Smear clip Art

[michiganbill]

I must be doing something wrong, when I use the offset for 3rd finish, most of my clip art is fine, but seem that the model is cut with a "smear" between the object and the surrounding base or other object. I am preparing to cut around the outline to clean up the smear (and then get a chisel out and clean up what I don't get with an endmill. But is there something I am doing wrong or could do better? Seem this has always been there since the 1st aspire I got in 2010, but now I am using it more and it seems that I am missing something.

Smear on Hearts & Banner

smear Settings

any help would be appreciated!
Bill

[mtylerfl]

Does your Toolpath Preview look correct? If not, then it’s a layout issue.

...or a bit spec issue...

Are your bit specifications matching the actual bit you put into the collet?

Could be a mechanical issue if the above have been addressed.

It would be cool if we could open and observe an actual file.

[michiganbill]

The tool path previews look great and yep the tools in DB match the bits. Most of the object turn out great. I am using mach3, maybe there is something in the Mach2/3Arcs(inch).txt that is "smearing" the steps or something.

The model seems to be good in that when I tell it to trace, it is fine.

I do appreciate the suggestions!

Traced Clip art with 1/8" EM

[michiganbill]

Will the forum allow a zip file? it is 6mb

[mtylerfl]

Possible helpful tip...

Create a small Base Height under each “troublesome” relief model you may encounter. This raises the model a teeny bit and may help avoid the “smearing effect”.

Disclaimer: Since you report that your Preview looks correct (no smearng), this tip may not apply nor be useful!

[michiganbill]

Thank You - I will try that now - I did just screen capture the Tool Path Preview shots, the look pretty crisp

Will be trying as soon as it gets off the machine

I will outline object, create thin object from it and add clip art on top and see what happens - thank you again.

[martin54]

What have you got your z heights set up in the toolpath setup form? If it is set really low then it could be that you are catching the material when the bit oves to a different area

I have my z1 & z2 set really low & have had that problem when the material wasn't as flat as I thought it was

[michiganbill]

you mean this?

[mtylerfl]

No need to make a separate component/model.

Just select a model and click the Wrench icon to see the Properties of the model. There is a Base Height field you can type a value into.

(There are several other ways to open the Properties menu, but no matter)

[martin54]

yes that is the one, looks like all the marks are at the surface which is why I asked, the other thing I meant to ask was if you were using a zero plane which helps to define the edges but I think that would affect the preview as well as the actual cut & you have said the preview looks good

[TReischl]

A bit of trouble shooting:

1. If you run the file a second time do the "smears" repeat in exactly the same place?

If they do not, then my best guess is that you have some Z axis issue. Sticky. Backlash. Play.

One of the things I do with all models is apply a tiny bit of general smoothing. Might help, might not.

BTW, it sounds like you have had this problem for a while or all along?

[highpockets]

Looking at the banner there's in shift in the machining of the model about half way through. That could be your stepper motor loosing steps. What feed rate are you using? You might try dropping the feed rate down to a low number and see if it helps. If so start working on a optimized feed rate.

[michiganbill]

Highpockets, , The shift halfway through the Banner is the mill changing from counter-clockwise to clockwise and indeed might be some slop in the CNC, but also is interesting that the smear is on the leading edge or approach to the clipart (you can see that as the tool path changes direction the smear goes from the right side of the model to the left.

TReischl , I looked back at some of the older things I had done but not sanded (my ooops pile) and yes, it has been there a while (I think event version 4 - so don't think it is unique to a version of aspire - have not upgraded MACH either). I am setting up a little experiment where I can dive into the feed rate and the G-Code with out sorting through all the lines.

it is interesting that when I cut the individual models (say the heart and banner stand alone) in offset there seems to be no smear, but that is part of my test as well.

Thanks all, will post some results when I can get them