Realistic limits for mastercam

[jaydenn]

I'm really struggling here the last couple days...

 

I'm trying to put together a preliminary program to make a "forging" out of a billet of steel.

The billet is around 8000lbs of 300M steel.

At this point, I'm merely trying to get a reasonable cycle time for quotation.

 

Timelines are tight and I really, really need automatic toolpathing to work in this scenario.

So far the results are grim.

The stock recognition is beyond terrible.

The air cutting is enormous.

The cut path generation time is unbearable. 15 minutes for a general opti-crap and 45min if I attempt to use a "stock model".

 

After much,much pain, my first tool will cut for around 40 hours.

 

I can't imagine trying to make a real program for this part in mastercam.

 

I guess my question is what is the practical limit for mastercams usefulness? Would you ever consider it for ultra-complex aerospace work? Am I kidding myself to think opti path will work on a massive forged component?

 

J

 

[Guest]

I'm really struggling here the last couple days...

 

I'm trying to put together a preliminary program to make a "forging" out of a billet of steel.

The billet is around 8000lbs of 300M steel.

At this point, I'm merely trying to get a reasonable cycle time for quotation.

 

Timelines are tight and I really, really need automatic toolpathing to work in this scenario.

So far the results are grim.

The stock recognition is beyond terrible.

The air cutting is enormous.

The cut path generation time is unbearable. 15 minutes for a general opti-crap and 45min if I attempt to use a "stock model".

 

After much,much pain, my first tool will cut for around 40 hours.

 

I can't imagine trying to make a real program for this part in mastercam.

 

I guess my question is what is the practical limit for mastercams usefulness? Would you ever consider it for ultra-complex aerospace work? Am I kidding myself to think opti path will work on a massive forged component?

 

J

 

You're struggling....

 

So would people consider it for complex Aerospace work?....it's used everyday!

 

The problem isn't Mastercam...it's between the keyboard and monitor....

 

Seriously, get with your reseller, someone that should know how the software works......and get some training.

[jeff]

We need more info if you want an accurate answer. Most people don't use the ATP feature, although some do.

Is your reseller able to help you out with the program? (if you can't share it here). 

 

"ultra complex" is relative, there are people here doing some pretty complex parts, but I am not one of them. 

[Guest]

We need more info if you want an accurate answer. Most people don't use the ATP feature, although some do.

Is your reseller able to help you out with the program? (if you can't share it here). 

 

"ultra complex" is relative, there are people here doing some pretty complex parts, but I am not one of them. 

 

ATP is "generally" only useful in Router because of the way parts can be setup on levels...and recurrent toolpath strategies

 

 

ATP (Automatic Toolpathing) overview

Automatic Toolpathing (ATP) is a Mastercam NET-Hook that automates the process of assigning toolpaths to geometry for recurrent parts using a machining strategy. A machining strategy contains one or more level names that are associated or mapped to Mastercam operations. ATP automatically scans all named levels in the current drawing and when a level in the drawing matches a level in the active strategy, the applicable operation is applied to all geometry on that level. The process continues until all named levels have been processed for the drawing.

Once a strategy is created, you simply choose a cutlist or files to be processed. ATP then automatically batch processes each individual part file using the chosen strategy. It then nests the toolpaths and posts the results to one or more NC files (depending on the number of sheets created).

ATP is especially useful in large projects with many different pieces of geometry. With a modest amount of setup, you can save valuable time by letting the software automatically find, chain, and toolpath all the elements of a job.

You can start ATP by choosing Run Add-In on the Home tab, and then locating and opening the ATP NET-Hook file (XATP.dll). You can also access ATP through the Machine tab.

Note: Be sure to run ATP in the same product that you used to create the part files. Part files created using a Router machine definition must be processed by ATP with a Router machine definition. If you try to process them with a Mill machine definition, errors will result.

[crazy^millman]

What system are you programming this with is my 1st question? What are the hardware specs on the Tower you are attempting this with? We took 7200 lbs of Aluminum and turned it into 350lb with over 32000 surfaces and 2000 operations. Had to be broken into several files with 3 different people working on it, but we had over 1200 hours on that project and it spanned 3 months.

 

Automagic or Automatic toolpathing? Do you expect to grab the whole solid and have Mastercam figure out the best tools and approach or just having problems with the toolpaths you are using? Opti-Path can do it, but I would use only surfaces and I would have a line of sight model to program to. No need to have everything picked when processing the toolpaths. Dumb the solid down before making the model you are going to rough with. No need for all the holes or other things. Yes it takes some work upfront, but when you can process a toolpath in 25% of the time you then realize that is time well spent. Use only what you need and it will process much faster than telling it to process the whole solid. Really comes down to how much work do you expect to have to do programming the part. I keep hearing how what people call high CAM have the same struggles. People use the same methods and techniques I am talking about here, but if you do it in a high need CAM then is okay, but if you need to go about it the same way in Mastercam then it is not.

 

Have a Roughing Model file. Then have a Semi Finishing Model and then have a a few Finishing model files if you see you start getting to high in the operations count. 500 to 600 operations about the limit and around 1GB to 1.2 GB.

 

End of the day really going to come down to the computer you are using. Here is what I would recommend as a start and would go up from here with no reservations.

 

Xeon with at least 2.7 GHZ and 12 cores or I-7 the best highest GHZ you can get. Might even think about parallel processors.

 

Quardo 5000M or greater card 24GB cards now.

 

(2) 1TB SSD set up to run RAID 1

 

64GB of memory or greater.

 

3D Mouse.

 

2 (27") monitors minimal

 

Best of luck and you are more than welcome to reach out to our group and we can go over some things to help you start headed in the right direction.

[jaydenn]

I know, I know.... It's always the 200lb gorilla pounding on the keyboard and never the tool to blame.

 

At this point in the process, I only have like 3 toolpaths in my file,so resources are not really taxed yet. When I generate a toolpath, the CPU(its a xeon 2.4) never goes above 10% usage. Lot's of available ram, lot's of processor headroom. It's just really, really slow.

Am I wrong to think that I should see the CPU reporting more usage if that's really an issue?

 

I might have to create some separate models for each level of roughing. You're right; a few hours spent now can save a ton of headache.

 

I should also clarify that I did not mean "ATP" in the sense y'all thought I did. That's my mistake for not being clear.

I was referring to the "opti-path" strategies for "automatic" bulk roughing. 

 

My real constraint is time. I have very, very little time to pull this off and my initial efforts are not giving me a warm fuzzy feeling!

 

J

[jlw™]

Jaydenn, an 8,000lb part would be a small part for me.  Can you send me a file and let me help you?

[gcode]

My personal best...

a large HY-100 casting for a submarine

It weighed 38,000 pounds when it got here

and 24,000 pounds when it left ..  3 mill operations and 2 lathe operations

 

I have a part on my list that weighs 97,000 coming in and will be about 74, 000 pounds when it leaves

 

I have no doubt that Mastercam can do your part,

Its not going to let you "window it and cut it" like CAM salesmen have been promising

for 25 years, but it can do it

[crazy^millman]

I know, I know.... It's always the 200lb gorilla pounding on the keyboard and never the tool to blame.

 

At this point in the process, I only have like 3 toolpaths in my file,so resources are not really taxed yet. When I generate a toolpath, the CPU(its a xeon 2.4) never goes above 10% usage. Lot's of available ram, lot's of processor headroom. It's just really, really slow.

Am I wrong to think that I should see the CPU reporting more usage if that's really an issue?

 

I might have to create some separate models for each level of roughing. You're right; a few hours spent now can save a ton of headache.

 

I should also clarify that I did not mean "ATP" in the sense y'all thought I did. That's my mistake for not being clear.

I was referring to the "opti-path" strategies for "automatic" bulk roughing. 

 

My real constraint is time. I have very, very little time to pull this off and my initial efforts are not giving me a warm fuzzy feeling!

 

J

 

I have seen many poorly planned projects where enough time was not given to complete the project. End of the day it takes what it takes. If someone has got the magic get it done in seconds button then I always willing to bow out and let them have it. Sorry big parts require some thought and some leg work. Once done you move smoother, but you can be doing other work while it is processing toolpaths. Other thing is remember the toolpaths will only process at normal CPU usage. On big toolpaths I will open the Multi-Thread Manager and set the priority for the toolpath to high and have seen some nice reductions on calculations doing so. Yes you must do this every time they don't keep the settings.

 

If it were easy then everyone would do what we do. End of the day it is not. Best of luck getting through this and getting the project done.

[Colin Gilchrist]

Getting computer hardware that is optimized for all the components to work together is often an overlooked aspect of creating a "CAD/CAM" computer. Does your computer have a SSD? Also, I totally agree with Ron. Many times you must do the grunt work of prepping the model in order to realize the time savings that can be achieved. Like Ron said; strip out everything that isn't needed. Adjust your tolerances accordingly when creating Stock Models and STL Files. (.02 Tolerances, instead of using the .001 defaults) Adjust the tolerances of the Toolpath calculations to suit your roughing strategies.

 

If you are just doing an estimate, then also change your Opti-path parameters. Don't go for "fine" step-ups in your initial toolpath calculations on the initial path. Get the motion to "look right" with the basic step down settings, then only "Fine tune" the path once you've got some decent preliminary results.

 

It would depend on your company/boss, but consider hiring a guy like Ron to come in and show you how to be more efficient in your programming methods. Sometimes thinking outside the box can pay huge dividends when tackling a project that is so big in scope, or doing work you haven't been exposed to before.

 

Shameless Plug:

For Air Cuts only, the Verification Module in Vericut now includes Air Cut optimization with a basic Verification license. That alone could save you 20-30% on your run time when used properly, if the path results you are getting have a lot of entry/exit and positioning motions...

[?Mark]

Colin, I didn't know that you're with CGtech. That's great, congrats on the new gig 

[Colin Gilchrist]

Colin, I didn't know that you're with CGtech. That's great, congrats on the new gig 

 

Thanks Mark, I started yesterday and just updated my profile. I'm excited about the future..

[Fred @ Slate Industries]

Surface Rough Pocket .003 Step down

Come back in 3 days  DONE !

[Matthew Hajicek - Singularity]

In my experience estimating cycle time is very difficult without going through all the programming work.  Little changes can make a huge difference in cycle time.  Even then you only have a rough guess; you don't know for sure until you have a good part off the machine.  If you're just going for a ballpark figure why even program?  Just look up the material removal rates of the cutters you plan on using (HSMAdvisor is your friend) and do some estimating.  I once had a boss who quoted by the pound.

[jaydenn]

In my experience estimating cycle time is very difficult without going through all the programming work.  Little changes can make a huge difference in cycle time.  Even then you only have a rough guess; you don't know for sure until you have a good part off the machine.  If you're just going for a ballpark figure why even program?  Just look up the material removal rates of the cutters you plan on using (HSMAdvisor is your friend) and do some estimating.  I once had a boss who quoted by the pound.

 

You're right. And that's what I'm doing now. Making a rough program to get a realistic-ish estimate.

The reason I don't want to just ballpark, is because we're working on a very short timeline, and like you said, little changes make a huge difference.

If we commit to the work, and the estimate was wrong, we'll be dead in the water. We're a small shop with few staff.

 

I will also do a MRR calc. and compare the two.

 

This is due diligence. I need my ducks lined up neatly before I commit to something I can't complete!

 

J

[jeff]

 

If you are just doing an estimate, then also change your Opti-path parameters. Don't go for "fine" step-ups in your initial toolpath calculations on the initial path. Get the motion to "look right" with the basic step down settings, then only "Fine tune" the path once you've got some decent preliminary results.

 

 

 

^^This!

If you know it's going to be a big toolpath,make your initial values big. Large depths of cut, large stepovers, etc... just to get the motion correct. 

 

If it's taking a LONG time for you, maybe it's something as simple as a filter setting or a stepover setting that is holding you back?

I don't know how experienced you are with Mastercam, but just a thought.

[Matthew Hajicek - Singularity]

You're right. And that's what I'm doing now. Making a rough program to get a realistic-ish estimate.

The reason I don't want to just ballpark, is because we're working on a very short timeline, and like you said, little changes make a huge difference.

If we commit to the work, and the estimate was wrong, we'll be dead in the water. We're a small shop with few staff.

 

I will also do a MRR calc. and compare the two.

 

This is due diligence. I need my ducks lined up neatly before I commit to something I can't complete!

 

J

 

Sounds like you need to program the part before you commit to programming the part.  What I meant was that any "roughing in" of the program will likely give you misleading results.  If you need an accurate cycle time estimate you need to program the part for real, no shortcuts except maybe looser toolpath tolerances for long-crunching ops.

[Newbeeee™]

 

This is due diligence. I need my ducks lined up neatly before I commit to something I can't complete!

 

What's your gut feel?

In my experience, you should be able to look at the part and (relatively) comfortably think 'yup, i'll 'gram it in x hours, fixturing will take y, and run time will be z'.

If you can't do that, then you shouldn't be estimating the part.

Mastercam will do anything that you tell it to do.

[mikenaturalice]

You're right. And that's what I'm doing now. Making a rough program to get a realistic-ish estimate.

The reason I don't want to just ballpark, is because we're working on a very short timeline, and like you said, little changes make a huge difference.

If we commit to the work, and the estimate was wrong, we'll be dead in the water. We're a small shop with few staff.

 

I will also do a MRR calc. and compare the two.

 

This is due diligence. I need my ducks lined up neatly before I commit to something I can't complete!

 

J

 

You have said that ^ every time. Maybe that is the problem? Not trying to be a smart arse, but sometimes it just TAKES TIME. There are no shortcuts to be had. Sounds like you just plain don't have time to do it, so pass. Why kill yourself and find out your bid is too high and it goes somewhere else anyways!?

 

I would also like to add a random thought sort of. When I use Inventor HSM I think how freaking great the 3d adaptive is! I mean it just wokrs most of the time, no fiddling around. But then I try to do some tweaks and NOPE, not happening. 2d contour, works fine if you have a simple shape, want to fine tune it NOPE, not happening.  My point being, all softwares have pros and cons, and I have found after using several, Mastercam is #1 in my book. Yeah, sometimes I have to 'dial in' a toolpath more than I would like, but ya know what, Mastercam will let you do it, and its not really complicated, it just TAKES SOME TIME. So to second what most others have said, you are going to have to do some work. There is no magic "program this part" button, no matter what software you use or the sales guys tell you.