What's new
What's new

CNC hole, how round and smooth?

drcoelho

Stainless
Joined
Feb 19, 2017
Location
Los Altos
New to CNC. Assume I have a Brother Speedio S300X1, and assume that I'm doing a routed hole (not drill bit). Are the CNC controls of the bit cutting a smooth circle, or is it digital, e.g. X/Y are incremented to create the hole. In other words, if I were to look at the cut in a microscope would I see X/Y ridges or a smooth surface?
 
It should be smooth except for cutter tooth marks, unless at EXTREMELY low feed rates. The mass of the machine cannot accelerate and decelerate fact enough to create tiny steps as you envision. In addition, the motion algorithms in the CNC control smooth the motion. The only time I have seen the actual increments of control motion is when watching through a USB microscope during infeed of a 0.008" drill at 6300 RPM. That was sort of like watching a 1" drill infeeding in 1/8" jumps! The drill survived somehow.
 
If you are using G02 or G03 in your gcode, the machine's servos will be tracing out that programmed circle to near the limit of their ability to track. On newer machines, servos have encoders with much more resolution than necessary.

If your CAM program puts out circles as a long list of straight line segments, then your parts will have that polygon look to them, and the holes will not be round. There will still be small, rounded surfaces between these line segments because your machine cannot instantly change direction and the rotating tool also has a radius to it.

You will never see X/Y ridges like pixels on a screen, but you may see a part with a hole that is supposed to be round but is instead a polygon with hundreds of sides.
 
You will never see X/Y ridges like pixels on a screen, but you may see a part with a hole that is supposed to be round but is instead a polygon with hundreds of sides.

More like a polygon with hundreds or thousands of sides, each with a corner radius.

Moreover, while I expect a new Brother will interpolate a hole a lot rounder than any of the machines I deal with, you should anticipate that hole not being perfectly round (I would expect a brother might be round within a few tenths, while the 20 year old machines I typically deal with might be out as much as .001 or so)

If the print wants the hole round within a few tenths, you might want to bore it to be safe.
 
All round movements regardless of the code (arc or point to point) are interpolated movements made up of a series of pulses on linear screws so all will have some degree of faceting to them. The only way to produce a truly round hole is with a single pointed rotating tool like a boring head. Since you mention microscope if you're talking small sizes a reamer will get you much closer than interpolation and can be had very economically.
 
Nice machine for a first cnc machine, be patient and careful, machine lots of air.... don’t break it!


Sent from my iPhone using Tapatalk Pro
 
My understanding is that Speedios are outstanding at interpolating circles. On my S500 I was interpolating holes this week with a 1/4" end mill to form tap M10 holes. I don't have a way to gauge them, I can't say how great or terrible the holes came out, but it worked beautifully for me. It all depends on what you need it to do.

I "think" that the Speedio is about the best machine you can buy for doing things like that without adding a '1' to the start of the price tag and buying a straight up mold-level machine.
 
More positive to say about Titan Gilroy, than negative. Too bad he's not a member, though I suspect he'd get picked on quite a bit here.

R
 
I got in an argument with our shop manager because he said he made sure that he had shimmed a plastic pipe round using a pair of calipers to make sure it was the same diameter all the way around. I pointed out that the shape could be and likely was a trochoid that will measure perfectly on diameter with a two point measuring device, but can be very far from round. He claims to be a precision machinist but didn't know what a trochoid shape was, which means he really has no idea what round is.

I was of course completely correct as shown by the documents I presented to him and my boss but I got asked to work from home for two weeks to keep out of the way! He can only hide incompetence for so long.
 
Even with a Mit bore mic (3 point) I can’t measure a discernible variance in any of the holes I have interpolated.

If I drop a .0001” indicator in the bore, I have seen everything from NO variance to .0002” variance. Typically NO variance, but the hole location will be out a few tenths.
 
My understanding is that Speedios are outstanding at interpolating circles. On my S500 I was interpolating holes this week with a 1/4" end mill to form tap M10 holes. I don't have a way to gauge them, I can't say how great or terrible the holes came out, but it worked beautifully for me. It all depends on what you need it to do.

I "think" that the Speedio is about the best machine you can buy for doing things like that without adding a '1' to the start of the price tag and buying a straight up mold-level machine.

Pretty much any machine that's not clapped out is going to make a fairly round hole. Speeds may vary.
 
The most likely out-of-round condition you'll get interpolating holes is a jog at the axis reversal points (quadrants). The newer and tighter the machine is the smaller they'll be, to the point of being practically immeasurable in some cases.
 
I got in an argument with our shop manager because he said he made sure that he had shimmed a plastic pipe round using a pair of calipers to make sure it was the same diameter all the way around. I pointed out that the shape could be and likely was a trochoid that will measure perfectly on diameter with a two point measuring device, but can be very far from round. He claims to be a precision machinist but didn't know what a trochoid shape was, which means he really has no idea what round is.

I was of course completely correct as shown by the documents I presented to him and my boss but I got asked to work from home for two weeks to keep out of the way! He can only hide incompetence for so long.

You sound like an absolute joy to work with.
 
Even with a Mit bore mic (3 point) I can’t measure a discernible variance in any of the holes I have interpolated.

If I drop a .0001” indicator in the bore, I have seen everything from NO variance to .0002” variance. Typically NO variance, but the hole location will be out a few tenths.

Try interpolating a round boss, say, 1/2" diameter, then measure that with a good tenths micrometer at various clock positions. This lets you pick up axis anomalies easier than with a bore mic.

Regards.

Mike
 








 
Back
Top