[oldcrowe]

So- I'm teaching myself CNC... and have run into a problem.

In Aspire, I designed a simple 2" X 2" square, generated a pocket toolpath, and sent it to Mach3.
The results, seen in pictures 1&2, show that the square is 1.986 in both X and Y.

Next, to verify that my motor tuning is correct, from Mach3 I cut a simple straight line 2" long with a .25 end mill- pic 3. I also am posting a screenshot (picture 4) of Mach3 showing this cut went 2.0018 inches... 2" plus .25 for the end mill= 2.25, and I cut 2.2505- close enough for me.... So, motor tuning is (almost) right on...

So, why is my pocket 1.986 instead of 2"? What am I missing?

I also have included the .txt file for the cut...

Any suggestions/direction would be GREATLY appreciated?

Thanks so much for your time...

OC

[routalot]

Could be flexing of the machine,have you tried cutting two identical shapes-one climb cut and one conventional to see what differences emerge?It might also be illuminating to actually measure the tool diameter.

[dazp1976]

Machine flex.
Backlash.
Run-out.

Take your pick.


Look at backlash first on direction changes.
Doing a calibration check sees you only move in 1 direction.
If the axis has a start delay in movement on the way back, that's backlash.
Causes smaller than wanted.

[oldcrowe]

have you tried cutting two identical shapes-one climb cut and one conventional to see what differences emerge?

I haven't, but I will!

I'll also check exact dimensions of the end mill- good idea!

Thanks!!

[mactec54]

So- I'm teaching myself CNC... and have run into a problem.

In Aspire, I designed a simple 2" X 2" square, generated a pocket toolpath, and sent it to Mach3.
The results, seen in pictures 1&2, show that the square is 1.986 in both X and Y.

Next, to verify that my motor tuning is correct, from Mach3 I cut a simple straight line 2" long with a .25 end mill- pic 3. I also am posting a screenshot (picture 4) of Mach3 showing this cut went 2.0018 inches... 2" plus .25 for the end mill= 2.25, and I cut 2.2505- close enough for me.... So, motor tuning is (almost) right on...

So, why is my pocket 1.986 instead of 2"? What am I missing?

I also have included the .txt file for the cut...

Any suggestions/direction would be GREATLY appreciated?

Thanks so much for your time...

OC

You should not be trying to cut such a small pocket at 100IPM you are also programing in Incremental, and you should be using Absolute

Try a different Post Processor you have a G70 that should not be in there as well and a G94 is not needed the Incremental programing and these codes won't change your under size pocket

Slowing down the feed rate might help, you need a very ridged machine to run at high feed rates

[CNCMAN172]

Check your steps per unit using a large caliper or linear scale. As others pointed out try to slow the speeds and feeds and do thing change. Backlash also needs to be tested it can cause what you are seeing

[mactec54]

I haven't, but I will!

I'll also check exact dimensions of the end mill- good idea!

Thanks!!

Most hobby users can't measure a cutter to get the size, the easiest way is to cut a slot at a slow feed rate and measure the slot this will give you a more accurate cutter size, and what size the cutter will cut.

[oldcrowe]

So- I'm teaching myself CNC... and have run into a problem.OC

OK...SO...

I finally had some time to get back to the CNC and have checked everything you suggested, also ran shielded cable on all axis and spindle to make sure not getting EMI...

I also ran a new test:

In Mach3 Wizards, Rectangle Pocket, I designed a simple 1.5" square .3" deep
I ran it at 30 IPM, .125 end mill.
The results are pic 1

I designed that same square in Aspire V-carve with .125 roughing cut, 7.5 degree tapered ball end mill- results are pic 2

Pic 3 shows the measurement of the rectangle in Vectric: 1.5"

So- the router is working perfectly if controlled only by Mach3...
The signal from Vectric, while it SAYS it's going 1.5", is commanding Mach3 to only go 1.349...

Any more thoughts?

I really appreciate your time and attention to my problem!!

[routalot]

The obvious place to check is the tool library of your Vectric installation.If you open the file you created to check all the parameters for the cut,one of the options will be for the tool-click on it to edit the tool.You don't need to edit the tool unless the description is incorrect but having gone to the tool library,does the tool diameter listed conform to the description?It seems an odd choice to select a tapered end mill at all for the test when logic would suggest using an identical tool with identical speeds and feeds for a comparison test.Perhaps you should define a tool having the correct characteristics and try again with Vectric and it may be worth ensuring that you have selected the correct post processor from the exhaustive list.That would be the only obvious alternative area for a mistake to creep in.The fact that the machine does the job as expected from within Mach 3 suggests that the machine calibration is correct and the errors from using Vectric need to be chased down.My suggestions will eliminate a couple of possibilities.Could you post an equivalent image of the toolpath information?

[machinehop5]

...is Vectric leaving .075 stock for a finish pass or this could be a tool comp diameter/radius statement problem possibly?

[ZASto]

The signal from Vectric, while it SAYS it's going 1.5", is commanding Mach3 to only go 1.349...

This is normal as you've choosen TAPERED end mill.
So your pocket looks like \___/

[oldcrowe]

...is Vectric leaving .075 stock for a finish pass or this could be a tool comp diameter/radius statement problem possibly?

Well, I didn't tell it to save anything for a finishing pass... and this is a problem I'v been trying to sort out for well over a year, with dozens of (failed) cuts. And the finishing tool is set up properly in the tool library...
So, I don't know...

[oldcrowe]

This is normal as you've choosen TAPERED end mill.
So your pocket looks like \___/

So, two things.

First, I took the measurement at the very TOP of the cut, right at the melamine. .151 too small.

Second, the whole point of V-Carve is to get the finished taper correct AT THE TOP of the cut. So yes, there should be (and there is) a taper, a very slight taper (3.6 degrees) but the entire process is incorrect. The top of the taper should be 1.5" exactly with the bottom slightly smaller...

[oldcrowe]

It seems an odd choice to select a tapered end mill at all for the test when logic would suggest using an identical tool with identical speeds and feeds for a comparison test.

So- the reason I did as I did was that I had cut the Vectric cut first, as I was going to cut the male plug to fit, but was disappointed (but not surprised) by the results of the female pocket, so decided to do the Mach3 Wizard cut just to convince myself (again) that the machine was cutting to the correct dimension and tolerance, which it was (obviously...)

But you do make a good point- I'll do that next and post the results, along with the gcode and screenshots of the Vectric setup data...
Thanks for your response!!

[CNCRim]

I hate to bring it to you so far everyone talking about machine and so on but my experience is the material you cut "WOOD", nature of wood tend to contract and expand due to temperature. Running in summer time and winter time will give you different dimension, even exactly same program and speed.

you can experience with PVC sheet, will get much better result, more stable.