The bigger hammer

[Stepper Monkey]

At the risk of starting a rather "spirited" discussion, I thought I might start a thread about design, suitability for purpose, and the tendencies toward massive overkill in this forum - it scares and confuses a lot of newbies I notice. If you read to the end, you'll see it also has more grim implications.

Whether through lack of thorough understanding the technical details of our rather new hobby, blind following of old design methods that came before in industrial machining, misinformation from sales literature, keeping up with the joneses, or just plain old-fashioned machismo, I note a strong tendency toward the Tim Allen school of design here.

When it comes to the contexts that many of us know machining as - industrial machines - I see a blind following of ideas that may or may not hold true in smaller-scale CNC machines. The cries of "real machines all have this feature" or "no high-end machine uses that feature" as justifications for thier superiority are common, but are they really true? It bears discussion.
"Real", "Industrial", "High-end" ships have foot-thick steel hulls, use multiple diesel engines and all have navigation radar, but I strangely feel the need to have none of these features on a ski boat.
I feel that if this were a watercraft forum, many here would be spending thier time building thier ten-foot Bass boats out of steel battleship plate, cramming the every last inch of them with 5-ton diesels, and figuring out ingenious ways to mount a 40-foot radar mast on top. Some of you are trying to figure out how to do that last one in your heads right now, aren't you?

In short, why are we so stuck on one definition of "best" when it actually varies from purpose to purpose?

I have so far been afraid to even mention in discussion new developments like carbon fiber rails (vibration attenuating!), aluminum screws with - god forbid - modern plastic bearing materials for nuts and ways, and even glass fiber impregnated thermoplastic or ceramic composite frames. They actually can well outperform the traditional methods of construction when used in smaller mills. They are used currently in manufacturing in robotic and linear units in our sizes as the best solution, period. They are available to us as well, though entirely unutilized right now.
They aren't a good solution however, in full-on turret mill-sized units. Since they aren't in Bridgeports, though, they must be crap, right?
Unfortunately, Bridgeport-sized construction methods ultimately may not be right for us.

I only bring this up, as being in discussion with (i.e. drinking with engineers from) a couple of companies designing CNC machines, I found out that quite honestly they are putting in higher cost and sub-optimally performing parts in thier offerings solely because of fears of our consumer resistance to anything different than the "big" mills as being inferior.

In other words, our attitudes, or percieved attitudes, are actually hindering our diversity of choices of machines out there, and costing us extra money, for what in the end is lower quality equipment for our hobby. That should make us all think.

[LeeWay]

I find your post interesting. I too have noticed some of the same things, however you don't want to make a bass boat out of paper mache' either. There are limits and boundaries that should not be crossed. That said, in this day and age of all the new materials and techniques of doing different things, there is more than one way to build a machine.
There is a lot to be said of tried and true methods though. Those that have gone before us obviously have a lot of time and money invested in achieving desired results around a particular budget.
This being a good hobby and having a multitude of sizes, costs, accuracies and availabilities of materials to choose from, we should be seeing more people pushing the design envelop.
I personally have an ongoing project in which I am using some different materials. Different for mills anyway. I am using 8020 extrusions for the structure. I have been following the polymer concrete thread as well. I will be filling these extrusions with something. Either that stuff or some non shrinking hydraulic cement that purchased for it.
The rest of it will be standard stuff. Ground ball screws and THK ways and blocks.
I am using something non standard for the table though.
It is some 1" Extren plate. It's a glass reinforced material. Here is the description from the seller.
"THIS IS A CONSTRUCTION GRADE FLAT SHEET EXTREN GLASS REINFORCED PLATE THAT CAN BE USED ON A NUMBER OF PROJECTS. THIS PLATE CAN BE CUT, DRILLED, ROUTED, AND MACHINED WITH STANDARD CARBIDE TIPPED BLADES AND BITS."
http://www.norvaplastics.com/misc/images/2212.jpg
This stuff should work very nicely. It is heavy, stiff and oil and water proof.
I will typically just have a vise mounted anyway.
This stuff was cheap enough to try out. It may very well be some great stuff for using in routers or bridge mills instead of aluminum, steel or cast iron.

I would like to see a lot more experimentation will different materials and techniques.

[Geof]

Originally Posted by Stepper Monkey ... I note a strong tendency toward the Tim Allen school of design here.... I found out that quite honestly they are putting in higher cost and sub-optimally performing parts in thier offerings solely because of fears of our consumer resistance to anything different than the "big" mills as being inferior. In other words, our attitudes, or percieved attitudes, are actually hindering our diversity of choices of machines out there, and costing us extra money, for what in the end is lower quality equipment for our hobby. That should make us all think.

I prefer the Red Green school of design but maybe that is just jingoism .

The comment about overbuilding or using outdated technology because of consumer inertia is quite correct. And it pervades all areas; if it didn't all those eighteen wheelers would be hybrid gas turbine electric because it is much more efficent. But it is not as macho as wrestling a ten speed gearbox and making a big diesel growl.

But I disagree a bit with the sticking with old stuff costing extra money...have you priced some of these fancy composites, ceramics, whatever? Better performing machines you may get by going ultra-up-to-date but lower cost I doubt.

[Stepper Monkey]

LeeWay:
That Extren stuff looks cool! Some of the pultruded glass stuff is awesome. I liked the pictures of the building going up using the stuff. Especially the lone guy lifting the huge I-beam into place by himself! Hope it works out for you.

As to what you said about existing methods - "Those that have gone before us obviously have a lot of time and money invested in achieving desired results around a particular budget." - I agree to a point. It's just that those who have gone before us spent all that time and money designing stuff somewhat fundamentally different than what we are using now at home and to a different market. This market is still small enough until recently, and even still, that the hobby CNC companies just aren't big enough to have done much R&D themselves. We are still to a large degree just copying methods of larger machines, that while close to our needs still aren't exactly optimal for us.

Geof:
I stand corrected. Red Green is definitely a better analogy!

When it comes to the ultra-high tech stuff, you might be surprised at the cost saved in manufacturing time. I have held in my hand just this week a prototype slide that has the bearing block and twin rail clamp on each end (I don't know what the technical term is) made from something glass filled - something probably similar to the Extren stuff. The trick was that it was pressure-molded to near final form already, and I'm sure any machining on it was probably just to true the three bores and fast as hell to do. Cheaper and faster than going from aluminum blocks for sure, and no anodization needed. They are CHEAP to make. About HALF their competitors.
Problem is, the marketing guys groaned and said they imagined that thier competitors sales guys and customers would rip it for being "plastic", even though it was probably a superior unit.

[digits]

Well, while I agree with you that progess in materials etc is often shunned, as others have said, it's often because of cost. I have an very nice 6x4" bandsaw which has a very hefty cast aluminium base - and weighs in at a mere 24kg, despite looking a hell of a lot more sturdy than the usual bandsaws. The downside was that it was over twice the cost of an all cast-iron traditional design. To me though, having something light enough for me to be able to stick in a cupboard when not in use was worth the money.

Having started off with a light weight X-1 mill, I really have found the need for a 'bigger hammer' - if anyone had told me that it would take hours to cut out shapes from 1/2" thick aluminium plate with an X-1, I would probably have gone straight for a larger machine.

That said, my X-1's DIY replacement is IMHO far from conventional - it has no scope for handwheels for any manual operation, and is a bridge design with the whole X-axis/bridge moving up and down rather than just the head. It is also almost entirely made of aluminium, which is why it doesn't weigh as much as a Tormach. The aluminium does however cost about as much as two X-3 mills - and may yet need to be filled with damping material. The screws, slides etc are however very solid, traditional items - and rather overspecced in a Tim Allen way. Why? Because the overall cost of larger screws and slides is not much bigger than cost of the smallest ones (when bought new) and having seen my little X-1 flex like a drinking straw, I don't want to take any chances this time!

What I would really like to see for the DIY market is a light-weight alloy/composite based CNC mill that is specifically designed for very high speed, ultra light cuts - e.g. 500IPM but 0.01" DOC with 1/4" mills. The cutting forces would be low enough to allow light-weight construction, but the required speed and accuracy of the screws/slides and motors would be extorionate, unless they were churned out in the same numbers as inkjet printer components. Perhaps we should lean of Epson or Canon for such a mill

[Geof]

Originally Posted by Stepper Monkey .... Problem is, the marketing guys groaned and said they imagined that thier competitors sales guys and customers would rip it for being "plastic", even though it was probably a superior unit.

My first reaction only reading your description is that I would probably "rip it" for being plastic. And this is based on experience making things out of plastic, admittedly not the really high tech stuff, and metals such as steel and aluminum.

I know the manufacturers go on about 'reinforced engineering plastics with exotice fibers and binders' with high strength and high modulus but they are only high either compared to non reinforced plastics or on a specific basis; i.e. modulus or strength divided by density.

When you look at the absolute values even the fanciest composites do not stack up well against many metals. So when you want something stiff and strong, with the emphasis on stiffness as is the case with machine tools and weight is not a factor then I think many times metals are going to be preferable. When weight is a factor probably the composite is going to be preferred.

Also another point about the composites is that they are not ductile and are very poor at accommodating stress accumulation as occurs when things are bolted to them. On practically any structure from a machine tool to a bridge the highest loading occurs at attachment points, bolts or rivets. Many times things are designed primarily for stiffness so the strength is in great excess and it does not matter if the loading on threads and around bolt holes is 10 or even 100 times the average load on the structure. There is so much excess strength the metal can handle this and many times because metals are ductile if one bolt or rivet is carrying an excess load the metal will deform plastically until the load is averaged across several bolts.

I will agree the composite parts can be cheap to make despite the high material cost. Unfortunately this is often the over-riding factor; make them at a lower cost, sell them at half the price of a metal alternative and don't worry that the customer eventually discovers they cannot handle the beating a metal takes and only perform half as well or half as long.

[Stepper Monkey]

Lets talk apples and apples here. When it comes to cost, if that is the sole overriding factor in a design, ANY design in ANY material will of course be compromised. Even in metal - just look at the X1.

Sometimes composites can be the best solution, regardless of cost, just like metals can be.
I love the knee-jerk attacks on composites, they are always predictable; if they cost more you get the cry "too expensive!". If they cost less it turns to "cheap cost cutting measure!" You can't win. Attitudes like this ensure that we will always have the unspoken assumption that any non-19th century production techniques are just shoddy cost- and quality- cutting measures.

I agree within limits of the point about composites being inferior to metals. They are in many ways, usually if you are using composites in place of metals as a substitute in a job best suited to metals in the first place.
There are hundreds of attributes any material has, not just two or three. In cases where the required job is best done by composites, metal can be inferior as well.

In other cases, when a required job can be done with a composite equally well as with metal, why should we care that metal is stronger? If the component is not one of the primary failure points of the unit, then why do we care so much whether a specific component is overbuilt by 4x or 20x? It is irrelevant.

[askman]

if you want high accuracy, composites are pretty much out unless you go hitech, which means high cost. extruded aluminum like 80/20 may be a much better choice for bang/bucks. anyway, high speed pretty much requires ballscrews which is expensive, or belt drive, which is not very accurate.

with mill, you will need to run linear rail, and , you can easily build a very rigid, high speed mill that is pretty accurate.(accurate enough or most of us). definitely good enough for aluminum.

anyway, I just got a 22x24 travel gantry router with 20mm abba rails and ball screws. it is very rigid. (it feel just as rigid as taig or x2) and I am getting .001" accuracy without measurable backlash. I am very much impressed. I am able to run it at 100IPM+ with stepper system.