I own and operate a small Lego-compatible toy company (www.brickarms.com) that sells small (1/2 gram) detailed injection molded parts. I had 8x8 steel molds made, but they are very expensive. I'd like to mill small mold inserts (3"x3") myself, for use in a Mini-Jector Wasp 45 bench-top injection molder.
I need to have high tolerances in the cavities, and no parts are over 1" in length. Is an X3 overkill? If I go with X2, will I still get .001 tolerances after CNC conversion (ball screws)?
I'm no stranger to the electronics of CNC, but with my experience being solely in making a CNC wood router, metal-working is new to me.
Any suggestions on which mill is a better choice?
I have a hard time seeing anyone of them doing well with die steels. Not impossible with the x3 I suppose but do expect issues with getting good surface finish.
As to the conversion and accuracy that depends on many things not just the quality of the lead screw. Fit up and adjustment of the ways is very important. Keeping high tolerances over the small distances discussed might be possible with a lower cost lead screw, but do realize if you go with fully speced ground lead screws that you will end up sinking a lot of money into the machine that could just as easily go into a Bridgeport. Your final costs going to the bigger machine (purchased used) might not be that much different from the cost of a new X3 and a conversion.
As an aside a couple of years ago I had an interesting conversation with a guy at Cabin Fever. He started up his own business much like you are intending to, that is with a converted import. Once his demand got to the point of being significant he went out and got full CNC machines from HAAS. This immediately solve a bunch of problems for him.
Now I did not relay the above story to turn you off to the idea of trying to build a business with such conversions. Rather You should not get your expectation up that high with respect to what you want to accomplish with this mill conversion. It will take considerable effort on your part to get the machine to the point of actually doing what you want. Keeping it in that condition will take effort also. You won't be able to CNC the machine and totally forget about its heritage.
I am almost certain that the rolled ballscrews are better than .004" per foot. Also, if your parts are usually in hte 1" range, I can almost promise you that the leadscrew error will not exceed .001" in 1"! If you were doing tenths work, it may not be suitable. Mach 3 has a leadscrew mapping utility if you have the means to do that.
I would recommend the X3. I've gotten pretty good finishes on steels with just normal end mills. If you used a carbide end mill, and a carbide insert facing tool when appropriate, you would get very impressive finishes. You just can't take off .25" at the same time! Make smallish finishing passes, you will be pleased. The X3 is an impressive machine, and it's low speed RPM's cater nicely to your needs. It's got a lot plenty of dovetail contact, by bench top standards. When my mill was in manual mode, I found it difficult to get it to chatter before the motor would bog or the tool break.
I owned an X1 for almost a year, the general quality of the machine is significantly lower than the X3. I suspect (though not certain) t hat the X2 has similar quality as the X1.
With the X3, you've got a larger table so you could potentially setup multiple vises and run several things at once. I have made molds at work before, many of those we make are from D2 tool steel. Having worked with the stuff, I can confidently say the X3 can take it.
I guess I will jump in once more since NC_CAMS has not alrready
I think you will get more error from using cheap bearings for your leadscrew support, than with the use of cheapish rolled ballscrews over a few inch lengths.
If you want to do it right, and greatly reduce your accuracy issues regardless of which screw you find, use angular contact bearings on all three axes. If you want the work done for you, and an assurance it is done to the highest quality, buy duplex bearings. If you've got mechanical inclination, you should be set with angular contact bearings. Then again, you need to have the proper tools as well. Duplex bearings are not cheap, but it's the only way to go if you want real accuracy. I know what it costs to have a mold made, I can assure you even with the price of duplex bearings, the machine will equate to a big savings.
Thanks for the input, guys. It sounds as if I choose an X3, then the quality of components in the CNC conversion is the most critical decision.
From what I can tell, the Syil-America Super XCNC kit looks pretty good, with double ball nuts on good ballscrews. The benefit that it is already assembled means I won't screw it up and add slop by during disassembly/assembly for the CNC conversion.
One last question for you experts: Is a spindle speed of 3800 fast enough to mill aluminum and achieve a smooth finish? I will be using a 1/32" cutter on my finishing path. I have read that running these tiny cutters at 7 IPM is typical, but I've also seen mention of spindle speeds in the tens of thousands.
The Syil kit looks pretty good. 3800 is not much for aluminum. Well, consider that a bridgeport has a max RPM of around 4500. It's alright, but at 3800RPM with a 1/32 you will be feeding very slowly. If you have to do it a lot, you might look into some sort of attachment or something to get that speed. Maybe make some hardware with your mill to mount a router on the side of the head for small cutter work? Then you can get your high RPM
seems to me that an imaginative person, with some good mechinacal skills could come up with something like a tapping rig, that would chuck up in the mill and double or more the rpm's.. with a really small cutter, seems that it wouldnt even have to be super-duper strong, just real straight and repeatable...
Grizzly X3, CNC Fusion Ballscrew kit, 3 500oz-in bipolar steppers, 3 203v Gecko's, Linear power supply from Hubbard CNC, Mach 3, BOBcad Pro Art V22, Rhino.
Besides accuracy, it needs a gentle and smooth tool pressure using those small carbide cutters.
Machine mass (IMHO) is a non issue since you could use a Dremel by hand and get all the tool pressure a 1/32" em could take.
Where to place it, by hand, is the tricky part
Perhaps Syil can help out with a 20-50k spindle for it?
Otherwise, the improvision by you could be the most difficult part.
It seems the postings of a machines value is primarily rated by max cu/in of metal it will whack away.
In my 50+yrs of making tooling and molds it is just the opposite. Machine and tool accuracy are paramount.
Cubic inches of metal chips piled up was of little concern compared to breaking the cutter.
For high speed, you could get a spindle speeder or side mount a grinder like this:
For those tolerances, you will definitely need anti-backlash ballnuts. The Syil X2 only has single ballnuts, where the Syil X3 has double ballnuts.
Thanks for all of the tips, guys.
I ended up buying a Taig 2018CR-ER from Andrew at www.computersculpture.com. It seemed my best bet for high resolution and for milling the tiny stuff I need. Besides, the upgraded spindle runs at 10000 RPM - which will be great start as I learn the intricacies of high(er)-speed aluminum milling.
If the 10K spindle won't cut it, I'll try an even higher-speed spindle. Since I'm milling with a 1/32" cutter on my finishing pass, and using a .00625 stepover for a smooth finish, I'd like to get more than 7 IPM when I'm cutting. Maybe 60,000 RPMs will be the way to go. Zing!
Truth be told, I picked the Taig after realizing that there was no way my wife would allow the Super X3 beast into our small garage!
BTW - Andrew from even called *me* to be sure my machine choice matched my needs, and he was helpful in recommending collet selection and material selection for my application. He really knows his products and I wouldn't hesitate to recommend him to anyone "on the fence" about machine selection.
Last edited by GammaWill; 03-14-2007 at 03:10 AM.