Sign in to follow this  
essid.mh

New CAM programming algorithms

Recommended Posts

To improve the quality and the precision of the complex surfaces obtained by high speed machining and increases the service life of the CNC machine or Robots, it is necessary to integrate the manufacturing parts process parameters as a boundary condition in the calculation of the tool path: Dynamics of CNC machine axes, the parameters (numerical and kinematics) of the computer numerical control, parameters of the smoothing mode, …
I am writing to inform you that I am in the process of  developing of   new  CAM programming algorithms  approaches  allowing to integrate the principal parameters of the CNC units (smoothing mode defined in the CNC unit, anticipative setpoint mode of the CNC unit, maximum capacity of the feed-forward control, the effect of the axes inertia variation,...) in high speed machining process.

The proposed new software belongs to industry 4.0.
it allows  :
1) simulate the dynamic motion of the CNC machine taking into account the effect of all the environmental elements of the machining operation (parameter of the CNC unit, smoothing mode, NC file , tool/part interaction)
2) suggest to the programmer:
          a) improvements in the parameters defined in the cam software:  machining tolerance, smoothing tolerance, smoothing mode, ...
          b) improvements in the parameters defined in CNC unit : max axis jerk, max axis acceleration, feedforward control,.... ( the new cnc unit allow their parameters to be defined in the NC file, like sinumerik 840D).
3) create modifications at the NC file  in order to improve the machining quality
 
These algorithms can be programmed in CAM software or in post processor or in CAM simulation software
The new algorithms developed   result in a significant improvement in the accuracy and roughness of the free-form surface parts produced by high speed machining (CNC machine and robot).

The proposed project is in the form of new mathematical models for calculating the tool path integrating the drive controller parameters as a boundary condition.

In addition to these three projects, several other guidelines can be developed in this area with the aim of producing a new intelligent generation of CAM software.

Are you interested in collaborating in order to finalize the study ?

 

Best regards  

Essid Mohamed

Teacher researcher : ISET SFAX and UGPMM-ENIS ( Tunisia)

https://www.linkedin.com/in/essid-mohamed-567865177/

https://www.researchgate.net/profile/Mohamed_Essid
 
 

Share this post


Link to post
Share on other sites
2 minutes ago, essid.mh said:

Are you interested in collaborating in order to finalize the study ?

How much do I get paid to consult????🤣

  • Haha 3

Share this post


Link to post
Share on other sites

This sounds like an interesting project

Share this post


Link to post
Share on other sites
On 9/10/2021 at 2:39 PM, essid.mh said:


The proposed project is in the form of new mathematical models for calculating the tool path integrating the drive controller parameters as a boundary condition.

 

I'm 99% sure the old Yasnac Matsuuras had this....I bought a 1994 FX5 for a place I worked, and I'm sure we had the high speed tolerance codes where we specified "?" (can't remember now) and that was a 0.1mm profile tolerance, and another code for 0.05mm etc...

Share this post


Link to post
Share on other sites

 

....and a bit more trolling :lol:....purely because you state "Industry 4.0" (which for those that don't know, is the WEF's 4th Industrial revolution) :rolleyes:

Below is your post on Practical Machinist, stating you already have the algorithm to allow for optimal tolerance blah blah (attached).

 

So how does this work - how does your algorithm "tell the CAM" to allow for the following different machines?

Can you please explain?


I want to make the same part on 3x different machines.
#1 machine is a large Haas VF8 machine - a big heavy table with slow acceleration and deceleration.
#2 machine is a much smaller Haas - smaller lighter table with the next generation faster processing control and much better acc/dec.
#3 machine is a gantry design machine - the table is fixed and the axes are a bridge/portal design. So although mass of component on the table is irrelevant, unfortunately I have an issue with the X axis where there is "some" out of squareness to the Y.
This has an early Fanuc control with no lookahead processing.

 

???

Capture.JPG

Share this post


Link to post
Share on other sites
2 hours ago, Newbeeee™ said:

I'm 99% sure the old Yasnac Matsuuras had this....I bought a 1994 FX5 for a place I worked, and I'm sure we had the high speed tolerance codes where we specified "?" (can't remember now) and that was a 0.1mm profile tolerance, and another code for 0.05mm etc...

IIRc Colin was making some really neat test parts on one of the new Yasnacs, it was very precise

i forgot the exact numbers already

 

On 9/10/2021 at 8:39 AM, essid.mh said:

Are you interested in collaborating in order to finalize the study ?

I sometimes dabble in code, what language is your software written in?

Share this post


Link to post
Share on other sites

Linux CNC does this. I believe it's g64.1 pxxxx for your corner rounding tolerance

Share this post


Link to post
Share on other sites
2 hours ago, Thee Kid™ said:

IIRc Colin was making some really neat test parts on one of the new Yasnacs, it was very precise

i forgot the exact numbers already

 

I sometimes dabble in code, what language is your software written in?

(It was a Yasda machine, with a Fanuc 31iB5 Control...)

A Yasnac isn't a control that I have a lot of experience with.

I do however, know a bit about kinematics, motion control, high-speed modes, and NURBS processing. Add in the correct Dynamic Codes for Tilted Work Plane, Tool Center Point Control, and Workpiece Setting Error Correction, in addition to properly calibrated kinematic values, and the machines work like magic.

Parameter Settings and Proper Kinematic Calibration. Both of these are the real Boundary Conditions of making accurate components on any CNC Equipment. Repeating buzzwords like "Industry 4.0" won't overcome the real-world obstacles everyone encounters when they try and build software that solves a hardware problem. 

Don't get me wrong; I'm all for progress and technological innovation, but I don't hear anything new in the offer being made here. 

You can write NC code to query a lot of functions on the three major brands of control: Siemens, Heidenhain, and Fanuc. But without some kind of understanding of how all the equipment was configured and calibrated, good luck with trying to solve something in a novel way.

  • Like 4

Share this post


Link to post
Share on other sites
8 minutes ago, Colin Gilchrist said:

(It was a Yasda machine, with a Fanuc 31iB5 Control...)

A Yasnac isn't a control that I have a lot of experience with.

ah ok, yeah I got mixed up, yours was the yasda, It was someone else with the yasnac 👍

 

  • Like 1

Share this post


Link to post
Share on other sites

The thing not being shared is the testing and prove out process of this. Where are your machines located for testing? How many builders are on board to allow for the testing? What is your regression testing progress look like? What are your gauge R & R parameters for ensuring that the improvements are living up to what is being promoted? How many different parts and material have you developed and tested these on? What tool brands have you also tested the process with? What are the parameters for such? What about holders, fixtures, cutting fluids, and other things that play an important part in machining and getting the best results? How does the algorithm take those factors into account?
 

I was approached years ago by a team at a highly respected University. They wanted me to help them with their study similar to yours. There were 14 people with 20 PhD behind their names. They had 20,000 hours into their algorithm. I asked them about the same question and got a bunch of well and hum good points. Then I asked them to show what it could program. It could only do simple stuff. Anything really complex or other things they thought it would be another 20k to 40k hours to do 3 Axis then maybe another 100k in hours to get basic 5 axis parts. That was 6 years ago still not able to handle the 3 axis stuff. Know of 4 other companies trying to do the same thing all floundering so curious how your experience is going to best all of them? 
 

What are your credentials for doing this? Background and other things to make anyone think your not trolling?

  • Thanks 1
  • Like 4

Share this post


Link to post
Share on other sites

Unite the Luddites - to Hell with the 4th Industrial Revolution :hrhr:

  • Haha 1

Share this post


Link to post
Share on other sites

A fair amount of that was promised with STEP NC... back in the 1990's. And here we are 25 years later with no measurable improvement beyond CAD interoperability which would be expected regardless because of natural progression. 

Call me a skeptic if you want, I prefer REALIST. There are few if any "standards" CNC Controls adhere to beyond CE basic electronics. Until THAT changes, this is nothing more than mental and code xxxxion. 

Don't get me wrong. I like new stuff, I thrive on new stuff, and new stuff usually comes from impossible ideas. 

Industry 4.0 ... more meaningless buzzwords from educrats. 

  • Like 5

Share this post


Link to post
Share on other sites
On 9/11/2021 at 8:25 PM, Colin Gilchrist said:

(It was a Yasda machine, with a Fanuc 31iB5 Control...)

A Yasnac isn't a control that I have a lot of experience with.

I do however, know a bit about kinematics, motion control, high-speed modes, and NURBS processing. Add in the correct Dynamic Codes for Tilted Work Plane, Tool Center Point Control, and Workpiece Setting Error Correction, in addition to properly calibrated kinematic values, and the machines work like magic.

Parameter Settings and Proper Kinematic Calibration. Both of these are the real Boundary Conditions of making accurate components on any CNC Equipment. Repeating buzzwords like "Industry 4.0" won't overcome the real-world obstacles everyone encounters when they try and build software that solves a hardware problem. 

Don't get me wrong; I'm all for progress and technological innovation, but I don't hear anything new in the offer being made here. 

You can write NC code to query a lot of functions on the three major brands of control: Siemens, Heidenhain, and Fanuc. But without some kind of understanding of how all the equipment was configured and calibrated, good luck with trying to solve something in a novel way.

I was just finishing up some 5 axis surfacing toolpaths and went really deep into attempting to get the best surface finish. I was very impressed with my little Hurco VM10Ui machine throughout the process...very capable controller. It does get hung up on some really small interpolation moves but that is manageable. Here is an example of a Unified Parallel Morph surface finish.

image.png.dc2fffe6f5990e6e490930713951be19.png

 

 

As far as surface finishes go, how much can servo tuning play a role? The machines run Yaskawa Sigma 5(?) servo drives, which from my understanding are very capable little units. I believe they can be tuned. Now, is that something the machine tool builder does when building a machine? Or is that something the end user can do? I am falling into the rabbit hole of what is "good enough" but if tuning servos can make for smoother motion, I am all for it.

When you start smoothing a 5-axis finishing toolpath, are there methods and techniques to help create smoother motion that don't just say, shorten segment length? Our controllers do offer G05.2 and G05.3 misc. parameters, which from my reading are similar to Fanuc AINano smoothing and the like.

 

Just curious to get others' opinions on developing a process that leads to the BEST surface finishes, but also at reasonable feed rates.

  • Like 2

Share this post


Link to post
Share on other sites

I have less than 0 experience with Hurco. In the 1,000's of shops I've been in in may career, I can count on one hand with fingers to spare the number of Hurcos I've seen.

That said, take what I am about to say with a grain of salt because that's gonna be about what it's worth; most CNC's can be tuned. Typically the factory machine servo settings are shipped with Acc/Dec settings that will allow it to safely machine it's max load capacities. The end result ultimately is if you're running small parts you're leaving some performance on the table. So, I'd call the builder and explain what you're after and chances are they may be able to either offer some advice, or put you in touch with someone that performs the services of servo tuning.

HTH

  • Like 1

Share this post


Link to post
Share on other sites
9 minutes ago, cncappsjames said:

I have less than 0 experience with Hurco. In the 1,000's of shops I've been in in may career, I can count on one hand with fingers to spare the number of Hurcos I've seen.

That said, take what I am about to say with a grain of salt because that's gonna be about what it's worth; most CNC's can be tuned. Typically the factory machine servo settings are shipped with Acc/Dec settings that will allow it to safely machine it's max load capacities. The end result ultimately is if you're running small parts you're leaving some performance on the table. So, I'd call the builder and explain what you're after and chances are they may be able to either offer some advice, or put you in touch with someone that performs the services of servo tuning.

HTH

100% agree with James here.

The tuning available will also depend on the knowledge and skill of the person doing said tuning. Servo Tuning is something that is on my list of skills to acquire eventually. So much to learn; so little time...

  • Like 3

Share this post


Link to post
Share on other sites
Just now, Colin Gilchrist said:

Servo Tuning is something that is on my list of skills to acquire eventually. So much to learn; so little time...

Right there with you Colin. It's DEFINITELY on my skills acquisition list as well.

  • Like 1

Share this post


Link to post
Share on other sites
On 9/24/2021 at 1:55 PM, Colin Gilchrist said:

100% agree with James here.

The tuning available will also depend on the knowledge and skill of the person doing said tuning. Servo Tuning is something that is on my list of skills to acquire eventually. So much to learn; so little time...

Yes it does not sound like something for the faint of heart! Not going to lie

 

On 9/24/2021 at 1:57 PM, cncappsjames said:

Right there with you Colin. It's DEFINITELY on my skills acquisition list as well.

Hurco is definitely a niche brand here in the US (which is funny cuz they're based in Indy, Taiwanese iron though). Over in the UK they seem to have a huge following based on what I have seen by listening to and watching MTDCNC youtube/podcast. All around it is a capable machine builder but it certainly isn't a high end machine. Mid range. Good for job shops. I guess they really try to sell folks on their controller for programming but I just can't justify it even though I learned how to program CNC on a Hurco. Mastercam is just way more powerful and capable. In the end I am running ISO G Code (basically Fanuc) but I do like the controller for setting up. Way more intuitive than Fanuc.

In a perfect world i would only have Heidenhain machines

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

  • Recently Browsing   0 members

    No registered users viewing this page.

Join us!

eMastercam - your online source for all things Mastercam.

Together, we are the strongest Mastercam community on the web with over 56,000 members, and our online store offers a wide selection of training materials for all applications and skill levels.

Follow us