Left click should select all and right click only I believe.

Vista(Epicor) here, Syteline at the last place, JobBoss at the place before that...

I don't have much to do with the software but from what I heard they all have their pros and cons.

Consider the only thing an outside post takes into consideration is the code in the NCI.

Translation of the 1027 line(which should be the t-plane I believe) of the NCI code into aptsource should either do a matrix translation of your XYZIJK or output a tracut of some type into the aptsource assuming your WCS is actually set the same as the machine X, Y, and Z axes and the t-plane is defined relative to the WCS axes with the same origin point as James said.

I always played it safe and output a max feed typically with curve 5-axis like you are doing if it was more than about a 45 degree transition, as I was not sure how the head of the Mag3 would position at rapid. Does the TCP control the tip of the tool in rapid accurately?

Should be a setting in the control definition under the Arc tab to allow or not allow helix motion

May be somewhat of a tangent, but now I am curious.

Does Mastercam now drive the solid model independent of how the surface on said model is defined?

Take a flowline toolpath for example. If the the surface of the model is defined by curves(isoparms) then doesn't the toolpath follow said curves?

In my experience there are circumstances where even creating surfaces from a solid model results in an inverted surface normal and very often skewed isoparametric definition relative to the part.

I am a few versions behind however, just curious as to driving the solid necessarily negates the need for modifying or rebuilding surfaces due to normals/isoparms etc.

As a general rule our actual cycle times are approximately 30% over the program cycle time due to what James said about acc/dec, pecks, etc.

I say approximate because since 2 parts may vary in the amount these things influence it. It can be used as a starting point, but really the only accurate way to schedule that I am aware of is to actually base it on the cycle time at the machine.

Yes, but only on tool changes/forced tool change sequences to check for cutter wear, etc.

Koma/Alberti were helpful and have a good product IMO. We are running a slim and a mid-sized one through tool changers. Give them a call, they came out to us and got all the info for the cut parameters we needed so they could suggest the best models.

It does happen on a rather rare basis mostly on big engagement cuts on .750-1.000 diameter holders that are probably close to 5 years old. Problem is we have hundreds of them. I checked out one that recently failed a bit more and found about a .0005" deviation from top to bottom(axially) on the inside so perhaps that is an indicator... Maybe stick a blank in as you suggested Ron and test it...

Hmmmm....

Libraries are really the way to go IMO for cutters. Better even than generating it with the vericut interface. They are a pain to build in Vericut but I prefer to build it manually once and get it over with. I only use import STL's in the tooling for holders. I don't use the interface to generate cutters. There is a decent library somewhere in the examples I think that you can modify. Of course if you don't have a standard tool list then the interface will probably be best.

Stock and design models I always STL out of mastercam in World coordinate system then I have a post that outputs a Vericut matrix so that when you simulate it it builds a coordinate system instead of trying to do all that manually. BTW the chordal tolerance you STL out your models on large parts has a lot to do with how fast Vericut will run. We don't have a MCX interface so I do still do that setup manually, but we have a Catia interface which is wayyy easier than manual set up.

I'd really suggest taking at least the first class CGtech offers. It will save a headache or three.

Finish on the inside looks fine, the color of the billet ones looks goldish to purple. On the black finish ones, we were told when they begin to crystallize they are done according to the manufacturer. The billet finish ones however don't seem to do this. We have asked them what else to check and got the "Wellllll.... " type answer. And yes... some have been way overheated. Hopefully that problem has been corrected, but it is difficult to tell which have been overheated and which not and to know when they are going to fail until... they fail.... in a part.

Greetings.

We are looking for a simple way to check heatshrink holders for end of life cycle. Anyone run into these holders eventually degrading and having tools pull out? I thought checking the bore for egg might indicate it was bad, but we have a bad one that is less than .00005" deviation which I imagine is within factory specs.

Thanks in advance

Is B set up with a +/- rotation? Worked with a machine before where B was absolute 0 to 360 and the + or - dictated the direction it would rotate.

I started with Mastercam 6... used it until MCX2(mr2sp1) and still use it some. About 3 years ago I started the transition to CATIA as the shop I work for was going away from it and only to CATIA.

I will say this:

1. CATIA is more expensive by I would guess 30%

2. You will have to buy a post software of some type

3. The learning curve wasn't so bad for me, but there are college level CATIA classes available locally here. Mastercam is more of a VOTEC class AFAIK locally, I was OTJ trained in MC.

4. The CAD side of CATIA is awesome.

5. The CAM side: (consider I am using x2mr2sp1, I do not know if any upgrades change the following opinions)

-On basic toolpaths CATIA is somewhat cumbersome as you find yourself making a bunch of clicks for stuff that is irrelevant. Geometry selection is easier usually though.

-The way the toolpath catalogs are in CATIA I do not like as much as Mastercam toolpath defaults

-The approach/retract macros in CATIA blow anything that Mastercam offers out of the water

-Full multi-axis toolpaths in CATIA offer more power IMO than Mastercam, BUT when things go wrong when processing, it is easier to FORCE mastercam to do what you want it to.

-Cycle toolpaths sometimes are a pain in CATIA next to Mastercam.

-I always created more geometry in Mastercam to drive than I do in CATIA

All in all, they both have their strengths and weaknesses. I prefer CATIA now primarily due to the the things I mentioned, but either package is capable on the CAM side. If it were up to me the deciding factor would be the type of work I did. If it is aerospace where models are primarily in CATIA from creation, and it is large multi-axis parts then CATIA I would choose. If it were not aircraft and/or primarily 3 and 4 axis work on smaller-mid sized parts I would choose mastercam.

I can't get to the file due to IT restrictions so not sure, but I have seen what you are describing where two surfaces meet and are possibly not exactly tangent therefore it has to stay normal to the end of the first surface, then back up to stay normal to the start of the second.

Could this be the case?

If I remember correctly... been a long time... something like if it can calculate the toolpath inside the containment boundary it won't do it as one would prefer, but if the containment boundary actually constrains the toolpath then it will. Might try a tighter boundary or extend teh surfaces to overlap your existing one.

Edit: Nevermind lol Just tried it.

You do have to have that box gcode mentioned on though if any of it goes outside the containmnet boundary.

Switch your transition type... it doesn't work with follow surface edit: or ramp

I find it rather amusing that you have to use an outside piece of software to verify anything that Mastercam does.

I agree to an extent, however when you are talking about complex multi-axis motion then the rules change to a degree. The calculations the post is doing becomes far more integrated into the code and the need to see this when you are talking about a high dollar piece of material becomes rather important. I use CATIA day in and day out and Mastercam now only some of the times, but regardless of the CAM system I would never... ever release a program to the floor without running Vericut.

For me it is a matter of speed and mistake proofing. While those quirky CAM issues that cause issues do exist... 99% of the mistakes that could potentially make it to the floor are my own. They come from a lack of understanding what the software is doing as well as having my head up my... oh wait.. it was operator error.

Sounds like a wing spar...

Even in Vericut if you do not have suffecient RAM for this, it will be necessary to break it up in sections. Vericut has features where you can compare the cut model to the design model in sections to reduce the necessary memory.

It may be possible to define your stock in smaller sections say 50" to use a similar concept in verify to get the resolution you want, however I am total agreement with J in that you aren't verifying code... it really isn't the best option. Vericut and a computer to handle that size part is the best option IMO.

There is a C-hook called Nohist that will do this. At least in X2 there is...

Edit: Ha! too late...

Lintol is an APT statement short for Linearization tolerance. Most other CAM systems use APT of one kind or another as opposed to an NCI file for CL data.

Basically when a tool swarfs from one vector to the next, the tip of the tool will not track in a linear motion, although you want it to. It will track in an arc or curve. So the function of LINTOL is to set a chordal tolerance for the post to recognize when the arc deviates from the linear path you want the tip of the tool to take and add additional points in so that it tracks closer to a linear path. I.E. break up the rotary moves and adjust the XYZ accordingly.

With a nutating head or table/table machine it has the potential to want to make a C-axis move of up to 90 degrees in a tiny linear axis move so as A-axis approaches 0(but not at 0) the potential is large for it to sweep around into a wall regardless of how many vectors the CAM system outputs. This doesn't happen all the time, but there is the potential.

I should have read it in your first post I suppose. Heh

So tool axis vectors that intersect with a 0,0,1 vector where it is transtitioning from swarfing one direction to another cause a potential problem the way any CAM system outputs CL data for a table/table or nutating head machine. The CAM system does not know what the rotary axes will do. The post decides this. That being said, this is the need for LINTOL in the post so you do not get a huge rotary and tiny linear move. The post needs to be able to break the rotary moves while maintaining the tip of the tool in the proper position else you will get the tool sweeping in/out of position.

Don't send anything out on the floor without verifying it with a gcode verification software IMO. I have heard of ICAM's, but never seen it in action so I can't say good or bad about it.

Sounds like fun.

What machine is it?

A couple things of my opinion:

LINTOL Your post needs to handle it IMO if it is a nutating head or table/table rotary machine. 4-axis, not so much.

ICAM and Austin NC are both options for other post processors. I believe support is better with Austin NC while power is probably a bit better with ICAM.

If you have a machine with a ceramic bearing spindle, hitting one bolt and trashing the spindle will pay for Vericut.

As far as G43.4, my experience has been whenever the machine is in 5-axis motion, it likes lots of points/vectors. It does not need them on locked axis cuts so much as it does when it is swarfing. I have been on the switch to CATIA so I never really used the newer 5-axis toolpaths, but in the old toolpaths the best luck I got was turning on point generators based on an angular deviation. I think I defaulted at .1 degree for our Makinos.

Mastercam is perfectly capable.

Do you need something specific or help with the whole thing?

This is no simple task if you don't have previous experience with it.

You would need all the probe routines set up as subprograms in the control. You would have to know how to pass variables back and forth in the control and possibly build subprograms to specifically call what you want. Then you would have to have your post set up to output the probe calls when you designate them to in the program.

In the past the guys that sold us a machine did some training for us with probe routines, macros, etc. I would suggest that route first if you have limited knowledge of it. Otherwise I would ask specifics and you would be more likely to get a helpful response.