Gouging showing in Verify, but not in classic backplot, not sure which to believe

[huskermcdoogle]

I am trying to helix swarf a negative tapered slot...  I think I have it pretty well figured out. However It appears in verify to be overcutting the top, does not show in classic backplot.  Which should I believe?

 

 

 

 

Undercut Tapered Slot.mcam

[huskermcdoogle]

Nothing?

[Jim at Gentex]

The problem with backplot and verify is that they run the nci file for the simulation.

So the answer is you can't fully trust either one.

The BEST way to find out what your toolpath will to is to run Vericut or some other verification software that actually runs the NC code in the simulation. Then you know exactly what you should expect to see when you run the NC program on the machine.   

[huskermcdoogle]

Can't afford any simulation software at the moment.  Someday when revenue gets a little better and we are booked up solid, I'll go after a package.  Firstly I need to get Verisurf, as that would help me tremendously on the surface creation aspect of things working from scan data or repairing xxxx poor reverse models that are provided to us. 

Ended up running the parts and they came out good as far as I can tell.  Used the output angles as my litmus test.

[Jim at Gentex]

On ‎5‎/‎10‎/‎2017 at 11:53 AM, huskermcdoogle said:

Can't afford any simulation software at the moment.  Someday when revenue gets a little better and we are booked up solid, I'll go after a package.  Firstly I need to get Verisurf, as that would help me tremendously on the surface creation aspect of things working from scan data or repairing xxxx poor reverse models that are provided to us. 

Ended up running the parts and they came out good as far as I can tell.  Used the output angles as my litmus test.

Yeah that's probably all you can do if you don't have verification software.

Find the trouble spots in either backplot or machine sim, then look at the NC code in the corresponding area of the program.

If it's going to do something weird you may get lucky and see it in the code, but that is a risky way to go about it! 

Depending on the dollar value of the parts you run, Vericut could pay for itself many times over in a short period of time.  

[huskermcdoogle]

Dollar value is less than fifty typically.  Never really exceeds a few hundred.   So it would take a while to pay back on that front.  Now if I could have very good confidence of my programming results, and could possibly simulate results of things like poor work offsets and such.  That could pay for itself pretty quickly with productivity and not burning through as many blanks correcting issues.

[David Colin]

I post your project in Vericut (with my machine/post of course) and it looks good.

Here is Vericut auto-diff (0.01 tolerance) result.

[huskermcdoogle]

Well dandy!!  Thanks David.  My eyeballs and calipers told me I was in good shape, but didn't have full confidence in the code.  Good to know that sometimes you can trust your geometry and selections in the tool path to get a good result.  Now if only verify would generate a proper part.  My guess is it has to do with rounding of the vectors, or how it smooths between points.

[?Mark]

Verify should be correct. Are you sure that your tool is showing correctly in backplot/verify?

I use camplete and vericut daily, but found verify to be pretty solid with the exception of rapids and transition moves especially in mutliaxis applications. In 3axis mastercam backplot/verify is very dependable.

[David Colin]

'Backplot' seems OK and i looked again 'verify', it should be ok but it's clearly not here (stock model isn't too).

I tried to modify tool definition and i could get good verify with a cutting diameter  <Ø17.

I still could have doubts but I exported tool as it was in Vericut and it verifies OK.

IMHO, it should be send to QC.

[huskermcdoogle]

Sent in.  We will see what they say.

[huskermcdoogle]

Apparently if you change the collision tolerance in the options in Mcam Simulator, it gouges less and less as you tighten the tolerance.  .05 mm gets a decent result. .025 is pretty much perfect.  So it pretty much comes down to a system tolerance issue....

[David Colin]

Yes that works for verify (but slow down overall) however stock model still isn't ok...

[huskermcdoogle]

Added that to the dialog with QC, good catch.