CAD/CAM/CNC "Associativity" Benefit or Futility

[Scott_bob]

I could sure use some insight on this subject!

To save time, I am not going to define the consept at this point. I am hoping that as you offer an opinion on the promises of CAD/CAM "associativity" and the resulting CNC code, that this illusive technology can be better understood.

Anyone have any success stories?
Disappointments?
Anyone seen any empty promises go unrealized?
How about a experience like, "I can't imagine not maintaining" it.
How much "extra work" does it take?

What is "it"
I thought about doing a poll here, but few participate in them...

What do you think?

[derekj308]

Hi Scott-bob
The term 'associativity' when used in the context of CAD/CAM refers to the ability of a design/manufacturing data (CAD/CAM, its broadly just geometric 'data') system to dynamically change 'downstream data' based on 'upstream data' changes. An example of this concept would include a simple model (lets use a 10" x 10" cast plate with a 1" machined hole in the middle), the casting model, the patterns for the sand mold, the fixture to hold the casting and the toolpath to machine it. If you create these elements of the manufacturing process in an 'associative' manner then any changes made to the 'upstream data' will effect intended change to 'downstream data'. The example would be if you changed the diameter of the hole in the cast plate CAD model from 1" to 2" then you would expect the patterns CAD model to update and reflect the new size (ie the hole in the pattern would change from 1" to 2" + shrinkage), the fixture would remain the same and the tool path would change to machine a 2" hole instead of a 1" hole. This is all done automatically by changing the 1" hole in the design CAD model to a 2" diameter.

The catch is, is that you need to define how these different elements of the manufacturing process relate to each other and if you want change of upstream data to effect downstream data or not. This concept is called defining design intent. Defining design intent defines the associativity between each of the manufacturing elements. The CAD/CAM system does not automatically define this associativity for you. The tools are there for you to use but it requires a lot of training and some experience and mentoring to make it efficient within a work environment.

The type of system you require is largely dependent upon the type of work you do. Repetitive work and variation upon a theme are real 'monty' for associative (also known as 'parametric') CAD/CAM systems but they require a lot of sound design up front so that any variations wont be inheriting crappy design intent. If you do mostly one off's then associativity won't give you any short or long term benefit since you are always 'starting from scratch'.

I am assuming from your question that your environment is 2D and your toolpaths are manually created based on 2D drawings. I'm also assuming that you have a number of CAD/CAM vendors all wanting to sell you their systems. If you have a good 2D user they will quickly pick up 3D and be able to create models than you can use in a CAM package to generate toolpaths. If you are doing toolpaths manually then you will quickly notice a dramatic decrease in programming time once your guys are up to speed with the CAM software.

Does the CAD system and the CAM system need to be associative? NO! Plenty of manufacturers run seperate CAD/CAM systems and are very proficient at concept to production manufacturing.
If they were associative would there be any production gains? MAYBE! If you guys are really switched on they will see the opportunities for gains and make them work. If your guys are average you would never see any gains. They will just imitate old practices using new resources.
Does the CAD system and the CAM system need to be from the same software package? NO! These days just about every CAD/CAM system can 'talk' to each other via a common file format called STEP 214. Occasionly you will get a hiccup in data translation but it can be easily overcome with a bit of tweaking.
Would it help if they were from the same software company? Yeah, it would help from the POV of the users being familiar with the software companies 'interface look and feel' but its not a requirement for productivity.

Summary: Do plenty of research and make the vendors show you how their system works with your products and processes and not just pretty pictures and models of someone elses products and processes.

Cheers
derekj308

[mrainey]

I've always wondered about this.

If a designer were to change, say, the corner radii on a pocket, how would the associative system "know" what to do?

Depending on the direction and magnitude of the change, and what other features were being machined with the pocket tools, a sharp human programmer might wind up making significant changes to his/her machining steps and tool selection in order to maximize efficiency for the new design.

For example - if a pocket had several .250 corner radii, the roughing tool might be .500 diameter. If the designer changed the corner radii to .500, a .500 end mill might no longer be considered as most efficient for roughing the pocket when a larger one would fit.

The associative system might change the .500 tool to a 1.000 tool. What if the old .500 end mill was also being used to machine other areas on the part, and a 1.000 wouldn't do in those areas? What if you had previously used a .500 drill to open up a hole in the pocket for the .500 end mill? Would the associative system pick up on this and substitute a 1.000 drill? If the system didn't have that 1.000 drill defined, would it stop and throw up an alert?

I guess what I'm asking is - how efficient would these new programs be, and how well could the system spot and deal with all the possible side-effects? There's a lot more to making intelligent machining choices than just looking at tool diameter and tool type.

How much do these systems cost?

[Scott_bob]

Originally Posted by derekj308 The term 'associativity' when used in the context of CAD/CAM refers to the ability of a design/manufacturing data to dynamically update 'downstream data' because of 'upstream data' changes. An example of this concept would include a casting model, the patterns for the sand mold, the fixture to hold the casting and the toolpath to machine it. If you create these elements of the manufacturing process in an 'associative' manner then any changes made to the 'upstream data' will migrate to change the 'downstream data'. The catch is, is that you need to define how these different elements of the manufacturing process relate to each other and if you want change of upstream data to effect downstream data or not. This concept is called defining design intent. Defining design intent defines the associativity between each of the manufacturing elements. The CAD/CAM system does not automatically define this associativity for you. The tools are there for you to use, but it requires a lot of training and some experience and mentoring to make it efficient within a work environment. Does the CAD system and the CAM system need to be associative? NO! Plenty of manufacturers run separate CAD/CAM systems and are very proficient at concept to production manufacturing. If they were associative would there be any production gains? Maybe! If your guys are really switched on they will see the opportunities for gains and make them work. If your guys are "average" you would never see any gains. They will just imitate old practices using new resources. Does the CAD system and the CAM system need to be from the same software package? NO! These days just about every CAD/CAM system can 'talk' to each other. Would it help if they were from the same software company? Yeah, it would help. derekj308

Well said, well said...
I knew someone would lay out a better scenario than I was will to pound out. You sound like you have some experience with the "theory of associativity". Thank you for your insight! Now what I am hoping for is that some one can tell me how their product cycle has been revolutionized due to these benefits. Because as you pointed out, this is not easy to do. The guys and gals you work with are willing to put out above "average" performance on the job, and the beneficial promises of associativity have materialized for you...

On the other hand (I have two) maybe what I'm hoping for is that nobody is going to be able to demonstrate the application of this technology in a real world scenario. Oh sure, I can demonstrate associativity tomorrow on a simplified example but tell that you and your partners on the job are using and benefiting from this theoretical technology. When the guys and gals you work with are only willing to put out "average" performance on the job, then the theoretical beneficial promises of associativity will never materialize.

What say you?

[Scott_bob]

Originally Posted by mrainey I've always wondered about this. If a designer were to change, say, the corner radii on a pocket, how would the associative system "know" what to do? Depending on the direction and magnitude of the change, and what other features were being machined with the pocket tools, a sharp human programmer might wind up making significant changes to his/her machining steps and tool selection in order to maximize efficiency for the new design. For example - if a pocket had several .250 corner radii, the roughing tool might be .500 diameter. If the designer changed the corner radii to .500, a .500 end mill might no longer be considered as most efficient for roughing the pocket when a larger one would fit. The associative system might change the .500 tool to a 1.000 tool (I don't know bout that). What if the old .500 end mill was also being used to machine other areas on the part, and a 1.000 wouldn't do in those areas? What if you had previously used a .500 drill to open up a hole in the pocket for the .500 end mill? Would the associative system pick up on this and substitute a 1.000 drill? If the system didn't have that 1.000 drill defined, would it stop and throw up an alert? (Nope) I guess what I'm asking is - how efficient would these new programs be, and how well could the system spot and deal with all the possible side-effects? There's a lot more to making intelligent machining choices than just looking at tool diameter and tool type. How much do these systems cost?

You're right on my track... The best associative systems are very expensive ^20k. Very difficult to use, requiring a lot of training to get the simplest of results. Steep learning curves full of frustration due to sheer complexity. I have never seen so many freak'n Icons, pull downs and window dialogues that open other window dialogues, now where was I...
Sometimes I just want to go back to something easier to use. Oops I guess I have revealed a bit of my issue. Oh well, that's progress (I hope). I'll get this stuff eventually, I just hope everyone else around me will get this stuff as quickly, or I am going to get irritated. After all, the work has got to get done along with climbing this freak’n learning curve and you know the boss is not going to want to hear that it's gonna take longer till WE get good at this stuff. If we can just find a way to hold off old Murphy and his law.

Come on, someone shed some light in this dark tunnel. Or, help me pull the plug before it's too late. I really have no experience to guide me here...

Give a guy a break,

[derekj308]

Hi Scott-bob
Many others before you have also pondered the 'should I' or 'shouldn't I' make the next step into modern manufacturing software. I would be tipping that the one's who didn't are still struggling or out of business and the ones who did are now glad they did and can look back at the hard road travelled and say it was worth it.
mrainey hit the nail right on the head with his questions about how the software would interpret the changes and what to do with them. The simple answer is that there are varying levels of involvement from the software but in the end a skilled human needs to make a decision which of the choices he/she is presented with is most suitable to accomodate the change. The level of programming required to automate a process of even simple manufacturing associations is complicated and needs to be justified by high volume and/or high margin products to make the effort worthwhile.

You may even consider hiring a skilled person with your software purchase to speed up the learning curve for your company. It would certainly be worthwhile in terms of return on investment as a whole. He/she could dramatically increase the rate at which all of your relevant personnel learn this new skill and apply it in their day to day tasks. Find a person with the right skill set for your business type who also has the software experience you are looking for. A well rounded designer/manufacturing engineer does not need to work with a specific CAD/CAM system. Its about knowing the principles of robust modelling and when associativity is required is what makes the person right for the job. Using a different CAD/CAM package for the right person is just semantics.

derekj308

[Scott_bob]

derekj308,

Ok, you can ansewer the theoretical question. I really, really want to know if you personally have benefited from the brave new world of associative CAD/CAM/CNC?
Are your product design processes faster?
Do you see fewer errors because of the higher level of technological integration?
Is associative engineering making your organization leaner?

Please, Please not just theoretical ansewers, I'm really, really interested in real world production ansewers.

Thank you for all your insight, and it's fine if you don't have any actual real world example. But if you do, please share it with us. I know I am not the only one interested...

[derekj308]

Hi Scott/bob
I do have real world production examples for you and I do work with this type of technology everyday as a manager of people who use this techology to perform their day to day tasks. I was also a full time user not so long ago so I can speak from experience. The company I work for produces many injection molded parts for use as simple moldings and also in assemblies. When we design one of these injection molded parts we also do what is called concurrent manufacturing. By this I mean we design the part and the tool in parallel or at least as close to parallel as we can.

Real World Examples

Old way

Prior to our 3D associative software our 'old' way of producing an injection moulded plastic part was to design a part on a 2D drawing, then give it to the toolmaker who would then have to interpret the drawing, work out how he will make the tool and then have to construct surfaces and/or profiles to generate toolpaths from. If the part is not designed for manufacture then the toolmaker will then go back to the designer, the designer makes the changes, the toolmaker revises his design until he finds the next problem. Then its back to the designer and the cycle continues. This type of process is a series process which adds unnecessary time to the ultimate goal of producing an injection moded part. We didn't know how the part would fill, we didn't know where gas traps would be, we didn't know if our part was going to be strong enough to withstand the loads applied to it and we didn't know how much time it would take to manufacture the tool.

New way

Our new way of producing an injection molded part is to design the part with only the basic features required and then specify how we are going to 'split' the tool. We then do a quick analysis using a plastic flow simulator to see if our part will fill they way we expect. Be mindful that this is very early in the design phase. We then perform a FEA (Finite Element Analysis) to establish if our part is even in the ball park of being able to withstand the loads that will be applied to it. Once we are satified that we can fill the part and that it will be suitable for the loads applied to it we build a CAD model of the tool using library parts and procedures we have defined to enable us to quickly establish the major parameters of the tool - size of the dieset, cooling channels, inserts, sliding core anciliary parts,etc. Remember the part is not fully designed and we can get to this stage in about 3-4 days which in todays terms is slooooow. Now we get serious and do some real work. We add in features to the design part and watch the changes propagate through to the tool. As we add a round (fillet) to the design model the tool cavity will be updated automatically to reflect that change. We are in a position to immediately see the effect of a change in our design model (the upstream data) in our injection molding tool (the downstream data). At this point in our design cycle we have absolute confidence in the way in which our part will mould, its strength and the design of the tool is basically done using the same source of data which is the design model. Once we have 'polished' the design to make the tool easier to manufacture (without sacrificing design intent I might add) we can now use all of the geometric data we just created (to validate the design of the tool within CAD) to use as surfaces and curves to generate toolpaths to machine cavities and also to machine electrodes if required.

We have only recently completed a prototype injection molding die (single cavity, 3 removable cores, sprue straight into part) that we completed in 9 working days start to finish. We spent 2 days writing the toolpaths and 3 days part design, molding verification and FEA verification. Prior to our new method of design and manufacture we would not have been able to complete it in under 30 days all up (compared with 14 days all up) and certainly not to the same level of accuracy we achieved not to mention with very little confidence of being able to produce a working prototype. We could easily claim a 50% reduction in concept to production of a protoype component in the material it will be mass produced in. We then delivered it to our customer and they were blown away with how fast we were able to give them working prototypes in the material required. We have the order and are now producing a 4 cavity production tool. I must stress that this in todays terms is not fast and could be done in much shorter leadtimes. However you look at it though, for us, we have reduced our product lead time by in most cases, half. We are very happy that we chose to select an 'associative' solid modelling software package. But as I have stated in my previous posts on this subject it’s our skilled, enthusiastic, 'think outside the box' staff that make this work not simply having the software. Its also about us working as a team. Sales, engineering and production need to function as one big team to make any 'part making' business work.

As far as making our organization leaner. No. We didn't and we don't intend to. But what we have achieved is a system where our customers are happy with the product and service we provide and they fill our new found capacity with new work.

I have set my keyboard on fire from too much typing and I apologise to those who now have red eyes from reading my encyclopedia wanna-be post. LOL.

derekj308 who now has RSI of the hand lol

[Scott_bob]

derekj308,

Awesome! Now don't you feel better having said that so consisly? Anyone in our buisness of engineered products can read between the lines and see a great deal more could be said about your process. You saw my hope to get to the bottom line and got there quickly. Nice... I notice you emphasized your skilled people working as "a team", and in previous post having people who are "switched on" and "highly motiviated and skilled" as being the key to using this level of sophistication in CAD/CAM/CNC.

How do we hire such people?
Are they willing to trade income for sunshine here in southern California?
Oh, the future looks bright except for the less than average people around here.

[derekj308]

Hi Scott/bob
What products does your business manfacture?

derekj308

[Scott_bob]

I have learned from experience that the Internet is not a place for total transparency. I protect my personal information by "attempting" to maintain anonymity. Having said that I also am a hopeless optimist when it comes to technology. In this country we use a phrase borrowed from your country; "Don't be like the ostrich that buries its head in the sand to hide from danger". I still find it hard to understand an animal this dumb. Of course, the human animal is no more intelligent when it attempts to do the same thing when threatened to extinction, by competition that will eat you alive... I cannot tell you how often I personally have tried to promote improvements that cost money to make more money, and have been ignored by ignorant or ostrich like leadership. The business world today is like life on your great barrier reef off the coast of Australia, very dangerous... If you are going to go swimming with sharks, you better be aware and prepared.

Oh, but you asked what we do in mfg... Well, we design engineered solutions for nearly all disciplines of businesses. Aerospace, automotive, commercial, semiconductor, off highway, you name it, customers from NASA, to McCaulla (chainsaws), from Chrysler to Amat semiconductor components. Nearly everyone needs at some point of their product, what we can engineer, manufacture and deliver prototypes in an average of 4 to 6 weeks, and production volumes to the millions in 12 weeks.

So here we are groping along attempting to migrate from 2D to 3D associative CAD/CAM/CNC. We buy as few licenses of Unigraphics as possible, send guys to training for CAM and then put off buying the license so any familiarity with the program is lost. After all you have to use it or lose it, or claw your way up the steep learning curve slipping and sliding 1 step back for every 2 steps forward. Sound familiar? I hope not...

Anyway, for us the benefits of associative engineering are obvious. We design parts, the fixtures for machining those parts, every part requires a mold, sometimes depending on production quantities multiple cavity molds. FEA is increasingly important to many customers, and moldflow for flow characteristics is available for engineering as well. Every part must be inspected of course so we do CMM on those parts too. Of course ECNs are a regular part of all engineering departments, so we have our share of engineering change notices. Our business model includes leveraging lower labor rates in Mexico for our high volume production processes. This is not easy, believe me.
Customer demands and expectations are only getting tougher with global competition...
I share your vision of the near future, "get up to speed with new technology, or go by the wayside, cause its only a mater of time before your competition passes you by"...

Enough said for now,

[weimedog]

I also would be interested in success storys. At one time I was tasked with writing a specification for a front end to a CAM package to
allow that concept of associativity to function. I used a solid modeling concept where
"attributes" (basically raw data to be later analysed) migrated thru the process either
thru boolean subtractions or thru the front end software...I assumed that there would
be situations where the process evaluation would fail and there would need to be manual
intervention of the re-evaluation process. Used software flags that would be
set to true or false based on the state of the evaluation of the process on those
specific "objects". Each object pointed to the prior one in a process model and to the
next one in the process model. (yea it was old C++ & object oriented programming
concepts). A set of "objects" would be the "process model"..each object in addition
to pointing to the next and prior would point to a geometry set and process definition
on how to machine that geometry. (Could point to the results of a boolean subtraction
of a toolvolume from a stock model defined by a PRIOR process.) The machining
process definition could be a series of "APT" like commands or a more automated
toolpath generation style function...like "solid areaclear" which would bounce a tool
around the inside of a solid model of stock to be removed. The system architechure was
designed to allowed a developer to blend any type of User interface they desired
to either the entire process modeler or each object as the process model was built.
This was to allow flexibilty on how to handle special cases where as a result of up
stream changes, down stream objects were required to re-evaluate or be removed and
replaced with different ones.

Basically it worked pretty well simply because I didn't attempt to automate the
entire process...gave a developer plenty of options on how to handle a stituation
where because of upstream changes, down stream portions of the process model
where no longer possible or even relevent. The system was flexable enough to
allow a completely manual toolpath with each process obect containing an old
command style toolpath pointing to a list of geometry features (surfaces, faces,
lines, arcs, splines, or "profiles") ....implemented in this way it would look completely
as a conventional command based system would look. Or it could handle a more
automated approach where areaclear, solid areaclear, profile, solid profile, or
even more automated surfacing routines could be bundled together and modeled
to define a machining process. The best part was the capability of blending BOTH
styles and still allowing the output of one process to be the input of the next.

Anyway I'm rambling and the concept IS possible as LONG as the developer doesn't
try an ALL automation style to the package.

Hard part was to market the concept. Most people in the industry were not interested
in pushing that edge until someone else did it first..an endless loop.

Also the NIH concept was tough as everyone (including myself) has a pet theory on
how CAD/CAM systems are supposed to look along with an architechure legacy with
current products. I was an "outcast" in my company as I didn't beleive everything
could be automated thru some crazy "knowledge based" scheme!. I believed that a truely
useful system had to blend automation and manual NC programming technics.
Bottom line was many companies did proof of concepts and a few actually implemented this
concept in bits and pieces.(I will never tell which ones were based on where I came from
as I have been out of that business for quite a while) ...and now many years later I would
really be curious who actually figured it out and built a functioning system flexible
enough to be practical and smart enough to automate where it makes SENSE!

Actually the part that really intrigued me was that the places where the
MOST progress was being made in pushing these concepts were overseas.
Pacific Rim in particular. Watching how my customers overseas were
relative to those here...it was only a matter of time (ten years ago) where
the most inovative SW product development will be done somewhere other
than the USA. But I guess thats where the Manufacturing is....

Anyone want to chat about SW theory on how to develop a system with
Associativity?? That might be interesting.