What's new
What's new

CAMworks. Beware!!!

martin_05

Hot Rolled
Joined
Mar 11, 2009
Location
Valencia, CA, USA
CAMworks broke a couple of my small endmills through an amazingly dumb bug in the program: If you specify a spiral entry into a pocket to be machined and the pocket has islands that prevent one of the spiral entries from happening...CAMworks seems to change the spiral feed into a RAPID straight-down plunge into the material. There's a faint indication of this but no warning message of any kind.

I got lucky, I went through a couple of 1/8in endmills before I discovered the issue. Had the pocket required a larger tool this would have caused a catastrophic crash and probably damage to my VF-3SS.

The support guy pointed me to these little red lines that indicate that there's a rapid move in the toolpath. My reply to this was (among other things): In the first place, under what conditions is it even acceptable to plunge into stock at full rapid rates? In other words, it's a dumb and dangerous move whether a little red line tries to tell you it is there or not....because you can miss the little red line.

It'd be interesting to hear from others who might be using CAMworks. So far I am grossly underwhelmed and almost sorry that I went with this package.

How do others like SolidCAM, FeatureCAM, MasterCAM and whatever-CAM compare in terms of reality vs. demos vs. marketing?

I am using it with SolidWorks, BTW, so integration is desirable (but not at the expense of potentially damaging the machine).

Thanks,

-Martin
 
CAMWorks, beware.

I replaced CAMworks with HSMWorks and have been super happy.

I had full 5-axis CAMworks.

HSMWorks as a product and as a relationship with the company have both been exceptional.

I tried out trials of several other CAM products before going with HSMWorks.

Keith Clausen is the reseller from whom I purchased HSMWorks. He's at:

Keith Clausen
888.275.2195 Office
888.275.9526 Fax
604.220.2696 Mobile

[email protected]
[email protected]
 
I like to run a simulation before cutting metal. I have the option to turn on the rapid gouge warning for checking for situations like that. There have been a few times when I became complacent, but should have run the simulation first. I'd recommend that you always run a simulation, if your software has a bullet proof simulator that doesn't hide the truth of the path from you. If you cannot trust the simulation, then I don't know what good it is.

I run OneCNC.
 
I like to run a simulation before cutting metal. I have the option to turn on the rapid gouge warning for checking for situations like that. There have been a few times when I became complacent, but should have run the simulation first. I'd recommend that you always run a simulation, if your software has a bullet proof simulator that doesn't hide the truth of the path from you. If you cannot trust the simulation, then I don't know what good it is.

I run OneCNC.

I am not sure that CAMworks has the equivalent of "rapid gouge warning". I'll check. The simulator doesn't seem to be bulletproof because it doesn't really seem to simulate what the machine is doing. It's weird.

-Martin
 
How do others like SolidCAM, FeatureCAM, MasterCAM and whatever-CAM compare in terms of reality vs. demos vs. marketing?

The "Helical Entry" screen in MasterCAM V9 has two options that toggle:

"If ramp does not fit:

O Skip
O Plunge"

The plunge is at whatever feed rate has been defined for the tool. I see no reason why this hasn't been carried forward to newer versions.

I agree, allowing a rapid below the "top of stock" as defined as the default is a recipe for disaster.

Dennis
 
Yes the spiral-in option on 2.5 axis does have a serious problem and must be watched. When running the simulation make sure you have Tool pause on collision and shank pause on collision. If these buttons are X'ed out you will not see the crash and if it is on cut you might not see the crash.
 
hi,
I purchased CAMWorks 2009 ( www.camworks.com) 3 months back after evaluating HSM , FeatureCAM and MasterCAM. I found it to be one of the fastest and easiest cam system on SW for 2.5 axis milling. Specially the knowledge based machining facility.

I am pretty happy with the machined parts produced after running the simulations.

I will be cautious for the scenerio you mentioned.

Joe
 
CAMworks seems to change the spiral feed into a RAPID straight-down plunge into the material.

Now that is just inexcusable. I take that as an indicator of the thought put into the program, as well as the programmers understanding of the end use. Camworks is hereby officially off my shopping list. Thanks for the info!

And why o'why O'mighty Why: If a program can show that it is rapid plunging into material, and thus knows that, why on earth can't it correct for that itself?

Does anyone have experiences with Esprit?

Is there a list of programs that actually, independantly and accurately simulates their own output? (Instead of just showing a graphic of what the programs programming section think it has done?)
 
How do others like SolidCAM, FeatureCAM, MasterCAM and whatever-CAM compare in terms of reality vs. demos vs. marketing?
-Martin

Actually I think at least Featurecam is under-marketed as far as capability and configurability.
Take your case from this thread.
The post has separate moves for rapiding to Z-Rapid plane and Z-clearance plane. One can override any and all moves below the Z-rapid plane in the post by defining a G01 F"custom", meaning below the Z-rapid plane you will never get a G00 move, rather a feed move with a feedrate specifically defined by you.
Rapiding into a stock below Z-clearance plane never happens unless you've defined the boundary to be inside the stock. IOW if you've told FC that there is no stock, then it will plunge there. It however will never attempt to rapid plunge in a pocket, as it is always contained within a stock.
Now, as far as the overrides, the plunge feedrate depends on 2 separate variables. One is the actual feedrate of the feature, and you can tell FC to use a % of the feedrate during plunge. Example, if your feature's feedrate is 10IPM and you've set this to say 50%, then the plunge is done at 5IPM. That is for straight or ramed plunges. You can even further specify a feedrate for straight plunges. IN cases where a spiral or linear ramped plunge is not possible, it will make a straight 90deg plunge, which is now governed by another setting. SO, if you've set the first register to 50% and set the second to 50%, then you'll get ramped feedrates @ 5IPM, but if a ramp is not possible then you get a straight plunge @ 2.5IPM. All without you ever having to manually monkeying with it.
What is even more configurable, you can set your own customized parameters based on machine, material or whatever and just recall them for each part.
Simulation is I'd say 99% accurate. The 1% is a bug and is something that happens in one case only where the outputted code is not the same as the simulation, and is due to multiple identical features being roughed and finished with the same tool. IN all other cases the sim matches the code exactly.


SO, since you've asked.....
 
Can't you view your cut in Camworks before you post and send to your machine? If I do that with GibbsCam I would have avoided a crash and saved the endmills.

I feel confident enough with GibbsCam that after I run the graphic and post process the part, I can load the program, set the tools, push the go button and walk away. I do that with great regularity on some very expensive parts and it has never let me down.

GibbsCam might be a small player in the industry, but I have learned in the 10 years I have been using it that I can do the same thing that a Mastercam user can do, but with about 2/3 fewer key strokes.
 
I can’t believe this is intentional, sounds like a bug or a setting is overlooked.
Surfcam will change automatically to ramp if the helical plunge will not fit.
I can’t remember the last time I ran a simulation in Surfcam before cutting , maybe I did when I first got Surfcam 14 years ago. Surfcam is not perfect by any means but I run dozens of new programs everyday without any serious problems. If there is a problem it is almost always driver error and rarely the programs fault. Our molds require hundreds of programs sometimes with many merged into one. Lead time doesn’t give us the time to run simulations. You have to have trust in the software you use.
 
hi,
I purchased CAMWorks 2009 ( www.camworks.com) 3 months back after evaluating HSM , FeatureCAM and MasterCAM. I found it to be one of the fastest and easiest cam system on SW for 2.5 axis milling. Specially the knowledge based machining facility.

I am pretty happy with the machined parts produced after running the simulations.

I will be cautious for the scenerio you mentioned.

Joe

Forgive me for saying this...and I could be completely off-base here...but your post sure sounds and looks like a plant from CAMworks. Why in the world would you go out of your way to give out the URL, etc.

If you are a legitimate end-user/machinist, please forgive me for saying the above.

Here's a list of the bugs/issues I just submitted to my VAR:

WARNING: This is the Internet and crazy people can post all sorts of crazy and contrived stuff. In reading throught the list below you are instructed to assume that I am an absolute idiot and that I do not know what I am talking about. Do not form any opinion whatsoever based on what you are about to read without first confirming through your own testing procedures the veracity of the claims. For all you know I can't put a round peg in an oversized round hole.


1- It is possible to generate toolpaths that contain a rapid plunge into the stock. This can cause tool and machine breakage as well as harm to the operator. This is a high-priority bug that needs fixing immediately. What I've been able to ascertain is that it can happen if one sets up a machined pocket feature with some islands and chooses spiral entry. If there isn't enough room for the spiral entry it seems that CAMworks decides that it is OK to, instead, plunge into the stock at full rapid travel speeds. The way I ran across this was to define such a pocket with normal plunge entry and then decide that I wanted to play with using a spiral entry instead. The toolpath was fine before, so I just selected spiral entry and moved on. How could I imagine that CAMworks would think that it was OK to rapid into the stock.

2- Rebuilding a modified part when using any AFR features makes a mess out of the part. Features don't move to follow the changes.

3- Producing a separate g-code program out of each of multiple setups results in programs that are named exactly the same. There is no option to name each program separately. Manual editing is required to fix it. This is not acceptable as this is a place where one could make catastrophic errors.

4- There doesn't seem to be a way to require and use different work offsets for different setups. A part with three setups, for example, will output a single CAM file with all the setups mashed together and all referenced to the G54 work offset. If it so happens that the second and third setups require re-clamping or re-fixturing the part and using a different (G55, G56, etc.) work offset CAMworks does not seem to offer a way to deal with this at all. The single monolithic CAM file will destroy the workpiece as it starts machining past the first setup without so much as a warning.

5- No option to automatically output each setup as a separate CAM file with sequential CAM program names and stipulated work offsets (G54, G55, etc).
One has to output each setup by hand and then go edit each file to change the program name and do a search and replace to change G54 to whatever might be needed. As program iterations are output this is an easy place to make a mistake by forgetting to make a change and crash the machine or ruin a part.

6- The provided Haas post creates a very unsafe program start condition. No tool or work offset is loaded and the machine is sent Z0 without missing a beat. This can result in a serious machine crash, a broken tool, broken or mangled workholding equipment or all of the above.

7- The provided Haas post does not end programs correctly. The tool is left at the workpiece and M30 is not issued.

8- Cutter compensation toolpaths cause errors at the milling machine when it is not taken into account that the tool might probe to be slightly larger than the nominal tool diameter used in CAMworks. The simulator does not have any provisions for this at all.

9- Contour operation. When enabling chamfer mode it is necessary to enable cutter compensation at the CAM level. This should be automatic. Because it isn't it is possible to bury a countersinking tool into the stock.

10- Exit from cutter compensation mode in a small pocket causes gouge when the machine moves from cutter comp to normal. The CAM code is wrong.

11- Haas post has coolant turn ON too late. In some cases the tool is already cutting before coolant comes on.

12- Coolant is being turned OFF too late, which causes coolant to be sprayed all over the toolholder's taper area during a toolchange.

13- Spindle RPM too high during retract. There is no reason to keep the spindle RPM at full tilt while doing a slow Z retraction from the workpiece.
I've watched the spindle run at 12,000RPM while the toolpath is told to move from 0.1in above the material to 1.0in above the material at 3 inches-per-second just prior to a toolchange. Either the spindle needs to be turned off earlier or the speed reduced to 1/10th (or whatever) of the cutting RPM's.

14- Program naming and setup comments are needed on a per-setup basis and these need to be able to make it into the CAM file.

15- No way to insert a programmed pause or stop to re-clamp or clean the workpiece.

16- Changing the origin for a setup does not produce a dialog or anything remotely similar to allow the user to set a different work offset (G54, G55,
etc.) for that set of operations.

17- I've seen it where the specified lead-in and lead-out feedrate is not respected and a different feed rate is used.

18- I've seen a markedly different cutting rate used than what was programmed for the particular tool. For example, a G01 at 150ipm rather than the programmed 100ipm.

19- General user interface issues with such things as windows that don't store and recall their position, the inability to resize dialogs to display full information on table columns that need to be wider, the lack of ability to edit tables in place (which requires cumbersome multi-keystroke navigation (for example: editing tool crib tool feed and speed and tool numbers), the lack of tool crib sorting capabilities, and more. In general, a very sloppy and time-wasting set of lazy interface decisions that cost in productivity and create more opportunities for making mistakes.

20- Using MS Access to edit and manage the database creates lots of opportunities for making serious errors. MS Access is not designed for this sort of ad-hoc database manipulation without a fairly massive amount of code wrapped around the database to, effectively, hide MS Access. I know, I have written many database applications, some using MS Access (which is a dog) others using MS Visual Basic to manage MS Access and SQL databases (better) and yet others using C++ to manage MS Access and SQL databases (by far the best in terms of control and elegance). Using MS Access as the interface through which my users would have to inteface with a database is just about the last choice I'd ever make unless I was able to put in so much code that, effectively, MS Access is safely hidden away.


Some of the above might have fixes through making changes in the way the program is used. Others could be bugs and yet other issues could just be the "personality" of the program.

Your mileage may vary.

-Martin
 
No, Martin, I dont think youre and idiot. Some of your complaints about Camworks are founded in truth. Most, however, are due to post issues or your lack of knowledge of how to use Camworks. I understand your grief, because when I first got Camworks the posts were lousy, and I didnt get much training. Now, 4 years later, I know the tricks, have made myself good posts, and my issues are minor and infrequent. A perfect CAM program? Not even close. However, I dont post crash code, and make good parts. Invest in some training, and either work with your VAR to get a good post and-or learn to make your own posts. Im sure this isnt what you want to hear since you seem to be on a rant, but thats what I would recommend, short of ditching it althogether. Considering your hatred of Camworks, that would probably be the best idea.
 
I'm sorry Rainman, but to me any program knowingly rapid plunging into material has issues far beyond training.

Can someone else with Camworks confirm this behaviour, and thus remove Martin's level of training and his posts from the equation?
 
Im sure this isnt what you want to hear since you seem to be on a rant, but thats what I would recommend, short of ditching it althogether. Considering your hatred of Camworks, that would probably be the best idea.

Before you get too caught up in this, martin_05 is the guy who ranted that G-code should be obsoleted and replaced because he didn't understand it. Check his oldest posts.
 
I've made 1000's of programs in Surfcam and never had a problem with crazy rapids into the part or anything like that. I've had some funky 4 axis G code but that was a post problem. I rarely verify any programs. Post it and cut.
 
Knowing how to use your software, understanding your software functions, understanding what your doing and having a solid post is the key to getting good programs....doesn't matter what system your running IMO.
 
Well, on that note a couple of relevant questions are beggin' to be asked:


1: Wouldn't G-code literacy be quite essential in the selection of one's CAM software and later on in the proper modification of the post processor?

2: Assuming the rapid plunge is a software bug in CAMWorks, should it be mandatory for the owner of the package be on a maintenance contract to receive the software update with the fix?
 
Before you get too caught up in this, martin_05 is the guy who ranted that G-code should be obsoleted and replaced because he didn't understand it. Check his oldest posts.

Ahhhh, I knew that name looked familiar.

I'm not a Camworks user, but the majority of his issues sound like training issues.
 








 
Back
Top