chips with it especially with the plastics you mention. Look to Onsrud for your tools if they go that small, or Harvey but they will be twice the price, Harvey for chamfer tools to deburr. The biggest deal is you need a machine that is still tight, those little tools don't like slop. I agree with what everyone has said including the Fadal.

This is just my opinion, but if you have no experience with cnc, nor equipment, and just want to prototype parts before going to molding for production I think you would be better off having them made and just get a second job instead of trying to do it yourself. If your quantities are not in the thousands a year cnc production may be a viable alternative with someone with the right machine. Plasitc machines quite quickly and design changes are much easier than trying to modify a mold.
 
I've done very little CNC milling work before (mostly whatever machines were at hackerspaces - wood or metal, no plastic) so I don't really know what I should be looking for here. A benchtop mill like a Taig or Sherline? Something like a CNC conversion of an X2? A small VMC eg Haas VF2? A mill specifically for micromachining eg Microlution, Datron etc?

You've listed machines that range from a few hundred dollars to a few hundred thousand dollars.
 
I have an older version of this machine (Officemill, same machine) and for what you describe it works very well. The competition is mainly Swiss made machines Kern etc. at about 3 times the price. Long time when I bought mine there was also a very nice and very expensive Mori.
Another vote for the Haas here. We were looking for a mill for tiny work and these came up as extremely popular with people doing micro-work. In fact, when they came out Haas considered them something of a toy so they were cheap. But they worked so well that Gene figured there was gold in them thar hills and tripled the price. People still pay it.

If you can find one, grab it.
 
I used to do a fair bit of what I think you're up to to on my 16k speedios, but I've largely stopped. Here's why:

The design parameters for injection molded parts and machined parts aren't just different, they're more or less diametrically opposed. The easy to reach corners when you're milling are the hard to reach corners when you're making a mold. Molded parts want to be even and thin. Machined parts you want to leave as much material as possible. Geometries that are easy to mold, like embossed lettering and small square corners, are hard to machine, etc. Geometries that are easy to machine, like straight 90 degree walls, are hard to mold.

What I found was it would take the better part of a day to get a medium complexity molded part modified for machining, programmed, fixtured and run to get a prototype.

Alternatively, I could just have a rapid prototype molder injection mold me the same thing for about the same time and cost, then I get the next hundred essentially free.

In expensive plastic, machining gets spendy because you're throwing out 90-99% of the plastic you paid for.

There's nothing wrong with machined plastic parts. But they're designed like machined parts. There's nothing wrong with molded parts. But they're designed like molded parts.
 
If you’re realistically looking at making a lot of something, do it yourself. Id start by getting quotes for other shops to make them. Cost and headache go up the more sides you need machined.

I love my Speedio. Your parts size and material is ideal for it. From when I put the deposit down on my Speedio to the following 30 days I think I spent $40,000 on just the machine. There’s a LOT that goes into the whole initial setup outside of the machine. It would make no sense for me to have done this just for money alone, but I could no longer deal with my parts that ‘will be done in 2 weeks’ taking 7-8 weeks to complete and to be coming in the door out of spec. There are time, space and money requirements to own a machine.
 
You may want to look into a machine made to mill graphite electrodes. They are fast, and since high speed hard milling has put a lot of edm sinkers out of work, there may be plenty of good ones for sale.
 








 
Back
Top