PDA

View Full Version : Gantry design question



D.C.
06-10-2012, 02:28 AM
First post and I'm intending to start my first cnc build soon, I'm starting to design at the moment but something is puzzling me.

Having a very rigid gantry seems to be crucial but all the designs I've seen seem to try and accomplish this by adding huge amounts of material and weight to the gantry which then means you have to start upgrading or doubling up on motors etc.

Has anyone done a gantry design by laminating aluminium honeycomb between either aluminium sheet or carbon fibre?

It would seem to provide a far more rigid solution for any given weight when compared to the mdf/plywood/aluminium builds, am I missing something blindingly obvious or would this be worth pursuing?

The cost shouldn't be that excessive if you look at the cost of honeycomb from easycomposites.

Aluminium Honeycomb Core - Stock Sizes to Buy Online - Easy Composites (http://www.easycomposites.co.uk/Category/Core-Materials/Aluminium-Honeycomb.aspx)

JAZZCNC
06-10-2012, 03:13 AM
It would seem to provide a far more rigid solution for any given weight when compared to the mdf/plywood/aluminium builds, am I missing something blindingly obvious or would this be worth pursuing?

Not quite blindingly obvious but this is one area lots don't fully under stand untill after they have used or seen the difference a heavy gantry makes.?

MASS makes a great deal of difference to the quality and depth of cut machine can make.! . . Yes a heavy gantry does slightly restrict rapid speeds compared to a light weight gantry but the reality is that most machines spend most of there time cutting not rapiding around. (Some exceptions do apply like drilling lots of holes)
So therefore the heavy stronger gantry with it's Mass can take far deeper cuts and still give a good finish and gets the Job done quicker than a light weight gantry can due to cutting deeper so less pass's.!
The Mass dampens vibration better and vib's transfer to the finish so again better finish.

Don't get caught up in the Rapid speed light weight race it's mainly for bragging rights and very little to do with useful cutting times and quality finish.

Also the doubling up of motors has very little to do with power to move a gantry and everything to do with ease of use and turning 2 screws to allow wide gantrys not to Rack or crab which they would with a single centre or side driven screw.
It's very possible and is often used to drive 2 screws with a single motor connected via timing belts and infact my machine use's just that setup. Timing belts are very accurate and efficient regards power usage and backlash etc so don't be put off using them.

D.C.
06-10-2012, 06:18 AM
MASS makes a great deal of difference to the quality and depth of cut machine can make.!
The Mass dampens vibration better and vib's transfer to the finish so again better finish.

Thanks for the reply, I was looking at carbon fibre precisely because of concerns about vibration, CF is insanely good at dampening that and a CF honeycomb panel won't flex at all until you stress it enough to induce catastrophic failure like when an F1 car meets concrete at 200 mph or the four horsemen of the apocalypse ride into town after a bad day at the office. It's why the 'budget no limit' aerospace & motor sport guys use it.

Is the extra mass adding momentum that helps the cutting process, or is it just eliminating vibration? Given the number of minute and rapid changes in direction a CNC machine must go through it seems counter-intuitive to add mass but I really do know nothing about CNC... (Damn it, where is the 'clueless dimwit' emoticon when you need it!) I kept coming back to that design direction especially after reading papers like this:

http://www.iaeng.org/publication/WCE2012/WCE2012_pp1941-1947.pdf

irving2008
06-10-2012, 09:14 AM
Interesting paper, but the structure they refer to in there, if I understood it correctly, is actually a support column not the moving gantry itself. While the paper starts off talking about the gantry all the study work is on the column design. So the jury is still out on whether a lightweight CF gantry beam is better. However, the information is useful as regards the gantry supports. The issue however, is for the small sizes we generally build, i.e. a gantry side 300 - 500mm high or so, does the complexity of forming such a structure give any benefit over a slab of 20mm ali?

JAZZCNC
06-10-2012, 10:46 AM
Is the extra mass adding momentum that helps the cutting process, or is it just eliminating vibration?

Bit of both really.!

I've had little contact with CF other than light weight motorcycle components and to be honest other than there light weight I've not been impressed.? The lower grade and quality components like covers, guards and shields etc were brittle and didn't stand friction wear very well.
The high end components like Wheel hubs, suspension parts etc were very expensive and the reason for this is down to the high manufacturing costs thru the need to use vacuum autoclave systems and high grade CF. . . . This really the reason why F1 anything that use's CF costs the earth.

So to me here lies the problem with DIY and CF.!! We don't have the budget for high grade CF and no access to equipment needed to use it correctly and get the best strength and longevity from it.

The honey comb principle I agree with and yes these transfer to CNC no problem and have been for years. IF CF gave any major performance along with even a small cost saving benefit the big boys would be all using it by now and I'm pretty sure there will be a few CF Gantry's, components or full machines stuffed in some corner of some company's R&D department.??

When it comes to Diy then light weight doesn't win over Mass and isn't easier of cheaper to acieve IME.!!

There's always got to be first that takes a gamble so with That said you go for it and we'll all watch learn.!!. . . I for one will be the first to congratulate you if it works and performs has expected. ( I'll do the Nah nah told you so has well. . . Lol)

Wish Good luck if you do.

D.C.
06-10-2012, 04:17 PM
Interesting paper, but the structure they refer to in there, if I understood it correctly, is actually a support column not the moving gantry itself. While the paper starts off talking about the gantry all the study work is on the column design. So the jury is still out on whether a lightweight CF gantry beam is better. However, the information is useful as regards the gantry supports. The issue however, is for the small sizes we generally build, i.e. a gantry side 300 - 500mm high or so, does the complexity of forming such a structure give any benefit over a slab of 20mm ali?

You are right, I stumbled across that while trying to find papers on the use of honeycombs or CF, there seems to be remarkably little available to be honest I was more taking away the general principle of 'honeycombs good if used correctly' rather than the specific application they refer to.




were brittle and didn't stand friction wear very well.

IF CF gave any major performance along with even a small cost saving benefit the big boys would be all using it by now and I'm pretty sure there will be a few CF Gantry's, components or full machines stuffed in some corner of some company's R&D department.??

There's always got to be first that takes a gamble so with That said you go for it and we'll all watch learn.!!. . . I for one will be the first to congratulate you if it works and performs has expected. ( I'll do the Nah nah told you so has well. . . Lol)

Wish Good luck if you do.

CF is brittle and really sucks for friction that's true, but every material has it's limitations hence why we don't use rubber for roof suporting girders. That said, it is a bit worrying that nobody else is doing it, I don't know if it's because it is a really stupid idea or because I'm a genius and smarter than an entire industry full of inteligent professionals. ( I vote for it probably being a stupid idea!)

Hmmmmm.

I think maybe the best thing to do would be to design 2 gantrys, if I can get the cost difference down to 50-100 I might take a punt just to see what happens. I was orignally intending to epoxy the surface fittings on but if I bolt through the structure instead if it all goes horribly wrong I should be able to fix it by adding a couple of thick reinforcing sheets. ( I'll engrave one with 'JAZZ WAS RIGHT!' if I do just to remind me what to do with my next good idea.)

mekanik
06-10-2012, 09:02 PM
You are right, I stumbled across that while trying to find papers on the use of honeycombs or CF, there seems to be remarkably little available to be honest I was more taking away the general principle of 'honeycombs good if used correctly' rather than the specific application they refer to.



CF is brittle and really sucks for friction that's true, but every material has it's limitations hence why we don't use rubber for roof suporting girders. That said, it is a bit worrying that nobody else is doing it, I don't know if it's because it is a really stupid idea or because I'm a genius and smarter than an entire industry full of inteligent professionals. ( I vote for it probably being a stupid idea!)

Hmmmmm.

I think maybe the best thing to do would be to design 2 gantrys, if I can get the cost difference down to 50-100 I might take a punt just to see what happens. I was orignally intending to epoxy the surface fittings on but if I bolt through the structure instead if it all goes horribly wrong I should be able to fix it by adding a couple of thick reinforcing sheets. ( I'll engrave one with 'JAZZ WAS RIGHT!' if I do just to remind me what to do with my next good idea.)

Hi DC
Welcome to the forum
I am really glad you raised this question, I have thought that a CF Gantry had to be more ridgid than all the ally extrusion, when i was @ the Model Engineering exhibition a couple of year ago there was a turbine driven photo graphic aerial platform (Heli Movie Camera thingy) and its space frame was made from 8mm thick CF composite plates.
I can see no logical reason why an exreemly ridgid gantry cannot be constructed using a mould and the vacuum infusion technique.
I made an item just using fibreglass and @ 12mm thick it was both ridgid and had substantial mass.
If i ever get round to building i would certainly give it a try.
Good Luck

D.C.
06-10-2012, 09:37 PM
Hi DC
Welcome to the forum

Thanks for the welcome, I think the CF is a non-starter for me though, I realised when checking the easycomposites website that there must be bug or browser conflict as the prices were not changing when differing thicknesses and sheet sizes were selected. What I had priced up as being about double\triple the cost turns out to be ten times or more expensive.

Ooops... :hysterical:

The aluminium honeycomb from there is still a good price though so I may look at doing a laminate design but it would be with thin steel or aluminium not CF, it's a shame but I'll have to shelve it for a few years and see if the price of carbon fibre comes down.


I did find a similar thread to this one on cnczone and apparently there is one company that do use composites for CNC but it is for an ultra precision machine:
DMG | DECKEL MAHO | GILDEMEISTER - HSC 20 linear | Precision Center (http://www.dmg.com/en,milling,hsc20linear?opendocument)

There is also this company making ultra high end machine components from composites beams with steel linear rails attached (look at the piccy):
Composite Machine Building Components (http://www.compotech.com/example_applications/machine_building_components)


They both use composites for extreme rigidity, as do photographers, astronomers, sailors etc etc so I think the idea is sound but the price needs to lose a zero off the end before it comes into the reach of the dirty plebs.

mekanik
08-10-2012, 10:54 AM
@DC
Nice links
I appreciate the CF rout would be the ultimate material but from my limited experience of working with fibreglass i can see no logical reason why you could not build an exreemly ridgid gantry, the project i tried to construct was a curved concrete mould, but unfortunately i could not get sufficient vacuum on the mould and it resulted in about a 400mm length being correctly cast, however this was both heavy and super ridgid, if you live anywhere near the Barrow in Furness area i would be willing to donate a roll of the FG for you to trial a gantry prototype on the understanding that we have lots of pictures and you keep us all updated, i am convinced this is the way to go, the only downside is the time it would take mould making ect.
Have a Nice Day

martin54
08-10-2012, 01:19 PM
Just out of interest what cloth were you using to work out your costs?? For something like this you would really need a lightweight cloth & this is where the costs spiral, not just the fact that they cost a lot more to start with but the fact you need 2 or 3 times the length to achieve the same laminate thickness lol.
Don't hold your breathe waiting for costs to come down, because there is less demand for certain types of cloth less of that type of cloth is woven so costs will always remain that much higher.

D.C.
08-10-2012, 08:32 PM
I wasn't planing on using cloth except maybe as a cosmetic wrap.
Carbon fibre also comes in ready made stiff boards and box section, you can laminate the boards/box to the aluminium honeycomb without needing to use a vacuum so it makes life a lot simpler.

Thanks for the offer mekanik it's very kind of you but I'm afraid I chickened out and decided to go for cheapy MDF job as my first build and then get more adventuous in a year or two.

After I've got a machine up and running I might buy enough to make up 50 cm beams of multiple different materials so I can test how each behaves side by side. Seems like a cheaper way of getting a definitive answer to the 'will it really work' question.

martin54
08-10-2012, 10:30 PM
If you are just gluing a flat sheet of carbon fibre to the honeycomb rather than laying it up I would have thought you would lose a lot of strength, the box section they do is all unidirectional cloth which means all the strength is in one direction which could give problems. Like you say the only way to know for sure is to try it lol.

If you are going down the MDF route then try to make sure the components you buy can be migrated to the next build, especially expensive stuff like the electronics. That is what I have been told by some of the forum members.

mekanik
09-10-2012, 12:06 AM
@DC
Have to agree with Martin,to get a ridgid structure i would make the mold and include webbing details then vacuum infusion,i suppose you could use the honeycomb to provide stiffening and to save the intricate mold work of the webbing detail but it still needs to cast in one operation.
Just out of interest what width were you thinking of making the Gantry.
regards
mekanik

D.C.
09-10-2012, 01:02 AM
I was considering it for a 4x8 build.

I'm not sure you would lose that much strength, plenty of people make torsion boxes using only glue for joining for example. The vacuum process is really to squeeze the different layers of cloth together around a complex shape, when you are dealing with a ready made flat sheet you can achieve the squeezing the old fashioned way with weight or clamps. If you were to try to make the gantry in one piece then yeah you would need a vacuum to ensure good contact in all areas.

martin54
09-10-2012, 02:39 AM
I was considering it for a 4x8 build.

I'm not sure you would lose that much strength, plenty of people make torsion boxes using only glue for joining for example. The vacuum process is really to squeeze the different layers of cloth together around a complex shape, when you are dealing with a ready made flat sheet you can achieve the squeezing the old fashioned way with weight or clamps. If you were to try to make the gantry in one piece then yeah you would need a vacuum to ensure good contact in all areas.

Sorry but that's not true at all, I could quite easily manufacture a complex shape without the use of a vacuum just by using the right weight cloth and simply laying the cloth around a mould or covering the part if I was just skinning it. The carbon fibre part that I made would look just the same as a part made using a resin infusion system but that's where the similarities end. The part produced by resin infusion would be far lighter & stronger than the part I made, what the vacuum actually does is remove all the excess resin which cant be removed from a traditional type wet lay up. The strength comes from the layers of cloth being closely bonded together, the less resin that is present the stronger the bond.

mekanik
09-10-2012, 09:27 AM
Hi Martin
You beat me too it.
OCLV as per my Trek frame, when i put the vacuum on the concrete mold that i attempted i was gobsmacked @ just how compact the the laid up glass went then when it's infused its far superior to any hand lay up ( i ended up having to lay it up by hand eventually as i couldn't get the mold vacuum tight)
@DC
Sorry mate i must have misunderstood your intention, i thought you wanted a maximum ridgidity gantry using composite construction, if this was the case it would need to be one piece.
If i ever get the time i might design a 4' gantry and make a mold,as it's a good size it might be possible that other members of the forum might find this usefull.
Have a nice day guys
mekanik

Tenson
09-10-2012, 12:51 PM
I think DC is talking about using pre-made ridged sheets. In that case they are already good and strong and he just needs to join them at right angles or whatever. Okay it might be a little weaker at those joins but he can probably do well with traditional methods like bolting each panel to some square section + resin.

martin54
09-10-2012, 02:17 PM
I think DC is talking about using pre-made ridged sheets. In that case they are already good and strong and he just needs to join them at right angles or whatever. Okay it might be a little weaker at those joins but he can probably do well with traditional methods like bolting each panel to some square section + resin.

Yes from a previous post he said using pre made sheet or box section to sandwich the aluminium honeycomb. The premade sheet will be good quality & strong so may well do the job but not as strong as if made as one complete unit. You could actually make the gantry in one quite easily using traditional wet lay methods, might even work out better with the increased weight but wouldn't be as strong.

D.C.
09-10-2012, 03:06 PM
Yes I was talking about using a premade sheet and strength wasn't really a great concern although I do agree that doing the whole thing as a complete unit would be stronger.

All a gantry has to do strength wise is hold it's own weight, the weight of the spindle + it's fittings and resist relatively small amounts of lateral forces and the occasional shock load, a single piece of 6mm plywood can easily cope with the strength requirements without failing.

The problem is rigidity and coping with those forces while still being in the correct position all along it's span to the nearest thousandth of an inch. Which is why I was thinking of CF sheets with the honeycomb acting like a torsion box, the strength of CF isn't needed but it's insanely good stiffness & vibration dampening properties are.

Tenson
09-10-2012, 03:14 PM
I like the ingenuity but I think it's kind of a waste of expensive material. CF is very stiff for its weight, but steel is equally stiff. Since the weight isn't much issue, use steel. Want damping, then use steel box section with foam or sand inside. For sheets of aluminium or steel use constrained layers (I mentioned green glu before, bitumen and butyl rubber sheet is also good).

I design loudspeakers and in this area we pay a lot of attention to damping sheet materials for the speaker boxes. The BBC did a great amount research on this and concluded that constrained layer damping was most effective, although making it stiff was not the goal. So IMO use stiff materials and constrained layers, but don't worry much about weight.

D.C.
09-10-2012, 03:25 PM
I like the ingenuity but I think it's kind of a waste of expensive material. CF is very stiff for its weight, but steel is equally stiff. Since the weight isn't much issue, use steel.

That is pretty much the conclusion I came to, CF has been used for CNC machines but only in the ultra high precision ones and I'm guessing the thermal expansion of steel and it's momentum put the CF ahead in those situations. For the size of those machines the labour cost is probably far higher than the cost difference in raw materials.

D.C.
09-10-2012, 03:27 PM
Having said that if the price of carbon fibre comes down I would still consider it just because it looks so damn good! :eagerness:

Tenson
09-10-2012, 04:00 PM
Spray paint your CNC machine then ;)

D.C.
09-10-2012, 07:24 PM
Damn it, why didn't I think of that?

martin54
09-10-2012, 09:53 PM
That's a great idea Tenson, any idea where I can buy carbon fibre paint ??:rolleyes:

Jonathan
09-10-2012, 11:03 PM
That's a great idea Tenson, any idea where I can buy carbon fibre paint ??:rolleyes:

Here (http://www.ebay.co.uk/sch/i.html?_from=R40&_trksid=m570.l2736&_nkw=carbon+fiber+vinyl) it is.

martin54
10-10-2012, 12:21 AM
That's vinyl Jonathon not looking for vinyl was hoping to be able to get a tin of spray paint.

D.C.
10-10-2012, 12:31 AM
No pleasing some people...

Spray it black and lumpy, it will pass form a distance.

martin54
10-10-2012, 01:40 AM
No black wouldn't look right, would need to be carbon fibre, looking for tartan paint as well if anyone knows where to get that.

Tenson
10-10-2012, 03:02 AM
Obviously you need two paints; one for carbon and one for fibre. You also need a mask to get them in the right patten.

mekanik
10-10-2012, 10:11 AM
Use one layer of the real stuff,epoxy and vacuum bag it,job done