Aluminum Saw

[QuintonAlame]

Me and a buddy of mine are thinking taking my stihl 460 apart and replacing the plastic with aluminum and a machine shop.  Im shure it will make the saw weigh more but all plastic will be replaced with aluminum.  But we going to keep the top orange plastic covers. We pretty much going to rebuild the saw with aluminum exp the motor and trigger, air filter housing.  But the rest we making aluminum.  Do you all think it would be a good idea.  We really want to do it to see if we can.

[weimedog]

For real?   Why?   Unless you have a pretty sophisticated CAD/CAM setup along with a the matching capabilities in a CNC mill (I believe Stihl has Catia for design, don't know what they use to cut molds with anymore.), those curves & surfaces are going to pretty tough to replicate....either that or your saw is going to look pretty blocky! To get close to a good finish on those surfaces, going to have to use either a lot of passes with a ball or have a 5 axis machine.....which do you plan to try? Even old tracer mills would have a tough time with the transitions....AND I want to see a aluminum air filter...

[beenthere]

Yep...WHY??

       

[clww]

It seems to me the engineers at Stihl thought about plastic, steel, aluminum and rubber when they designed their saws. You're going to take a several hundred dollar saw and turn it into a piece of junk for what?

[smalldog]

What???  Even when I had access to a cmm room, AutoCAD, a whole machine shop and plenty of time, I wouldn't do it. Is this for a school project? Something you can polish to a mirror finish and sit on a shelf?

Here's what you do. Put some oil and premix in that thing, and go find a big tree that fell over in a storm and cut it up. Git-er-done!

[Al_Smith]

It would be very time consuming and for all intents not improve the saw one iota .What it would do is add weight to the saw .

A solid machined piece of aluminum most likely would be stronger than the older diecast parts of yesteryear .Those were replaced with high impact thermal plastic and have been proven to be much more robust than the previous .

I almost think I've been fiddling with forums too long or I'm loosing my mind because I think I've had this conversation before. Then again there was another well meaning fellow that was bound and determined he could install Stihl parts in a Poulan saw and revolutionize the industry with that innovation .Odd I haven't heard back how that worked out .

[Ron Scott]

Really a waste of time.

[mikerat]

Your 27 years old yes I think you should do it!!!  Do I think you will make a better saw? NO but look at what you may learn, so then some day when your a old guy like me you will be able to give sound advise to younger guys based on your expeirence. When I was about 24 I had a old Dodge Power Wagon with a rusted out cab, I guy needed some work done  which I did and he had a newer cab. I swapped the new cab for the old one and it was the job from hell. Everyone said don't do it in hindsight they were right but it was a great learning expeirence. So when I look back now I laugh but it gave me some good insight and I think you will learn a lot of stuff.

[QuintonAlame]

I mean the plastic back handle (housing) we want to do it just to see if we can and yes we wwe got acvcess to a 5 axis mill all that and some of you are right some lines are going to be hard to replicate but if its a success i willl send a pic

[Al_Smith]

  Now come on now Quinton,time to fess up .Either you or your double posted the same type of brain storm idea a year or so ago and got about the same comments back as now .

Flying casually though cyberspace are a few of us that remember everything .

Never forget that cyberspace just like the night has a thousand eyes .

[weimedog]

I mean the plastic back handle (housing) we want to do it just to see if we can and yes we wwe got acvcess to a 5 axis mill all that and some of you are right some lines are going to be hard to replicate but if its a success i willl send a pic

Real curious as to how you will build your "CAD" model. Not like that info is floating around and it would take time to measure & create something close. If your place has things like AutoCad & Point Control...forget it, its going to be a rough approximation of that Stihl, maybe functional but blocky and heavy without a lot of secondary operations done essentially by hand. Its going to take a fairly sophisticated CAD system and then a reasonably good CAM package along with good surface machining capability to get in the ball park. Most machine shops have things like SurfCam, Point Control, Gibb's and things of that nature and unless they get a pretty good model, they will have a tough time evaluating those surfaces to get a good toolpath so manipulation of the model is required..(time!!). As far as CAD systems...there are a several that will be able to get a good visual model even at the low end...but can they create a good enough model to machine? So ..what systems do you have at your disposal to play around on? Of course if one has a lot of time to jerk around, you can aproximate almost anything by defining geometry to get what you want as a cut....a few surfaces...few lines & splines to get what ends up to be a toolpath close to your intended model...but you have to have a lot of free time & acess to some real capability! Hey...if you do, I'm jealous!  And yea..it would be fun!

[Jeff]

I mean the plastic back handle (housing) we want to do it just to see if we can and yes we wwe got acvcess to a 5 axis mill all that and some of you are right some lines are going to be hard to replicate but if its a success i willl send a pic

I wanna see the photos of the failure too. 

[Al_Smith]

CAD  or just about any CNC program would take a programer a month of Sundays to get the bugs worked out of .A novice would take eternity .

It isn't just as simple as most people  think it is .I have enough trouble just trouble shooting the danged machines let alone rewriting a program .Fact is I'd stand a better chance of cutting out said super handle with a hacksaw and a file .

Good Lawd I once worked on a machine that cut automobile pistons and only 11 existed in the entire world .That danged thing used four different programing languages .What a joy that was .

[weimedog]

I think Stihl was a Catia Customer, but I don't know for certain..and if the design model was available along with a either the Catia CAM based system or secondary one like DP Tech or SurfCam...its actually a day or two of work. If you have to generate it from scratch... ...man I don't know. Its a lot tougher trying to input a physical model vs. just designing one from scratch! I've done both...many times. Working from scratch you can model around the system capabilities and limitations. Trying to measure a physical model....lots of variables play havoc. Things like mounting locations, throttle mounting locations internal to a handle and the blend from that geometry into the handle geometry can be interesting both to model from a measured physical model and to machine. Oh man does this concept ring back memories! Its certainly do-able..but its a time thing! If was a certain pacific rim backed concern..its worth the time to build model & machine the molds required to produce aftermarket parts! But there is a economic reason to futz with it.....for one handle for play? As a replica vs. tooling? I wish I had toys like that at my disposal and the time!

[Al_Smith]

Oh you can duplicate parts as I've done many times but you have to have a valid reason to .No CNC on my part just hours upon hours of milling machine and lathe work .

Some things like a Stihl handle aren't that hard to find .On the other hand a valve spring retainer for a Ferguson TO 20 vintage 1951 or a D4 Cat push rod made in 1943 are not something you just trot on down to the local NAPA dealer and buy off the shelf .

[weimedog]

The whole CAD/CAM and CNC world is one of the most miss understood and miss represented technologies out there. Years and years of hype and expectations have built this twisted impression of what that technology can and cannot do. Don't have time to write a book but a few things come to mind:

1) All CAD systems are not created equal....the ability to model geometry accurately and the ability to evaluate (Many have no idea what that implies or even means) has a whole lot to do with the way the model data is defined and stored.
2) All CAM systems are not created equal...in addition to the types and way data is stored, they are slave to to things like "offsetting" code and intersection evaluation code which in turn is slave to the type and integrity of the model.
3) Bottom line is certain types of CAD and CAM systems, because of how they infact define & evaluate geometry; are better in dealing with certain classes of geometry than others. (Surfaces, Solids, Lines, Arcs, and Splines...and sub classes of things like splines...nurbs, b-rep, faceted models etc)
4) And depending who wrote the code along with how they define data, some CAM systems are better than others even when their geometry definitions are similar..obvious I guess but when your looking at all the flash & sizzle......many are blinded to the roots of a given environment. (A lot of the history is lost even to the vendors these days as to the roots of the algorithms they use)

Bottom line? If this guy has the right combination of CAD/CAM AND data, making an aluminum replica of a plastic handle isn't earth shattering. If they don't, its basically going to be like you doing it by eye & feel on manual machines...or worst if they haven't learned basic skills on their machine & system.

AND there are some many who are so called "experts" in the CNC field who never developed the skills to run their machines in "MDI" and with that big knob on the front of the control and think manual style machining techniques won't work on CNC machines.....

And the definition of frustration is a shop that has learned a specific system well..the interfaces....but not the underpinnings, takes on a job that's geometry is outside the performance envelop of their system. That job turns into an exercise in working around the limitations of the system....!

[Al_Smith]

Walter you're starting to talk the stuff I do when you mentioned MDI -manual data interface . That was one of the methods that piston profiler worked on .

Then you get into robotics which are supposed to work on the Carpathion plane .Okay fine but is that tool center point ,machine center point and which of the seven axis is it ? Liniar interpolation ,circular interpolation .I get so confused at times I don't know if I'm walking or astride  half the time . No wonder I find simple things like rebuilding saw engines  relaxing .

[weimedog]

Yes..a long time ago...far far away in a different life, I lived, ate and breathed CAD/CAM/CNC as I sold & supported end user CAD/CAM products and then later, software tools to those who were building CAD/CAM systems and Smart NC Controls.  A solid modeling based set of software tools for a company now owned by Dessault. Not many places that dealt in that world I haven't visited of that time period at one time or another. Actually help companies spec out their products as well. My last customer was a famous German car builder.  Way Before that I was a manufacturing engineer specializing in the stuff..then An Applications Engineer for a couple of CAD/CAM vendors...a product manager for a CAM product and all around BS artist in that realm for longer than I want to admit. The thing that disappointments me is every time I go back and catch up with the technology...how little has really changed other than interfaces. But....Now I reminisce and wonder why I didn't push dirt & farm sooner in life. Probably would still have my first family if I had led a more conventional life.

[Al_Smith]

The role of a controls engineer is one I'm glad I never went into although I had the chance early on .Now of days high speed manufacturing is changing so fastly that what might be state of the art today will be passe in 6 months or less and obsolete in three years .You can't even find parts to fix the darned things a lot of times .

They used to for example built rock solid machinery for the automobile industry that had a planned life of about 20 years .That's now down to about 7 .A lot of it is aluminum though which will make a bunch of beer cans so it's not a total lose .

[Jeff]

Aren't there duplicator machines that actually use some sort of cad instrument to measure a model, then reproduce from those measurements?