what is a reasonable amount of time to learn 5 axis

[The_kid]

Hello all,

 

I'm curious to see what the opinion is for the amount of time it would take someone to learn 5 axis programming more so 3+2 and what sort of resources and OJT would a person need to be fairly proficient at it? 

[Guest]

It's going to be determined by your comfort level working with planes. In other words, if you program now by working your way around a 3 Dimensional CAD Model, you're most of the way there for 3+2.

 

Simultaneous... There's some learning involved with that. So many variables....

[Guest]

From recent experience......

 

If this person is already well versed in using planes 3+2 is pretty straight forward.....if someone is till copying geometry to different levels of using different files and has no concept of using WCS & Planes, it will take longer.....how long, difficult to say, everyone learns at a different pace, some people can get comfortable with the concept in a matter of days, others, it can take months.

 

Then, does a "reliable" post processor exist for whatever machine/machines code is being created to run on? if not, does this person understand what the correct code would be?

 

It really is a loaded question wherein MUCH depends on the individual themselves, frankly, some people will n ever get it.  That's not meant o be offensive in any way, it is just a matter of fact.

[Bob W.]

I have a few questions in this area.  I desperately need programming help at a high level (orthopedic molds, R&D, aerospace, etc...) and there are two schools of thought.  Find and hire the best and throw enough $$$ at them to get them to join the team or hire a motivated individual with little experience and hope they pick it up quickly.  If there was an individual that hasn't ever programmed real world parts in a shop environment but they had played around with the educational version a bit on their own time with 2.5 and 3 axis parts, how long would it realistically take before they would be at a level where they could take some of the programming load off myself?  Typical parts include landing gear struts, molds for joints, low volume production where getting the program to the machine quickly is key, etc...  The problem is I don't have any extra time to devote to training as I already work 12 hour days just to get done what I need to get done.  Any thoughts?

[Guest MTB Technical Services]

Hello all,

 

I'm curious to see what the opinion is for the amount of time it would take someone to learn 5 axis programming more so 3+2 and what sort of resources and OJT would a person need to be fairly proficient at it? 

 

Like JP said, it's really a loaded question.

 

The hardest part for many is understanding the math behind it so you know what you are looking at in the NC code.

 

Here are links to a series of 5-Axis Machining articles I wrote for my blog.

http://www.mtbtech.net/blog/2012/11/13/5-Axis-Machining-Demystified-Part-One.aspx

http://www.mtbtech.net/blog/2012/11/14/5-Axis-Machining-Demystified-Part-Two.aspx

http://www.mtbtech.net/blog/2012/11/24/5-Axis-Machining-Demystified-Part-Three.aspx

http://www.mtbtech.net/blog/2014/09/08/5-Axis-Machining-Demystified-Part-Four.aspx

 

These are also available as PDFs if you like.

[CJep]

Bob I think the most difficult part is having an understanding of the tool paths. Having a working knowlage of the geometry and then which toolpath to associate to it as well as tool axis control that relates to the capabilities of the machine. Being able to trouble shoot the toolpath and knowing what parameters to change to fix the issue. I've seen instances in the advanced multiaxis tpaths where folks have thrown too many control functions at the toolpath and they get nothing. All the controls canceled one another out. I've fought with toolpaths that have progressively gotten worse only to find there was an obscure switch in the last place you'd look that fixed the issue.

 

Back to the OP, I've been working with the multiaxis tpaths for well over 5yr, I am still learning different ways to use them and different ways to trouble shoot my way out of issues. The folks at CNC continue to add new methods to the multiaxis strategies,... Have you checked out the tilt strategy in gouge checking of the high speed surface finish tpath? It's the easiest multix tpath you'll ever write.

[Bob W.]

Bob I think the most difficult part is having an understanding of the tool paths. Having a working knowlage of the geometry and then which toolpath to associate to it as well as tool axis control that relates to the capabilities of the machine. Being able to trouble shoot the toolpath and knowing what parameters to change to fix the issue. I've seen instances in the advanced multiaxis tpaths where folks have thrown too many control functions at the toolpath and they get nothing. All the controls canceled one another out. I've fought with toolpaths that have progressively gotten worse only to find there was an obscure switch in the last place you'd look that fixed the issue.

 

Back to the OP, I've been working with the multiaxis tpaths for well over 5yr, I am still learning different ways to use them and different ways to trouble shoot my way out of issues. The folks at CNC continue to add new methods to the multiaxis strategies,... Have you checked out the tilt strategy in gouge checking of the high speed surface finish tpath? It's the easiest multix tpath you'll ever write.

Thanks for the insight.  If someone had the aptitudes for this sort of thing and has only played with the home learning edition of Mastercam, how long do you think it would take for them to be a productive programmer in a shop environment with 3-axis 3+1, and 3+2 where they are designing the manufacturing process and programming the parts?  I know this is a wide open question, I'm just looking for thoughts or a guesstimate on the subject.  Personally, I can program just about anything and very efficiently.  I'm just trying to figure out a realistic time frame for training a newbie to be productive at advanced programming.

[Guest MTB Technical Services]

Thanks for the insight.  If someone had the aptitudes for this sort of thing and has only played with the home learning edition of Mastercam, how long do you think it would take for them to be a productive programmer in a shop environment with 3-axis 3+1, and 3+2 where they are designing the manufacturing process and programming the parts?  I know this is a wide open question, I'm just looking for thoughts or a guesstimate on the subject.  Personally, I can program just about anything and very efficiently.  I'm just trying to figure out a realistic time frame for training a newbie to be productive at advanced programming.

 

Bob,

 

The best way would be to treat it like an apprenticeship and actually give 'homework' based on your real parts.

You then review the work at your facility as part of yournormal workday.

The HLE will only get you so far as you need to be able to post the code and look at the various changes from Misc Integer switches and Misc Reals.

SInce you have MC, Vericut as well as the actual machines, an in-house training program on real parts could get someone very far in 6-12 months.

 

I would love to see CNC Software develop an apprenticeship licensing method that could be tied to the customer license at an additional cost.

Maybe something similar to the way SolidWorks network license that allow you to check it out and use it elsewhere.

Dassault does this with CATIA as well. Maybe somehow limit it to a total number of unique files allowed to be saved.

Thee apprentice would have have a license that could be used off-site that needed to be renewed every 3-6 months.

[Mick]

I recently went through this. I had programmed 5 axis in theory, but never in practise, having never had access to a 5 axis machine.

 

Just over twelve months ago, we put in an Okuma MU500VA-II 5 axis machining centre. I went to the Thinc facility in Charlotte, and got practical training in the use of the 5 axis machine, including 3+2 and simultaneous 5 axis, This happened while the machine was being installed.

 

When I got back, I began programming some test parts, in both 3+2 and 5 axis, and spent some time on the machine. To be honest, it has taken me about 3-4 months to get confident with the machine, and the programming of it. Admittledly, I was a relatively experienced programmer in 2-4 axis.

 

Next on the list is probably a dual turret lathe (Okuma Multus U3000) so that will be a new area of exposure

[Guest]

Thanks for the insight.  If someone had the aptitudes for this sort of thing and has only played with the home learning edition of Mastercam, how long do you think it would take for them to be a productive programmer in a shop environment with 3-axis 3+1, and 3+2 where they are designing the manufacturing process and programming the parts?  I know this is a wide open question, I'm just looking for thoughts or a guesstimate on the subject.  Personally, I can program just about anything and very efficiently.  I'm just trying to figure out a realistic time frame for training a newbie to be productive at advanced programming.

Bob,

 

If you're talking about someone with no practical job shop experience, they are just someone that is computer saavy......you're looking at months and months. They would need to learn how to design and machine, metals, tooling, fixture design concepts based on part design. To have any hope of accomplishing something useful in a year would be a tall challenge.

 

If you're talking someone that has been involved in machining, not just operating, myself personally, I would give them 6 months and I would expect to see continuing improvement of the next 6 months.

 

It's tough, though not impossible to turn a person with the right aptitude & attitude into a net positive but it won't happen quick.

[Guest]

Bottom line, don't expect immediate results. Patience. You didn't learn it over night, neither will anyone else.

 

If the shop has a tool like CAMplete (only available for certain machine tools), it can aid in getting the programmer proficient faster because they can in short order see how what they are doing affects the machine without having to go out to the machine and see for themselves.

[CJep]

Bob,

 

The best way would be to treat it like an apprenticeship and actually give 'homework' based on your real parts.

You then review the work at your facility as part of yournormal workday.

 

This is a good idea, you will definetly see how motivated the individual is concerning learning multix programming. I would think that someone who has solid background in CNC machining and highly motivated should be able to do some basic multiaxis programming 6-8 months.

Does your reseller offer multiaxis training as far as Mastercam is concerned?

[The_kid]

wow some really good input guys.  I have the basic experience for prismatic parts and have a good idea of how surfacing needs to work.  I need to work on some of the high speed tool paths and surfacing.  This shouldn't take long.  I've done programming for a VTL with live tooling etc. and have a varied background of 8 years in the industry.  So I'm not completely in the dark.  I need the practice and the input. I like to see the different strategies of how someone can go about doing the same thing and which is most efficient both for program creation and actual machine run time.  I've always wanted to program least for the last six years. I understand that it is not cost effective for anyone to train so I try to learn on my own time but this only takes you so far. Trying to get ideas for parts that vary with degree of difficulty as my skill progresses. I'm a very quick learner and highly motivated. Anyone got suggestions of resources that doesn't cost a ton? I'm even willing to take classes at night for this sort of thing.  While I'm not THAT green I think having the parts set out to practice on would be a huge help.   

[The_kid]

6-8 months of constant 8-10hr days?  

[CJep]

I understand that it is not cost effective for anyone to train so I try to learn on my own time but this only takes you so far.

 

I would have to say the opposite is true,..

How cost effective is it to have someone not familiar enough with what they are doing to crash a multi thousand machine?

 

If we don't train the next generation of machinists, then what,...

[K2csq7]

IMO, this ability is something you either have or you don't. 

 

If you have it, and the right teacher/classroom... < 1 month to independently program some simple parallel surfacing, Swarf a wall or 2, and Curve an edge with some side tilt. (experienced 3 axis programmer that has good spatial orientation skills)

 

If you don't, get out the checkbook and call the MTB service dept. Schedule some repairs now.

[Smrk]

I am very experienced using WCS and modelling. Also i programmed 3+2 for years, very variable jobs, so not only  copy and paste. But in september we went to five axis and it was completly new world. Especially Tool axis control page :-) It takes me about 1 month to learn what which toolpath do and how and when to use it. Hardest part for me is postprocessor and machine simulation, i am still arguing with reseller about that.

But i'm still cautious and paste program pause into g code before five axis part of program :-) I am still learning.

[DODGERFAN]

There was a webinar on Friday with some good info for you.  Here is the link for a replay.

https://www.youtube.com/watch?v=5tEmIaGOFeI&feature=youtu.be

[Del.]

About an hour.

[DODGERFAN]

About an hour.

58 minutes.    It always takes fish longer. 

[THEE THAINZ™]

Not when he swims down stream.......

[CJep]

 

Del., on 15 Dec 2014 - 11:39 AM, said:

 

About an hour.

58 minutes.    It always takes fish longer.

 

 

That's all the longer he can hold his breath,..

[Jayson Kramer]

The_Kid. so do you feel comfortable with planes and how they work?

As for files have you looked at some of the shared files on the FTP to play with a practice with?

 

Have you done any Horizontal programming of even just indexing using planes on the VMC with 4th. These are helpful tools when moving into 3+2.

 

Just some thoughts and questions help with moving forward.

 

PS the answer is 62.5 min.

[Matthew Hajicek - Singularity]

I firmly believe that a CNC programmer should start out with a file, a bandsaw, a bench grinder and a drill press.  Gotta walk before you can run.

[RaiderX]

I firmly believe that a CNC programmer should start out with a file, a bandsaw, a bench grinder and a drill press.  Gotta walk before you can run.

or at least a bridgeport, manual lathe and a drawing. nothing like stepping a job through manually to help understand the basics that are being multi tasked in a CNC.

 

5 axis learning curve? well if you can do a 5 axis job on a three or 4 axis machine your gonna love hitten it all at once on a 5 axis. much easier in my opinion. Im talking 3+2. simultaneous motion is an animal all its own. kinda gotta jump in and get the feel for it. some get it much faster than others.