[duvlitho]

Hello fellow Novakon owners,

I was hoping to see if any of you had incurred a problem similar to mine. I am setting up my Pulsar with stepper motors in Mach3. I was performing some test cuts to verify the mill capability after measuring/adjusting steps per pulse and backlash. My current feature I am trying to machine is 1" circle, 3" circle and 1" box with a .25" endmill. I performed to axis calibrations with a 4" gauge block and the circles are uniform when cut. My machined results are .020" smaller than the dimension I wanted. The error doesn't scale from the 1" to the 3" circle (same .020 error on 1" and on 3" circle). I think I am just missing something simple... I'm almost there! My cam software is VisualMill. I didn't have this problem with my converted G0704 so this error is new to me. Any thoughts of what I should look at? I thought I went over everything but obviously I'm missing something! Thanks in advance for your help.

Dustin

[Titaniumboy]

What is the measured diameter of the endmill?

[duvlitho]

Titanium - I am using a .25" 4 flute endmill. I drew this up just to test the capability. Attachment 297954

[duvlitho]

Attached is my code. I parsed it out in excel and it seems to make sense to me with the diameter of the cutter ((MAX X - MIN X) + diameter) & ((MAX Y - MIN Y) + diameter) for each feature.

[Titaniumboy]

Titanium - I am using a .25" 4 flute endmill.

The nominal size is .25", but what is the endmill diameter when you measure it with a micrometer?

You always have to measure the actual size of an endmill when you run a CNC machine.

[duvlitho]

Agreed... .2493 I should have posted that before orginally. Thanks for the help.

[LeeWay]

Yep.
That has bit me before too. Especially true with router bits. Not trying to be punny.

[Titaniumboy]

Agreed... .2493 I should have posted that before orginally. Thanks for the help.

Well darn, those endmill measurements don't explain your problem.

Perhaps try generating some code by hand and see if you have the same issues? That way you can eliminate the CAM as a potential problem source.

[Titaniumboy]

I did a quick glance at the code. That is a ton of code without very many comments at the safety lines.

I don't see any G41/G42, so the code is not using any cutter compensation. I've been bit many times by old settings in the tool tables.

I would try indicating your spindle center line to the center of one of your incorrectly sized holes. Then enter some simple MDI code or write a really really short gcode program to do a hole of the correct size.

[LeeWay]

I will say that I calibrated mine using the Mach 3 calibration aid. I don't recall the exact results, but I think each axis did change a little. I used a 2" dial indicator. You should look into getting one of those as well as a micrometer if you don''t have them. I mention that because you use a block. Dial indicator is much better. Shars has a good selection that preform well for not a lot of cash.

[mklotz70]

Any chance you had the CAM leave the .020" for a finish pass?

[duvlitho]

Thanks guys for the help. I use gauge blocks to setup the pulse per inch in Mach3. Attached is a pic on how I do on the y axis(similar for x & z). I do use a dial test indicator to zero my y movement before moving 4 inches in y. You'll notice in the pic that I took the pic before the dial test indicator scale was zero'd... it was performed after the pic. After 4 inches of movement I had almost zero error. I have had excellent results ringing a 4" and 2" gauge block together to calibrate over 6 inches. Just make sure your setup is tight and so that you don't induce error from your indicator stand moving.

I received my lake shore carbide order late Saturday afternoon. I redid my code for the new feeds and speeds. It cut spot on when measured with my calipers. I'm now very happy with the results even with the caliper measurements. I'll make some test parts and measure on the granite block with gauge blocks. I'll have to revisit indicating my toolholder and collet that was giving me error. I'll measure it later this week.

dustin

Attachment 298090

[LeeWay]

Ahh. Sounds like you know what you are doing.
Glad to hear that is working better. I bought a few tools to try from them. I started out using Chinese tools and some of them actually work okay. HSS mainly. I am still going through the supply of those that I had bought for some sizes. They last a pretty good while for us. Using premium bits in those cases would likely not do any better. Small end mills that do mostly aluminum at less than optimal RPM's. I am talking about 3/16" and 1/4" mills. I use Kyocera end mills for our 1/8". They compare well with the one I received from Lake Shore Carbide in the same size. Less the coating. All of my specialty tools are premium brand name tools though. You can tell a difference in those. No doubt they are worth the extra cost.