Mastercam verification limitation

[CamMan1]

I am programming some aircraft wing components that are very large. The material size is 220.00" long by 20.00" wide by 3.00" thick. The parts finish at 218.00" long by 18.00" wide by 2.500" thick. These parts are fairly complicated. I need to be able to verify theses parts in Mastercam. I am verifying at high quality with my Tool tolerance at .005 and my STL tolerance at .005. verifying true solids not turbo. I have found a limitation in verify. If my stock is defined over 177.00" long the verification looks like crap and the STL compare looks even worse. I have tried box, soild and stl for my material. Doesn't make a difference. Has anyone else seen this isssue before and if so does anyone have a workaround.

[gcode]

so does anyone have a workaround.

 

Vericut

 

if you're running X6, you might try the new Moduleworks verification

[Namliek_J@TheeCircle™]

Has anyone else seen this isssue before and if so does anyone have a workaround.

 

The problem with verifying in MasterCAM is that it is pre-post. Just because it looks good in MasterCAM verify doesn't mean it is going to work the same exact way after the code is posted.

 

Best bet is to run some type of verification after posting such as Vericut if possible. Whats great about Vericut is that the code that I verify is the same exact code that goes into the machine.

[MLS]

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.

[Aeroguy]

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

Doesn't exactly inspire confidence in MC now does it.

[gcode]

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

Doesn't exactly inspire confidence in MC now does it.

 

 

actually , parts this big push the technology of all software/ hardware.

at my last job we did some C-17 wing skins that were 150 x 25 feet x 1.25" thick

We started them in V9 and finished up in X using nothing but Verify

 

Verify works well for most work if you've got decent hardware.

It gags on the stuff moldmakers churn out, but I can choke Vericut with some of my impeller projects too.

 

It would be interesting to know what hardware CamMan1 is running.

[CamMan1]

actually , parts this big push the technology of all software/ hardware.

at my last job we did some C-17 wing skins that were 150 x 25 feet x 1.25 thick

We did started them in V9 and finished up in X using nothing but Vefiry

 

Verify works well for most work if you've got decent hardware.

It gags on the stuff moldmakers churn out, but I can choke Vericut with some of my impeller projects too.

 

It would be interesting to know what hardware CamMan1 is running.

 

I am running a Dell Precision T7500 with dual 3.00GHZ Xeon's with 16Gigs of Ram, Dual 80Gig 10,000rpm Raptors at Raid 0 running Win7 64bit and dual Dell 30 inch monitors. Even with this Rig doing 170 inches of verification is slow. I appreciate everyones input on this matter and If it were up to me we would have a seat of Vericut. Unfortunately for me I have been programming parts like this for a while with no mistakes which leads management to believe I don't need something like Vericut. For now I will have to live with doing small sections at a time. Makes it a pain though and just allows more room for error. I just decided to do a 72.00" section and it went a lot faster.

[gcode]

That's a nice machine

what video card are you running... thats the biggest issue with Verify problems

[CamMan1]

That's a nice machine

what video card are you running... thats the biggest issue with Verify problems

 

Ya I forgot one of the more important ingredients. It's a Quadro FX4600 with 768Megs of Ram. I would like to step up to a FX5600 but thats not in the cards right now either.

[EX-wccprogrammer]

Send me the MCX file, I will run it through Vericut for you. I have a pretty top end system that has handled everything that I have tried. Could be fun...

 

If you can't part with the MCX, make a STL of the part model, and send with the NC file.

 

I am running a Quadro 6000 with 6GB on board. Plus 24GB of RAM.

[gcode]

Ya I forgot one of the more important ingredients. It's a Quadro FX4600 with 768Megs of Ram. I would like to step up to a FX5600 but thats not in the cards right now either.

 

 

If you do upgrade your card, go with the new Quadro 4000 or 50000

the performance is double the old FX series and the prices are better too.

You can get a Quadro 4000 from Amazon for about $750.

 

I've got an FX4800 in this machine and some of the new machines here have the Quadro 4000

The 4000 is a much better card

[mayday]

I am running a Quadro 6000 with 6GB on board. Plus 24GB of RAM

 

wow,

whats the 1/4 mile time on that

[EX-wccprogrammer]

wow,

whats the 1/4 mile time on that

I ran the benchmark file a while back, and I think it ran about 1:05 at that time.

 

I think I will run it again, and post results.

[gcode]

I noticed something today.

You should use your filters where ever possible.

I just did a very simple high speed 2d Area toolpath.

Verify was hanging in the transition between passes taking over a minute to get through

the toolpath

The posted code was 3775 bytes of point to point data

I turned the filter on ( arcs in XY only)

and the toolpath runs through Verify in under 10 secs and the posted code is

1447 bytes of arcs and point to point

 

 

 

[TheePres]

Probably not a spar at 20" ....

 

I use Mastercam to verify wing skin toolpaths, but only to look for major issues, crashes, wrong tool (caused by MC holder issue), resolution are set at .020 for both tool and STL.

 

Final NC code gets verified in VeriCut.

[Mick]

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

Doesn't exactly inspire confidence in MC now does it.

 

Verify works fine for standard types of cutting. He is talking extremes here, with a lot of data...

[gcode]

Aeroguy is a hater.. if its Mastercam.. it sux

[BenK]

Aeroguy is a hater.. if its Mastercam.. it sux

 

The only question I have for someone like him is if you hate the software that much why don't you use something else and stop complaining? The grass is always greener right!?!

[gcode]

The only question I have for someone like him is if you hate the software that much why don't you use something else and stop complaining? The grass is always greener right!?!

 

 

He's moved from Featurecam to Mastercam for reasons beyond his control and he's not happy

 

He'll come around.. or not..

I had similar feelings when I switched from Tek-Soft to V8 Mastercam a lifetime ago

[BenK]

Ya I know I just think there is better ways to handle it. If he put the effort into learning why/how Mastercam worked instead of complaining about it he could really solve some of his issues. But no matter what we always have the control, even if that means switching jobs or buying our own software package.

[Aeroguy]

Aeroguy is a hater.. if its Mastercam.. it sux

 

Yeah, it does actually. I've been doing this for a long time, with a few other CAD/CAM packages, and never scrapped a part due to toolpath issues. Offset and set-up errors yes, but not toolpths that I created. And yet, no matter how many times I backplot, verify or overlay a cut stock stl file, I still end up with crap more often than not, due to either bad code, or just a generally crumby toolpath.

Now I'm a pretty smart guy on most days, but Mastercam is driving me friggin insane with it's limitations, constant chaining requirements, outdated alogrithims, usless spline machining, and generally unfriendly surfacing toolpaths. For a simple part, Mastercam is just fine, but for something complicated,..... forget it.

 

Believe me, if there were other jobs out there that didn't use Mastercam then I would be there. But thats not the way it goes in reality, and I still need a paycheck.

[robk]

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

Doesn't exactly inspire confidence in MC now does it.

 

Using the same software to check itself is pointless IMO. Our customers expect/require us to use differnt software for program verification than the software that was used for toolpath/posting creation. Vericut all the way when it comes to verifying our programs.

[within a thou]

Yeah, it does actually. I've been doing this for a long time, with a few other CAD/CAM packages, and never scrapped a part due to toolpath issues. Offset and set-up errors yes, but not toolpths that I created.

 

Wish I could say I was perfect in programming like you but I have made mistakes could these mistakes been caught if things inside off mcam worked better. Definitely. But mcam is the jack of all trades master of none. Ask my reseller he will tell you how vocal I have beeen to him about how every release since 9 this software is going backwards. CNC Software is trying to cater to a new market with all these new toolpaths but they dont bother fix the simple things that their long standing customers (who have paid for their houses, cars, PBR habits and their retirements funds) run into constantly. I run Verify with stl compare. and every part comes shows up on the screen as a friggen rainbow even though it is programmed correctly. Ask my reseller (who has been great help and done a great job at in house, who are not the problem CNC Software is) and all he says is cnc fixed it in X6.

 

CNC Software needs to take a second to look in the mirror, address the long standing issues they have ignored. fix the simple things before they keep trying to continue to bait people in with the latest version of X and all its new bells and whistles.

 

I started using this software in V8 and V9 was a improvement but as soon as they turned it windows based in X this has gone down hill. When it went to X so many people were "ya its different because it isnt dos based any more, give it time you will see it will be better" and I listened and patiently sat back and waited to see what would come. It has gone so far downhill that all the people who did instal X6 who are now trying to explain to their bosses what happened with all the scrap parts and crashed machines on the floor thankfully I didnt upgrade to the new version. Read through any of the posts on here and you can see how well it is worring for them. Hell they even pulled the upgrade because it is that bad.

 

To sum this up you sold us a car with square wheels and as long as we are willing to keep putting the presents under your kids christmas tree you will give us free oil changes on a car that wont take us where we want to go. It must be nice to work for a company that can xxxx off their customers the way you do and still have them come back. How did you do it? I would like to be able to take dump in a bag and convince someone to spends thousands on it. What venue o you use for marketing other then "keep giving us money and in time this bag of xxxx will start to smell better"

 

 

 

[GoetzInd]

For a simple part, Mastercam is just fine, but for something complicated,..... forget it.

 

I don't know man. I've seen a lot of cool stuff made with mcam. By both me and other users. My beef with the software is that it takes some time to figure out all little quarks. But having said that, iv'e never seen a part that mc couldn't do. Within a thou makes a good point on fixing some little stuff that has been lingering for a long time. It seems like the software is always 95% functional. And instead of fixing the other 5 %, they put more on there plate.

 

Mike

[Guest]

This is something I posted in the Beta forum sometime back, I am not giving up any secrets as I am only discussing my issue at that time, when I was still making chips for living.

 

Yes I really used to do that

 

You guys are not alone on this

 

 

Posted 19 January 2007 - 03:28 PM

I tried the one path at a time today after doing 10 at a time ran into a wall. I could no longer save and import the .stl file, It would verify and then when I saved I would get a 0k .lwn, which is causing problems, so even doing one path at at time it doesn't matter, I can't save and continue on.

 

The .stl at that point is 135 megs, I am wondering if there is a 128 meg cap on the .lwn file.

 

Either way it doesn't matter I am at a dead end on verifying this job. I am going to have to rely on backplot and pray for no gouges. 'frgiin 55k part, 3 of them and I have to resort to this.

 

I am not happy at this moment in time. I had to put the job down today because I finally got to frustarted to continue trying to play with it.