V-carving (again)

(William Adams) #6

What sort of stock are you using?

How well do things match the preview?

Have you adjusted your feeds and speeds per: https://www.precisebits.com/tutorials/calibrating_feeds_n_speeds.htm


(Michael) #7

If you are running a #302 and your letters are 1" in height, your sample cut points to the Z not being properly zeroed.

Note: When I zero my V bits I take a small piece of paper or backing to Velcro/ double sided tape and lower my Z until its just barely over, then adjust to .01 and lower until bit just grabs paper and allows to spin, no further as your cuts will end up gouged when it would otherwise come out looking rather clean.

I will switch to phone and add image of test cut

I am running a file now, scooped up your image and converted. While my conversion was not perfect, here are the results.


(Michael) #8

Single cut, # 302. Cut center is @ center of material.
Be sure to barely kiss whatever you use to zero IE- piece of paper. Get the V bit close using 1mm, .1 then final approach using .01 can be used…
This is essential or job will not come out properly.
Paper should spin on center radius but not move xy.


(Philip Hack) #9

Hi Mike,

Thanks for that. I don’t have a #302 cutter, just using a 60 degree bit from Inventables now. I do notice on yours and mine, that the smaller letters actually look better. They have straighter lines. If you’ll look at mine, you can see that it’s most dramatic on the N. Also, at the top and bottom of the N, it looks like the cutter missed a pass and just left the outline there. It looks like a serif font. Very weird. The O also has a weird misshapen part of the inner circle on both yours and mine. Looking closer again, ALL the letters have that weird serif.

I’m just doing tests in luan, but that shouldn’t matter, right?

Will, do you have any insight into this?

Thanks, guys!


(William Adams) #10

I believe the 60 degree endmill you have is too narrow for the larger letters — that matches the preview, no?

Please see: V carve + inside pocket? and c.f., https://wiki.shapeoko.com/index.php/Carbide_Create_Basics#Clearing_area_around_drawing

1 Like

(Neil Ferreri) #11

@philiphack How are you zeroing your Z?
Got a link to the bit you’re using?


(Michael) #12


How wide is your 60* bit at its widest point? and are you sure it’s 60*, my #302 1/2", 1/4" & 1/8" 60* v bits all fit in the profile of the letters fine.

I see what your talking about on your O, mine is not like that, kinda looks weird in my photo but its cut through a knot in the wood and looks perfectly round.

You may want to go through your machine and check v wheel tightness, belt tightness etc… basically the daily chores.

If you want just clean letters for the bigger ones, I find finding the largest bit that will fit in the pocket to provide a clean cut and use that.
A 1/16th SE 2FL Endmill would fit inside General Store and provide a very clean look.


(Daniel Olson) #13

When your bit is near the material surface, can you wobble it left or right by hand? Your z axis might be coming off the rail.

If that’s the case, lowering your router relative to the z axis clamp or raising your material with a thicker waste board might help.

Can you take a picture of your machine setup, with the bit in position, as if you were cutting?



(Rick) #14

That very much looks like something moving where it shouldn’t, in amounts that would seem to be in excess of what I’d expect from deflection. As Mike says, I’d definitely check out belt tension and v-wheel fit as those are the likeliest culprits. Somewhat less likely is the stock itself moving - how do you have it held down?

If all of those are fine, then I think you get into the weirder stuff, like the stepper motor gears being loose. I say “gears” because it looks like the errors are in both X and Y directions - it seems unlikely that both would be loose.

I’d really start with the belts, though.


(Philip Hack) #15

Hi All,

Thanks so much for all your suggestions.

So far, I’ve rechecked my belt tension and all seems ok (though there really is no way to know as there’s no actual way to know just how tight they are).

There is no deflection when I try to move the bit in the spindle, nor can I really make the carriage move in any direction when I try. The whole machine just moves as one, which seems to indicate the assemblies are all tight. The V-wheels seem tight as well. The stock is held down by 4 clamps and is not moving at all.

My 60 degree V bit (specs attached) is .23 inches wide.

I just carved an N and an I. I watched the bit as it went. I can actually see it making those squirrely curved moves. Like there is something wrong with the code or the interpretation of the code.

The original letters in “GENERAL STORE” are about .125 inches wide, again significantly less than the bit width.

As far as the simulation goes Will, it doesn’t reflect at all what it carved. Attached is a JPG of that as well. Thanks in advance for all your help!



(William Adams) #16

The preview matches the cut.

Post the file and a screengrab of your tool definition?

Have you tried the inset technique I suggested?


(Philip Hack) #17

Hi Will, I haven’t tried the inset yet as it seems likely that there is some other underlying issue.
Attached is the screen grab of the tool and the file as well.
60%20End%20mill%20definition N and I.nc (6.7 KB)


(Philip Hack) #18

Sorry - here’s the definition from the tool creation menu and another carve of the same letters in a different font, which seem to me to exhibit exactly the same issues with the N as the other font.600%20definition


(Philip Hack) #19

Looks like my cutter width is wrong. It should be 6, not 3.175.

I’ll change that and re-carve.

1 Like

(Philip Hack) #20

Here’s the new carve. It looks way better to me. I’m going to try the full logo now and see what it looks like.

Cockpit error. I never would have found it without all your help and suggestions.


(Philip Hack) #21


(Philip Hack) #22

Here’s the full carve. Still a few anomalies (the top of the T and the top of the S, but it’s SO much better! Should I expect that level of accuracy, or can I do better somehow?

Also, the machine was going really fast in terms of IPM. I wonder if I slow it down if it will make a difference?

1 Like

(William Adams) #23

The top of the T and S look like mechanical issues:

It is also important to be sure that the collet is correctly tightened, the endmill fits correctly and doesn’t slip, and the router is mounted securely in the mount, and that the mount doesn’t shift. Note than endmill pullout can happen gradually, especially when profiling against tall walls.[4]

Also feeds and speeds may be a consideration: https://docs.carbide3d.com/support/#tooling-support and see https://www.youtube.com/watch?v=S9bceJxpqG0 for concepts on this and https://www.precisebits.com/tutorials/calibrating_feeds_n_speeds.htm for a testing technique and see the series #MaterialMonday: #MaterialMonday on YouTube

Beyond that it’s usually a matter of Calibration and Squaring the Machine c.f., http://docs.carbide3d.com/shapeoko-faq/how-to-calibrate-the-machine-for-belt-stretch/


(Dan Nelson) #24

Maybe I missed it, is your bit really 6mm, or is it 1/4” (6.35mm)? If it’s really 6mm do you have a 6mm collet or 1/4” collet? Is it a 1/4” bit that you’re defining as 6mm? That tiny little 0.35mm makes a difference.



(Neil Ferreri) #25

Hey @DanoInTx, he posted a drawing of the bit in a post above. It has a 3.175mm shank and a 6mm max cut diameter.

@philiphack I’ve used those bits from Inventables on a different machine. The angle was closer to 58° than 60°. Unfortunately, you may have to fine tune it with tests. Setting your zero accurately on flat stock is also very important.