CNC Mill: Sieg KX1, KX3 or TS M1?

I doubt JS would use a four flute end mill to machine a perfect slot, it's not what they are for. I suspect that a two flute cutter aka a slot drill is probably what you meant.

Is it possible to comment on cutter deflection without knowing the feed rate, cutter accel and depth of cut?

1/ grab a dial gauge and a spring scale and measure how much force it takes to drag your table around the backlash. 2/ assume something fairly standard like a 6mm end mill and apply those forces to it and see / measure what happens.

I could possibly move the table roughing or hogging, but if I'm roughing or hogging I don't care about a couple of thou, if I'm finishing and I care about cutter deflection amongst other things, I can't put enough force into the work to move the table around, but don't take my word for it, this is easy enough to measure.

what you're getting here is people who wanted ballscrews, who bought ballscrews, and who don't want to hear anything else about anything else.... you don't HAVE to go to the expense of a ballscrew to eliminate backlash either, if it is that much of an issue, but like I said, go and measure your table, numbers don't lie.

I don't think I said anything about the number of flutes, but 4 flute instead of 2 is legitimate if like me your spindle rpm is limited to

2500

in the sense that cutter deflection != zero, and 2 thou backlash is easily exceeded by cutter deflection, yeah, it is.

in the sense that it is also easy to measure the amount of force needed to overcome friction and move the tables around backlash (eg circular pocketing) then you can work back from that force to material removal rate and work back from that to mill dimensions and mix mill dimensions with the force to get mill deflection

what gets me here is this whole attitude that just because the cnc software works in hundreths of a micron bolting it on to a machine tool is suddenly going to make that machine tool inherently more accurate, making something fairly simply like a carb manifold accurate to a thou in every direction is both un-necessary and a bloody sight harder (and slower) than people assume, but just because the software says it is so people assume it is so, and don't want to hear different.

If I had a quid for every time I've offered to take a made part to a CMM and compare notes to what the gcode said it was making, for free, hell I'll buy you beer and lunch if you like, and not one person has ever taken me up on it, why not, what can they possibly be afraid of from the output of a certified and calibrated CMM?... the ONLY thing it can tell them is the actual dimensions of the object, ergo they don't want to know, so why don't they want to know.

but then I'm just ranting aren't I, and those who have splashed out on ballscrews have the technical high ground merely because they have splashed out on ballscrews.

Like I said, if my CNC conversion ends up being good for a thou I will be absolutely over the moon.

Nobody apart from you has suggested that you had to have ballscrews for backlash elimination, as far as I can tell.

Regards, Tony

Yes.

Rubbish.

Regards, Tony

well that's your opinion and you're entitled to it, which is why I suggested the other guy actually measure the force required to move the table around backlash.

let's wait and see if he is the first example I have ever seen (excepting linear rails and ballscrews) where the cutter can make finishing cuts and overcome friction and stiction and move the table around the backlash slack

I get the distinct impression I have trod in something................................

John S

That's very generous of you. However, in this case, I was simply agreeing with you

Regards, Tony

Easily done ;) Some just want to make a meal of it!

Wayne.....

Sorry to bring back controversy, but are linear rails much better than slides? How much better? Bronze or linear ball-bearings?

-- Peter Fairbrother

(who may be overimproving his tiny lathe to the extent that it bends...)

Linear rails are less friction and stiction than traditional slides and gibs.

As to "better", you can whack a linear rail and forever bend it out of true, you can't do that to a slide.