CAMplete Vs ICAM Vs Vericut

[Greg Williams]

Hi All,

 

Has anybody dome any back to back testing of these 3 products? I am pretty familar with Vericut so I know what is good and bad about that, just wondering how the other 2 stack up as they have the added function of posting code from CLS/NCI files

[Guest]

You're going to have to spell out the comparison. As you mentioned ICAM and CAMplete both do something Vericut does not; Post Process.

[Mick]

I'd be interested in your comparisons Greg. I have someone local wanting g-code verification with some other features, and while they are keen for Vericut, for a reason not mentioned here, they need to look elsewhere at other products.

[crazy^millman]

ICAM is a good product not limited to machine, control or cross platform CAM integration. It is a Cradle to Grave Product for a company.

 

Vericut is a great product with a lot of functionality and tools, but does not do the Post. Cross platform CAM, but again they do not make their own PP and are slaves to someone else to make a good code for the machine that has been tested and proven out, before they can finish giving a customer the complete simulation product.

 

CAMPlete is a great product, but it is limited to machines and controls they support. They also support cross platform CAM integration. Cradle to Grave as well, but again limited to certain machines and controls.

[Guest MTB Technical Services]

PM me

 

Watch the video in HD with Full screen and you can actually read the NC code.

 

 

[Guest]

Should CAMplete not meet a need, I'd go with ICAM in a heartbeat over Vericut. They've got some nice tools no doubt, but... I'd still rather go with the option that contains a post.

 

JM2CFWIW

[Greg Williams]

Thanks for the replies. The parts are small and very accurate 4 axis gear like parts about diameter 20mm x 100mm. The machine will be Matsuura LF-160, I am very familar with Vericut and I have a trial license of CAMplete, so I knida only know enough about CAMplete to be dangerous at this point. I know nothing about ICAM.

 

I really only need the following features,

 

- machine simulation and interferance checking

- Verification of Gcode

- A stock compare (Gouge/Excess) function against the original part (Camplete does not appear to do this, Vericut does)

- The package needs to work to a high tolerance, Gouge checking to 0.025mm is not good enough

- The package needs to work at a reasonable speed with a tight tolerance, I can accept 20-30 mins to verify a part but that is about it, at the moment CAMplete seems to struggle with this.

- Posting is not so important, as I can create posts

- I don't need any path optimization tools

[Guest]

I really only need the following features,

Let's go through the list...

- machine simulation and interferance checking

CAMplete TruePath - check

 

- Verification of Gcode

CAMplete TruePath - check

 

- A stock compare (Gouge/Excess) function against the origianl part (Camplete does not appear to do this, Vericut does)

CAMplete TruePath will gouge check against the original part. Stock compare... do you mean like getting a report at the end of the session that states "your stock is .1mm³ on the plus side compared to the model" or something similar to this?

 

- The package needs to work to a high tolerance, Gouge cehecking to 0.02mm is not good enough

That's a tall order. STL files are only good to .025mm and honestly, that's pushing it.

 

- The package needs to work at a reasonable speed, I can accept 20-30 mins to verify a part but that is about it, at the moment CAMplete seems to struggle with this.

Are you talking about running a collision check in that set period of time or are you talking material removal simulation?

 

PM me. I'd like to take a look at your project if possible to see what my rig will do with your parts... if you can of course.

[Joe788]

Thanks for the replies. The parts are small and very accurate 4 axis gear like parts about diameter 20mm x 100mm. The machine will be Matsuura LF-160, I am very familar with Vericut and I have a trial license of CAMplete, so I knida only know enough about CAMplete to be dangerous at this point. I know nothing about ICAM.

 

I really only need the following features,

 

- machine simulation and interferance checking

- Verification of Gcode

- A stock compare (Gouge/Excess) function against the original part (Camplete does not appear to do this, Vericut does)

- The package needs to work to a high tolerance, Gouge checking to 0.025mm is not good enough

- The package needs to work at a reasonable speed with a tight tolerance, I can accept 20-30 mins to verify a part but that is about it, at the moment CAMplete seems to struggle with this.

- Posting is not so important, as I can create posts

- I don't need any path optimization tools

 

Might as well throw NCSimul in the mix as long as you're looking at all comers.

[#Rekd™]

Siemens NX has their own machine simulation.

 

 

More info here.

 

http://community.plm...ation/ba-p/3271

[Greg Williams]

Foghorn, Cool name by the way,

 

CAMplete TruePath will gouge check against the original part. Stock compare... do you mean like getting a report at the end of the session that states "your stock is .1mm³ on the plus side compared to the model" or something similar to this?

 

Well sort off, I am looking for something similar to Vericut Autodiff, Basically have any fetures been missed, I know this can be dome in Mastercam, Compare to STL, but not all code will come from Mastercam.

 

That's a tall order. STL files are only good to .025mm and honestly, that's pushing it

Yep I agree

 

 

Are you talking about running a collision check in that set period of time or are you talking material removal simulation?

Yes material removal

 

I will need to hack up a test file, The parts I have cannot be shared

 

Joe, good Idea I will give Ncsimul a look as well

.

[Watcher]

I think one important point that CAMplete offers and others don´t is the ability to manipulate the tool vectors and change the posted code. Perhaps ICAM can do it but VERICUT for sure won´t.

[MIL-TFP-41]

I think one important point that CAMplete offers and others don´t is the ability to manipulate the tool vectors and change the posted code. Perhaps ICAM can do it but VERICUT for sure won´t.

Thats one thing I don't ever do in Camplete for a finished product. I want my CAM file to match what is in Camplete as close as possible. Now i have done it in Camplete for testing purposes, then gone back to the cam system to match what I did. Same with re-ordering operations...it can be done in Camplete, tho, once again, I like the CAM file to match what the output is going to be.

 

Now I am on a older version of Camplete, it does simulate material removal, albeit at a slow speed. There is no "autodiff" function like vericut (maybe new versions, but not the one I am on). It does gouge check against the part. The posting is flawless and things like feedrate leveling and toolpath linerization (if you aren't using RTCP) are awesome tools.

[crazy^millman]

Yes ICAM does somethings with Smartpath to controls some things.

[Guest MTB Technical Services]

- The package needs to work to a high tolerance, Gouge checking to 0.025mm is not good enough

 

Good luck with that.

To get close to 10 micron resolution AND come within your 20-30 minute limit is impossible in the practical sense.

As a practical matter, I don't know that ANY system can actually do it.

They all rely on tessellation in one form or another.

 

- The package needs to work at a reasonable speed with a tight tolerance, I can accept 20-30 mins to verify a part but that is about it, at the moment CAMplete seems to struggle with this.

 

If you don't need to simulate the material removal process, you can speed up the gouge check, overtravel and interference checks.

 

I would suggest you investigate if ANY of the systems can actually meet the tolerance requirements you are specifying.

If they can then, I would say you need to toss out your time limit.

[Guest MTB Technical Services]

Yes ICAM does somethings with Smartpath to controls some things.

 

Yep

It's a unique approach.

It analyzes the kinematics of the the machine environment including part and fixture components.

and re-processes all approaches and retracts.

 

http://www.icam.com/...ATH_Product.php

 

[Guest]

Once they went to x64 support, things got a LOT faster. I'm on V2012 Build 569, even the material removal. I never use that feature. I always use Mastercam Verify, then just collision check in CAMplete. Working that way hasn't failed me yet.

[crazy^millman]

Thanks for the replies. The parts are small and very accurate 4 axis gear like parts about diameter 20mm x 100mm. The machine will be Matsuura LF-160, I am very familar with Vericut and I have a trial license of CAMplete, so I knida only know enough about CAMplete to be dangerous at this point. I know nothing about ICAM.

 

I really only need the following features,

 

- machine simulation and interferance checking

- Verification of Gcode

- A stock compare (Gouge/Excess) function against the original part (Camplete does not appear to do this, Vericut does)

- The package needs to work to a high tolerance, Gouge checking to 0.025mm is not good enough

- The package needs to work at a reasonable speed with a tight tolerance, I can accept 20-30 mins to verify a part but that is about it, at the moment CAMplete seems to struggle with this.

- Posting is not so important, as I can create posts

- I don't need any path optimization tools

 

Greg sounds like you need to use an old school process. Take all your models and blow them up 100 or 1000 times bigger than what you need to make. When you go to post the programs you will use a scale factor and have it give you all the output with the decimal moved over to the correct place. Gives you the ability you are looking for and also would be surprised how much better the part will come out scaling them up and making program for them at that size verses trying to force most system to do things it was really never designed to do.

[#Rekd™]

Not sure about the other programs but I had an unexpected move (Verified in Vericut) on a 3 axis machine. After a lot of investigation that part of the problem was how the actual control on the machine handled the gcode. The min arc radius in Mastercam was too small so the control did something else.

 

CGTech sent a guy down and ran a test part and took a video. They also added new updates to their software for this problem.

 

 

[Watcher]

 

Good luck with that.

To get close to 10 micron resolution AND come within your 20-20 minute limit is impossible in the practical sense.

As a practical matter, I don't know that ANY system can actually do it.

They all rely on tessellation in one form or another.

 

True dat.

 

All them will tell you they do. In theory it is feasible, but you will have to lower your cutting tolerance (When the system allow you to set your cutting tolerance, like VERICUT) to a level that will make it painfully slow. It would be ridiculously slow. Sometimes even with large tolerances it is. All them will tessellate data. I don't think with a medium model (>100mm) it's realistic to work with such low values. Vericut will even warn you about that if you try to do so.

 

 

(Sent from my Moto X via Tapatalk)

[Greg Williams]

Good luck with that.

To get close to 10 micron resolution AND come within your 20-30 minute limit is impossible in the practical sense.

As a practical matter, I don't know that ANY system can actually do it.

They all rely on tessellation in one form or another.

 

Yes I agree. But If we never asked for the "Impossible" then man would have not walked on the moon 45 years ago. My customers ask for the impossible every week and they expect to get it.

[Colin Gilchrist]

I'm reasonably sure that Vericut does not Tessellate the model. Their process for representing stock is different from most of the other players on the market. You can absolutely specify as tight a tolerance as you need. Your only limitation will be the amount of RAM that is needed.

 

When Vericut creates a block of "stock", the process it goes through is much like filling up a balloon with grains of sand. Each grain of sand is a "block" and the size of those individual blocks are based on your cut tolerance. So a tolerance of .01 mm would take about 10 times the amount of RAM as a tolerance of .1 mm. It should be no problem for Vericut to handle accurate stock models down to .001 mm, but you'll need a lot of RAM, and it will take a long time to generate the stock, let alone run the simulation. You mentioned .025 mm as your target for verification accuracy. I'd recommend running your simulations at .01 cut tolerance, and you'll have plenty of resolution to see exactly your needed results.

 

One of the really cool things about how Vericut creates and manages stock is that they are not intersecting and cutting triangles on the fly. Their process does a Boolean comparison between the cutter and the cut stock model. If the tool touches a polygonal block, it gets removed from the database. If the tool doesn't touch a block, it isn't removed. The running of the actual verification process is extremely quick as a result of how they have engineered their software, and the resolution is only limited by the performance of your computer hardware.

 

You also mentioned that the parts are small to begin with. This will greatly improve Vericut's performance, since the performance will be directly related to the total volume of the stock definition that needs to be filled with those polygonal blocks. The smaller the part, the more accurate you can make the resolution without sacrificing performance.

[Matthew Hajicek - Singularity]

So a tolerance of .01 mm would take about 10 times the amount of RAM as a tolerance of .1 mm.

 

Since it's volumetric I would expect 10X tighter tolerance to take 1000X more RAM, unless they have some pretty clever tricks.

[Guest MTB Technical Services]

Vericut's own website and documentation states it uses a Polygon model.

Polygon= Tessellation.

 

They may do something unique with regards to boolean removal processing but it's still

a polygon based approximation, albeit 3D, since they aren't using the inherent BREP topology.

[crazy^millman]

Greg, did you consider my suggestion? All and I mean all do it with STL, POLYGON, TESSELLATION all means the same thing just different ways to word it to throw off the consumer. Please consider like I suggested go in the opposite direction by going larger. Yes all your tools will have to be larger scaled according to the scale you find works to achieve what you are after. We would take and make 1000 time bigger prints out of shapes we need to make and then take the optical comparator and compare them on that to shapes we would grind where were trying to hold .0001" and we would hit it every time. Same concept here you are trying to hit a tolerance not easily done using today's developed process so adjust the process for your needs and make it work for you and not what was not designed to work that way, but getting around those limitations and working to make it work for you.

 

HTH