Incorrect pocket diameter

[MFchief]

Hi folks,

My name is Terry, I'm a fairly new member with my first post, just been trying to read and learn so far.

Posting because of a newbie problem that is puzzling me. I have an 1100 and have been trying to use the Newfangled Wizards to drill and then circular mill some holes in 6061 Al, but the size doesn't come out correctly. The operation seems to go OK, I drill to .250 diameter, then using a decent quality .250 4 flute carbide mill I circular mill to a desired diameter of .770, but the hole only opens to
.722? I do not have any tools in the tool table except # 1 is an indicator, so I'm accepting the default in the wizard of tool # 4, setting the diameter to .250 and 4 flutes. When I get to the second page of the wizard I set the pocket diameter to .770, then set some very mild step over and step down numbers?

I'm don't have a clue what I'm doing wrong, since I'm new at CNC, any help would be appreciated.

Terry

[philbur]

I think as a first step it would help if you post the G-code.

Phil

PS: I mean on the forum.

[MFchief]

Phil,

Thanks for the reply. Since I was trying to run a program for 15 holes, and it would be a little long, I ran the wizard this morning for just one pocket over a pre-existing hole so I didn't have to drill. Below is the code generated by the wizard, the requested pocket was 0.785, the measured pocket was 0.736.
I set it up for 50% stepover and 0.06 step depth to a final depth of 0.28.
Terry

(Program Posted for Aluminum )
G0 G49 G40.1 G17
G80 G50 G90 G98
G20 (Inch)
(***** Circular Pocket *****)
M6 T4(TOOL DIA. 0.25)
M03 S4000
M9
G00 G43 H4 Z0.25
X0 Y0
G01 Z-0.06 F4.3
G3 Y0 X0.13 R0.06 F8.6
Y0 X-0.125 R0.125
Y0 X0.25 R0.19
Y0 X-0.25 R0.25
Y0 X0.2675 R0.2588
Y0 X-0.2675 R0.2675
X0.2675 Y0 R0.2675
Y0.1338 X0.1338 R0.1338
G00 Z0.25
X0 Y0
G01 Z-0.12 F4.3
G3 Y0 X0.13 R0.06 F8.6
Y0 X-0.125 R0.125
Y0 X0.25 R0.19
Y0 X-0.25 R0.25
Y0 X0.2675 R0.2588
Y0 X-0.2675 R0.2675
X0.2675 Y0 R0.2675
Y0.1338 X0.1338 R0.1338
G00 Z0.25
X0 Y0
G01 Z-0.18 F4.3
G3 Y0 X0.13 R0.06 F8.6
Y0 X-0.125 R0.125
Y0 X0.25 R0.19
Y0 X-0.25 R0.25
Y0 X0.2675 R0.2588
Y0 X-0.2675 R0.2675
X0.2675 Y0 R0.2675
Y0.1338 X0.1338 R0.1338
G00 Z0.25
X0 Y0
G01 Z-0.24 F4.3
G3 Y0 X0.13 R0.06 F8.6
Y0 X-0.125 R0.125
Y0 X0.25 R0.19
Y0 X-0.25 R0.25
Y0 X0.2675 R0.2588
Y0 X-0.2675 R0.2675
X0.2675 Y0 R0.2675
Y0.1338 X0.1338 R0.1338
G00 Z0.25
X0 Y0
G01 Z-0.28 F4.3
G3 Y0 X0.13 R0.06 F8.6
Y0 X-0.125 R0.125
Y0 X0.25 R0.19
Y0 X-0.25 R0.25
Y0 X0.2675 R0.2588
Y0 X-0.2675 R0.2675
X0.2675 Y0 R0.2675
Y0.1338 X0.1338 R0.1338
G00 Z0.25
M5 M9
M30

[hub]

I'm fairly new also.. But it could be a number of things.

- Tool offset value on diameter (wear compensation) wrong
- Backlash
- Wrong settings for lead screws (steps/inch or mm)
- Among probably a few other things

What controller/controller software are you using? (The 1100 doesn't ring a bell)
Are the dimensions wrong only when doing a circular pocket?

Hub

[MFchief]

Hi Hub,

Thanks for your help. I am very new at this, but the machine is a Tormach PCNC 1100 with Mach 3 (Tormach version), and is essentialy brand new. I did not set any tool compensation, and looking at the offsets page shows none being used. If this is not the right place to check it, I would appreciate being corrected! I have not done any other projects other than a couple of canned programs such as the roadrunner from Mach 3 and one I downloaded called Knight, which is a 2 1/2 D profile of a horse's head with a 1/16" ball mill, and it looks fine. I don't believe there is significant backlash, since the mill uses ball screws, is new, and it repeats to position very well when using a probe.
Terry

[hub]

I would check each axis anyway. Tell it to move a distance and measure if it did move correctly. Then, move it back, and measure if it returned to where it started. If not, there is some backlash.

Also, do some other stuff, square pockets, circles, text, and so on. And measure if they are correct within tolerance.

Hub

[hub]

Also, looking at the gcode.. the R0.2675 x 2 = 0.535 dia. Adding the tool dia (0.25) to that should make a 0.785 hole..

[R.DesJardin]

Not using a reground end mill are you?

[Andre' B]

I do not know about the final size but the first few lines should be causing the control to complain.

Ignoring the Z moves.
You first position at X0.0 Y0.0
Then a G3 arc to X0.13 Y 0.0 but you want it to use a radius of 0.06
You need a radius of at least 0.065 or 1/2 of the linear distant between the starting position and ending position.
Same thing on the next line with the F0.125

G00 G43 H4 Z0.25
X0 Y0
G01 Z-0.06 F4.3
G3 Y0 X0.13 R0.06 F8.6
Y0 X-0.125 R0.125

[MFchief]

Holes, the end mill is new, it's an Atrax from ENCO.

Andre', Is this a bug in the wizard, or am I not entering something correctly?

Hub, I guess the same question applies to your comment about R0.2675?

Thanks
Terry

[MFchief]

All,

How to say this?? First a sincere apology for wasting everyone's time. The endmill turned out to be off size.

After Holes question about re-ground, I took another look and the mill is strange, at least to me. It measures .2020 which doesn't translate to anything that makes sense to me, it would be 5.13 metric, closest fractional would be 13/64, but it's a little smaller than that?

I have never purchased a used endmill and have never had one sharpened, so I'm at a loss to figure out what I've got here?

Re-cutting the code with a proper .250 endmill results in the correct .785 pocket, so I guess that ends the questions anyway.

Again, I apologize for not checking it physically before posting the question.
Terry

[BobWarfield]

That endmill was out by a lot--did you purchase it on eBay by any chance?

I've gotten some that were out that much from eBay and advertised as new. They were regrinds or rejects one.

Still, it can be surprising how often even name brand endmills are off a little bit in diameter. Some manufacturers make a real point of emphasizing theirs are not. The standards are pretty lenient.

If accuracy matters, I like to measure the EM rather than taking the manufacturer's word for it.

Cheers,

BW