Seeking informations on Big CNC

[Axel1966]

Hi all,
I'm Axel, from France.
I was searching informations for a DIY Huge CNC, for 2,5D milling on various types of raw materials.
Just landed here, and see a lot of interesting informations.
Thanks for all. And please excuse my bad english.
Cya.

[domino11]

Welcome!

[Axel1966]

Hi all, thanks for the welcoming.
I've read a lot and list all my needs about my final CNC project.
The Mechmate conception match my needs.

I have still serveral questions remaining before to start.
Appreciate I'm graduated engineer in accoustic, not in mecanics. :o

1/ Is light right ?

The Mechmate is based on steel bars.
From chassis to trolley, this is pretty heavy.
Is there any technical benefit of an heavy structure ?
Otherhand, stepper motors may use less torque if mobile parts are lighter.
Last, aluminium profiles may be less expensive in France.
According an aluminium structure offers comparable strenght and rigidity,
is it possible to swap steel for aluminium ?

[WTI]

Heavy is good as far as cutting goes.

You see many homemade CNC rigs filled with sand to add mass and dampen vibrations.

[WTI]

Generally, you need aluminum that is 3 times thicker to have the same bending strength as steel; so you not only would pay more for an aluminum structure (I don't think you will find aluminum cheaper than steel), it might end up weighing the same.

Welding steel with a MIG welder is easy and cheap (anyone can do it), Welding aluminum is much more difficult.

[Axel1966]

Thanks James.
I don't want to weld aluminium (and nothing else than electronics if possible...).
I thought rivets and corners braces was a fast and nice way.

Throught, the MM main longitudinal beams are able to support 10 tons each (22.000lbs).
Isn't it a bit oversided to carry the Gantry and Car ?

Why a lighter profile can't do the job, especialy a tube profile ?

[danilom]

You are only looking it in terms of static load. When the gantry starts to roll there are a whole lot of new forces around, and it needs a heavy table to absorb it.
My gantry tube is 3mm thick, my friends is 4mm, his gantry moves a bit smoother as the vibration from motors is dampened a bit and added mass keeps it firmly on the rails.

[Axel1966]

Thanx Danilo. I understand what you mean.

I thought the gravity system of the V-rails was perfectible, especialy with a light gantry.
Otherhand, the cut of the V-rails seems to me leading to unpredictable results,
event with building the special guiding tool. Small variations may produce a wrong result.
So I was about to invest into a stainless steel trolley-rail system witch is not so expensive compared to the needed v-rollers :
http://www.michaud-chailly.fr/custom/images/modele/gt/b2-cza-3080-zz_d/guidage-galet-economique-zingue-inox-b2_cxx-pmod1.jpg
With the Gantry and Car fixed horizontaly and verticaly, the movement should be as smooth as possible. Appreciate the rails are straight to 0.05mm.

So, the next factor may be, will the structure be stiff enough to absorb vibrations and kinetic energy ?

[danilom]

If you don't like V-rollers or the weight of a Mechmate the whole design will not work for you and you will be forced to make too many design changes, look into cncrouterparts linear carriage (http://www.cncrouterparts.com/extended-linear-carriage-with-abec-7-bearings-p-35.html?cPath=21) and see how they use aluminium for structure.

[bozona1]

This feels like Déja vu.

Axel1966

IMHO, the MM design has been worked to the point where no more changes are necessary. You're trying to re-invent the wheel. Anyway, if you do a bit of reading you'll see that you are asking the same questions that have been asked many times before. Sorry but i didn't mean to come down on you like that. :o...just had to say something. I hope to build one in the near future and I will not change one single thing.

Don

[KenC]

Axel, WOW! Stainless steel trolley-rail system that is straight to 0.05mm!!! Do they come in 3m length & still straight to 0.05mm? Can they take 300kg load?
I surely could use many of those for my store room & retractable canopy

[Axel1966]

I respect your point of view Don. Thanx Danilo, i've checked all the possible drive solutions.

An engineering is an instant vision to solve a problem.
A car is a car. Appreciate the way cars a builded now is not exactly the same than 20 years ago.
One day, the 2006 MM will be out of date. Should we definitely reject evolution ?

An exemple, Some MM are chain drived. IMO it's not a bad idea for many reasons.
First, it works pretty well.
Second, it's unexpensive to find, to build and to maintain.
Third, it's more simple than axis-pignon-belt-gear-rail. Just a chain a gear and two tensors.

I ask questions to the comunity to check principles.
As far i'm not an expert, my questions may be stupid, or not.
Compared to 2006, steel is much more expensive in 2012 (200% here).

I need a good CNC router. I don't have an extensible 10.000€ budget.
Two solutions :
1/ no CNC router = no budget problem
2/ try to find solutions around the MM, to make the build possible

I need to check the relevant possible evolution and minimize the desing changes.
I submit my thinks to the comunity to check where i'm wrong.

Weight means money. Definitely.
Is the weight an indispensable factor or can light but stiff structure can be an alternative ?

[Axel1966]

Thanks for your reply Ken. Sorry, I was writing, I did not read your post before.

300kg load ? Uh ! It's huge for the gantry and car, I'll use a mill not a V8 ! :D
Do you mean 300 kgf ?
According to the manufacturer informations, a four trolley guidage system can hold 320 daN (witch is comparable to kgf within 2%).
I was wrong. The profile is built with a tolerance under 1/10e mm to be exact
the roller trolley is built with a 0.05mm precision (cf. manufacturer).
I hope both can come in reasonable shape.

[danilom]

don't know whats the problem with rack and pinion, here it costed 22eur /m
you can just stick it up with tape and its finished?
chain on the other side needs tension, adjustment and further adjustment over time and maybe even be more expensive.

Steel tube is still the cheapest building material I know, a 6 meter U-profile 180x70mm costs 110 eur, you can cut it, tap it, weld it. Aluminium extrusion in that size and length cost surely more than 2 times that, and with extrusion you need all sorts of joints etc which you can't make yourself easily.

Please don't be discouraged by no means, find the material and parts and compare the prices, my 2 year experience with machine and the people coming over to see it says... you cant build it easier and cheaper.

[KenC]

Axel,
I'm thinking about having shelves to stow my plywood on a multiple drawers, I now stack all my sheets on the same pile, & pulling selected size sheet from the pile is too much strains on my back... 320kgF is ~32 kg... Can even take my weight... useless for my purpose. But I reckon you can build toy desk top CNC with it. Should be cheap & fast to complete.
If you money is short, time is your friend...
Ever consider hand filing your own rack gear from square bars?
I believe no one here resist evolution... as long as it is an improvement & worth the effort... Ever seen a hammer with 2 handles in your local stores?
BTW, ever cost in the time you spent on looking for a "better" "more Economical" solution? Most likely, one will waste more in opportunity cost than that few thousand bucks you saved on material...

Think "Standing on the shoulder of the giant"....

**PS, Even consider using scrap timber for the table & main beams?

[WTI]

Thanks James.
I don't want to weld aluminium (and nothing else than electronics if possible...).
I thought rivets and corners braces was a fast and nice way.

Throught, the MM main longitudinal beams are able to support 10 tons each (22.000lbs).
Isn't it a bit oversided to carry the Gantry and Car ?

Why a lighter profile can't do the job, especialy a tube profile ?



If you want to build a MM, there is going to be welding - there is no way around it. MIG welding is much faster than you can drill and tap holes for bolts and rivets, and a MIG welded joint is stronger than the steel itself.

You can learn to MIG weld steel in a day with a used $500 Miller welder (do they have Miller and Lincoln in France?). Although aluminum can be MIG welded (it requires an expensive spool gun add on and gas), it is usually TIG welded. TIG welding is 3 weeks of classes and a $2500 rig just to get started.

The reason the MM main longitudinal beams are so strong is that with the stresses of the moving gantry and the load of the router being plunged into the material are much greater than you think. You want to hold tolerances to .001" right? The roof on your house is fine until you pile 2m of snow on it...


******************
Today, I want you to call a metal supplier in your town and ask for the same piece of material (lets say 50mm x 50mm square tubing with a 6mm wall, 4m long) in both steel and 6061 aluminum. Report back to us the price difference.
*******************

[Axel1966]

Hi all, thanx for your awnsers.
I only studdy a CNC router for one year, every advice is crucial to me.

@Dalino
I seached a lot for best suppliers here. I don't have the same prices.
UK or Germany are less expensive than France. Serbia looks like even better !
The best price I get for U-Profile 180x70mm is 26€/m while you have 18,5€/m.
Same for the rack you get 22€/m while I get 32€/m.
For the transmission, pignon, 2 pulleys, belt and axis = +-45€ without fasterners... :(

That's why I turned (at this point) into :
100x18mm renforced aluminium structure at 4,5€/m (can be doubled or tripled if necessary)
Chain 5€/m, chain holders 2€ each, 1 pignon 9€, 2 tensors 6€ each, 2 springs 0,40€ each.
Definitely, it's less expensive.

Otherhand, Aluminium is fast and easy to be cut, just drill holes and fix a rivet, it's all done.
No need of grind, degrease, paint... A rivet cost 0,02€ each, plate or corner braces only 0,60€ each...
With steel, I will need an angle grinder (200€) preferabily a metal chop saw (450€), an arc welding machine (80€)...
And just for the CNC construction because I won't use it in the future...
Aluminium must be faster and easyer and not that expensive. The question is : is it a good idea ?


@Ken
"nani gigantum humeris insidentes" I was born near Chartres and Studied there, I well know this aphorism. ;)
That's exaclty what I'm trying to do... I guess it's at last missunderstood.
I apreciate my request seems ridiculous for your eyes. Ok, duly noted.

Among your inacurate comments, this must be corrected :
Kilogram-force (kgf) is equal to the gravitational force on a mass of 1 kilogram.
1 kgf = 9.80665 N = 0.980665 daN
That means you can hold >313 kg (not 32kg) on 4 trolleys, until 9.8 meters per seconds, whitch is fast enough.
Up to 10 plywood plates (2440x1220x18mm), not that bad.
So it can match your needs. My pleasure to help.

@James
Thanks, it's one of the best awnsers I ever had. I understand what you mean.
I know Welding is strong. Parts have to be cuted and drilled too, and maintained in place while welding.
I dunno if I'm able to do it in a single day. Throught, there's finition to do after that.
We have MIG engines in france, starting @ 180€, I guess a good one may cost at last 500€ (660 US$).
Maybe the solution is to rent a professional one.
It's too late here to have the exact prices you requested. (Week End). I'll get the information on monday.

I'm Sure the kinetic forces of the moving gantry and the car can be calculated.
It should be depending of the speed and the weight of the moving parts.
I'm more comfortable with sound reflexions or isolation calculation,
but I'll study that to know exaclty what amount we're talking about ! ;)

Thanx all for your interest and your comments.
I'm still at the begining of the road and your help is greatly apreciated
:)

[domino11]

There are lots of people on the net who have designed their own cnc routers. Some have used MDF, some Aluminum and some steel. If you want all the benefits of the Mechmate proven design, follow the plans. If you want to do your own design to make use of materials more available to you locally you can do that as well. Your end result will depend on how good of a mechanical designer you are.

[Axel1966]

Thanx for your interest Heath.

[Axel1966]

I can not find any information about the gantry and the Y-Car's mass. :(
Of course, it depends of many factors, especialy lenght, accessories and tube sections in use...
No need to have a precise number, just an aproaching value can be enough to have an idea.
Let's plan on 1500mm lenght for the gantry. (as Y-Car's should be closely the same weight for all MM)
Any idea mates ?

[IN-WondeR]

Gantry weight including the Ycar and all options, could easily move past the 100kg mark.

[Axel1966]

Thanks you so much Kim, I make a table with this value :cool:

[IN-WondeR]

Note that I said it could easily move past 100kg... So it's not exactly 100kg, it could easily be more...

[Axel1966]

Yes, I got it kim. ;)

So, a 100kg Gentry+Ycar at 0,170 m/s (400 IPM) makes a kinetic energy of 1,43 Nm.
A 120kg Full Gentry at the same speed makes a kinetic energy of 1,70 Nm.

I wonder if the calculation is right. The kinetic force seems very low... :(
Maybe the MM is able to go faster than 400 IPM...
What should be the good speed to deal with ? (aproximatively, throught...)

[WTI]

Your rapids probably can be a lot faster than 400 IPM, but for cutting, it depends on the bit, the size of your spindle, and the spindle rotation speed.

[Axel1966]

Thank you James.

Into this post, J.R.Hatcher says his MM max. speed is 1350 IPM :
http://www.mechmate.com/forums/archive/index.php/t-748.html
So fast! Even with a good spindle and low deepness, maybe too much for safety IMO.

So, let's calculate again with several values !
Ec for 120kg @ 500 IMP = 2.68 Nm
Ec for 120kg @ 750 IMP = 6.04 Nm
Ec for 120kg @ 1000 IMP = 10,74 Nm
Ec for 120kg @ 1350 IMP = 19.57 Nm
In the formula, it's speed². So speed is an important factor.
Maybe more important than weight.
Throught, these values seems to me low...
(If a Mechanic Engineer may verify...)

Let's see with half weight gantry
(aluminium made, according rigidity must be similar... -45 to -65% weight less usualy)
Ec for 60kg @ 500 IMP = 1.34 Nm
Ec for 60kg @ 750 IMP = 3.02 Nm
Ec for 60kg @ 1000 IMP = 5,37 Nm
Ec for 60kg @ 1350 IMP = 9.78 Nm

Even if the calculation is not totaly exact, with half weight,
aluminium will deal with half less kinetic energy.
Interesting, even if it's not the single factor.

What do you think ?

[Gerald D]

I think you are at the wrong forum - you should go to CNCzone.com instead.

[Axel1966]

Thank you for your interest Gerald, and for all your work and investment here.

A lot of people think about kicking me out, while I just want to understand principes before to start.

Just a question.
Today, the MechMate is 1.x
Did you ever think about a MM 2.0 ?
What could be the future of this marvelous machine called MechMate ?

[IN-WondeR]

Axel

That question have been up many times. The MM community is pretty set on their ways, and by such they are not keen on taking changes and new design ideas into effect.

I have also been one of the few which has been excluded a bit on this site, when I went my own way.

No offense Gerald, I have gotten alot of information on CNC's through this site, and I have put it to good use over the years. So for that I thank you.

[Gerald D]

. . . .I just want to understand principes before to start. . . . .

What do want to start?:
- if you want to start to build a MechMate, then follow the plans
- if you want to start analysis of CNC machines to design your own, then find some people with better experience, that is at CNCzone.

[Axel1966]

Thank you very much Gerald.

I have no more questions except those you did not awnser.

[WTI]

Axel,

I think what Gerald is trying to tell you is that the MM is done being designed. It is already version 2.0 (for a while the second version was being called Mamba, now it simply IS the Mechmate).

All of the engineering of whether it is stiff enough, the motors large enough, the weight heavy enough, the V-rails accurate enough - has already be worked out years ago.

Gerald worked for 6 years to help us all build our own MM and is now "retired".


So if you want to build a proven design (there are already well over 150 MM working and making money), that lets you own a $35,000 router for $5000, just download the plans and build the machine.

If you want to talk about re-engineering the plans, making it lighter (which will make the cut quality poorer), using sprocket and chain (gets filled with wood fibers), using ball screws, or any other ideas that were tested and rejected many years ago; then the place where people like discuss those things is the DYI Router Table group on CNC-Zone:

http://www.cnczone.com/forums/diy-cnc_router_table_machines/


Here on the MM forum people just want to quickly build a commercial quality machine from proven plans - not reinvent the wheel.

[Axel1966]

Thank you very much James.

[Kobus_Joubert]

I personally know of a couple of people that came to my house to look at my MM....then I hear all the KNOWLEDGE poring out...I will change this, I will do that etc.
I warn them and say to them to just stick to the plans like I did.
What happened, most of those start the build, change things and run into all sorts of problems...Bottom line 2 years later they still don't have a working MM.
I am happy with the MM as it is....too stupid to re-design it, and it works for me.;)

[KenC]

Agree with Kobus, we don't have to go far, we are not short of such "knowledgeable" individuals posting in this forum...

[MetalHead]

I am going to have to agree with Gerald and the others on the building of the Mechmate.
If you want to design a machine,this is not the forum for that. We are here to help you build a MechMate and the reason everyone does basically the same build has also been covered. So feel free to read as much as you like, buy a set of plans and get busy building a MechMate. We will be there for you on those questions but you will find this group not to focused on talking about non MechMate systems.

[Axel1966]

Oh, come on !

I was friendly and gentle with anyone here.
I studdy CNC router solutions for more than one year, I did not land here with no reason...
to come.

I said it at the start, I'm not a mech engineer.
Like in all forums, you have people of all kind. Sorry to be of an unexpected one.
There's no offense to try to understand the meaning of a complex structure you're going to build during weeks.

I'll have to build a MM with half the comfortable budget needed.
Unfortunately, parts are more expensive in France than in the States, when you can find them. Those are two very important factors to me.

That's why I started to question each point that could have an incidence on budget.
Starting by the question "Is Light Right ?"
Instead of any pertinent awnsers, I had "do not reinvent the wheel" ; "Go to CNCZone". A bit short to understand, especialy, if your English is not realy good.
So I tried to present arguments to see what could be wrong.

An understandable awnser comes at last in message #32, thx James :
(...)making it lighter (which will make the cut quality poorer), (...)
Good ! So weight is important to cut quality. That's all I wanted to know at this point.

So I'm now working on my table dimensions. I will make and submit a rough as soon as possible, to valid the global shape. I studdy meanwhile "welding for dummies".

According 6 years is old for technologics these days,
I wish we will not have a new version of the MM in a close future...
I don't want to spend months of work for an obsolete or dysfunctional table.

Thanks again for those who gave their help without judgement.
I promiss to explain better (as possible) the reason of my questions,
to avoid bi-lateral missunderstanding or waste of time.

I try to be a suitable MM builder. ;)

[WTI]

Axel,

Like on a job site, we have to pick on the new guy a little bit - just to make sure he is tough enough to hang around with the rest of us.

Looks like you will be fine, and we look forward to helping you build your MM.

=

Do you have a scrap yard in your city? Or Craigslist on the computer? It is pretty easy to find inexpensive I-Beams that can be substituted for the less common C-beams.

Sometimes giving a couple of bucks to one of the scrappers is worthwhile if he'll set aside material that comes in and call you.

[Axel1966]

Many thanks James,
Well I make my searches in this way...
A have a contact with a steel provider who cuts for the industry.
So he have 3000 tons of unselled ends, i'll find my stuff there with a piece of time.
I don't worry with the metal.

I guess I'll rent a good MIG Welder for a week end when all the cutted parts will be ready.
I own a dedicated (not so) old computer. This is not to be buyed.

I guess it will be more difficult to find rollers, morors and controllers cheaper.
We'll see. I'm just at the start and I know I must go slowly. :o

[Red_boards]

Axel,
There are very few MM builders with fewer skills than myself. It can be done. The welding takes practice. There is good info here and on welding forums about welding, but it's all about getting the "feeling" for yourself. One of the nice things about the "heavy" (thick) steel of the MM is that a poor welder can get away with grinding and rewelding (a lot, in my case).

If you're an engineer, then you have the ability to read the plans and figure out the electronics - for me these were the most interesting, but also the most worrying, because that's where the $$ are invested and so you don't want to fry components.

The thing I noticed during the build was that things magically fit together right at the end. If you change something from the plans, you suddenly discover that there is insufficient clearance for a bolt head or something when you assemble things. I've seen this a few times on the forum.

I've also noticed that builders who spend too much time worrying about fancy and "optimised" components often seem to bog down, go quiet, and seldom finish.

I guess I read some threads in the Structure and Mechanics and other parts 10 times during the build. Each time I understood a little more what was being discussed. Also, I found a few local builds to follow (surfCNC and Tony, in my case) because that helped with sourcing components.

[Axel1966]

Thanks Red,
Your comment is very instructive.
For Electronics, I feel confident. At last that reminds me University. :D
Even if I would, I'm affraid to be forced to advance step by step.
For three reasons :

First, I'll start when all will be checked, the whole plans learned and building lists verified.
Second, money because MM is not our family priority, just mine...
Third, I don't have a workshop ready for metal work, so I must invest on the necessary tools (at last a bench mounted drill, correct angle grinder...)

I guess it will be exciting, from start to end. :)

[KenC]

Axel,
1. You must produce the list all by yourself, consider it an examination on how much is your understanding on the plan.
2. Who doesn't worry about MONEY, :mad:
3. You can build with minimal tools. I built mine with a circular abrasive disc chop saw, an angle grinder, a tiny 350W bench drill (which is the cheapest bench drill which I can find for around 40euro) & a borrowed stick welding machine (I've bought a 120euro Chinese 140Amp welding machine since than), a power hand drill & a sledge hammer.

[KenC]

Axel,
Since you are a acoustic engineer, can you suggest a solution to the vacuum pump noise?

[Axel1966]

Thank you for your advices Ken.

There's solutions for the vacuum pump noise, of course. I thought about it.
The most simple may be to put an Electrical Relay on the spindle circuit :
the spindle is turning, the vacuum starts.
The cutting spindle should makes more noise than the vacuum.
If it's a workshop vacuuming station, you may isolate the wacuum in a box,
with specifications to avoid warming and aspiration noise.
Is there a dedicated topic where I could post roughs and schemes ?

[Axel1966]

I awnsered here : Quieting a vacuum system (http://www.mechmate.com/forums/showthread.php?t=3715) ;)

[Gerald D]

You were asking about using aluminium instead of steel to reduce mass, but see what this guy says about mass and vibrations:

. . . . Low frequencies needs more mass and surface to be reduced. . . .

:)

[Axel1966]

Absolutely true... :o

[normand blais]

Salut Axel beau texte sur l'acoustic ,mais ce qui m'impressionne le plus c'est ton traducteur . C'est tu une personne en chair et en os ou un logiciel? Tout est tellement parfait .il n'y a pas d'erreur de syntaxe .J'ai jamais vu rien te tel,rendu la pas besoin d'apprendre d'autre langues . C'est quoi le nom du logiciel de traduction que tu utilise?

[normand blais]

aussi j'oubliais connais tu ce site de cnc . Beaucoup d'info dans la langue de Molière

http://www.usinages.com/

[Axel1966]

Merci Normand, oui je connais le site. J'y suis également inscrit. Très bonne adresse pour l'usinage en général avec une petite section CNC. Ils parlent plus de l'usinage de précision, mais il y a beaucoup d'informations.

Je n'utilise aucun logiciel de traduction, je fais ça avec ma petite cervelle.
Ceci dit, tu es sans doute très indulgent, il doit y avoir largement à redire... :o

[Axel1966]

Hi all,
I move forward in the understanding of the steel construction of the MM.
I wonder if the flattening of tube tube ends (see #13 or #17 below) is structural (to give more strenght to the tubes)
or is it to make the welding more simple ?
How to flatten tube ends easily (without heavy equipment) ? With a vice ?
Another question : my supplyer have various tube diameters starting from 12 to 80mm tubes,
but only 2mm thick, is this ok for the table structure and to be flattened ?
http://i234.photobucket.com/albums/ee35/Jasonmm92/jason-1.jpg

[KenC]

I used hydraulic jack to flatten my pipes. Vice can be use if the pipe wall thickness in less than 2mm.
Sledge hammer can do the job too. Honest.

[Gerald D]

Flattening was only to make welding more simple. 2mm will be okay - you could flatten that with a vice.

[Axel1966]

Many thanks :)
I'll submit soon my table dimensions to the comunity wisdom. :o

[Axel1966]

Hi all,

Well, I bought a welding box, grinder and most of the metal work tools today. Budget 313€.
I'm affraid I started up without any idea of the final budget... :o

I spent all my freetime this week thinking about the main table and wakling around the house with a meter in hand.
I know the initial design is simple and perfect.
Unfortunately, I don't own a 60m² workshop, only a blind room in the cellar of my rented home.
First, that means the MM may be dismantled to move move with me someday.
Second, the wide side must be the Y axis otherwise I won't be able to feed the MM without diging some place into walls.
Third, because of this small room, I'll have to store new wood boards under the MM.

The max working board must be 1220x2440mm.
I dunno if this would be done someday, but I'ts wise to plan a 4th axis improvement to lathe r250mm stuff.

You know all about the specifications... :o

I tryed to adapt less as possible the initial design.
Bad news, mine is of course more complex and will cost a bit more (assumed).
But as far I'm not a mech engineer, I'm might be wrong on many subjects.

Please be indulgent. Here's the rough :
http://giemastruc.free.fr/Forums/MM_01_Perspective.jpg
Front view :
http://giemastruc.free.fr/Forums/MM_01_Front.jpg
Right view :
http://giemastruc.free.fr/Forums/MM_01_Right.jpg
Top view :
http://giemastruc.free.fr/Forums/MM_01_Top.jpg
Possible 4th axis :
http://giemastruc.free.fr/Forums/MM_01_4Axis.jpg
Elements view :
http://giemastruc.free.fr/Forums/MM_01_Eclaté.jpg

Let's explain a bit.
So the MM will be close to a wall corner in the rear Y axis and right X axis. A door opens near the left X axis. I only have two meters open space in front of the Y axis.
The control box and computer will be racked under the left X axis.
I thought the less bad idea was to weld each X axis parts together.
The two "ladders" on Y sides will be bolted on X axis, and several beams (to be bolted) will make the working table and the storage table.
The front Y "ladder" is wide open to help on load/unload.
I did not put the reinforcement tubes on X left, I may adapt triangulations to the control box dimensions "on the fly".

I tryed to be as optimized as possible on materials.
I live in the countryside, and not all the steep profiles are avalaible.
I manage to find UPN instead of C profiles.
The X beams are made of UPN 200x75mm.
The legs and work table are made of UPN 70x40mm.
The storage area are made of Square 40x40x2mm tube.
Reinforcements are made of Round 40x40x2mm tube.
Fixation deck are cuted into a flat 40x5mm steel bar.
As UPN don't have parallel profile, it's impossible to use sandwich bolts.
I thought about tapping the holes, they are at last 8,5mm thick.
I thought about using Allen M12 (X legs) and Allen M10 for beams.

What do you think ?
Before going forward I would like to check all this with the community.
Feel free to reveal my mistakes, I'm bashing proof ! :D

[danilom]

Don't know about that 250mm round stock, will it rack the table, maybe move legs a bit further upfront?
Also, gantry that wide has not been used until, mine is for 2050mm board and I think that's max. Maybe some more additional braces where the X motors are?
My gantry tubes (100x50) are 2550mm 3mm thick and I can sag them by pushing down with hand. They might vibrate at your length.
This really pushes the design to the limit

[KenC]

Is it mandatory for you to cut 1220x2440 board?
I concern more about safe working space...

[Axel1966]

Thanks Danilo,
You're right, my gantry tubes (2940mm - to be confirmed) should be at last 5 mm thick for this lenght to avoid low frequencies vibrations.
A topic is talking about huge Y dimensions : http://www.mechmate.com/forums/showthread.php?t=2083

I forget to say I need (reasonable) precision, rather than speed.

For the additional braces I don't know. The 10.20.451-452 are 5mm thick and could be increased to 6 or 8mm but I don't know the max cut possible with laser cutter.
The experts will say.

[KenC]

Increasing tube height is a more efficient solution, using a 50mmx150mmx3mm or 30 x 200mm if you are brave enough) square hollow section as gantry beams, it improve the deflection factor batter & its lighter too in comparison to 5mm thin 50x100 square hollow section...
What precision are you expecting? 0.0001mm? 0.1? or 0.3mm

[Axel1966]

@ Ken
Well 1220x2440 is the standard board size, available anywhere in France.
It seems to me logical to use this format, especialy I build a MM to work about furniture.
Entertainment units, beddings and cabinets calls large boards.
I wish using those boards would be possible.

[Axel1966]

@ Ken
Thanks for this precisions. Sorry, you post faster than I can reply :)

The larger gantry beams may induce a modification of the 10.20.451-452 parts, right ?
Those are lasercut parts, a very sensible subject in the MM building through.

When I talk about precision, I don't know exactly.
0,1mm will be pretty enough.
The most important is to get an accurate cut, especialy on curves.
The edges must be without defauts.
If the MM shape is good, I guess, it might be a question of motor reduction, rigidity and vibrations...

[Axel1966]

For example, I would avoid this :
http://www.mechmate.com/forums/attachment.php?attachmentid=10225&stc=1&d=1282474513

[KenC]

In my humble opinion, 0.5mm is already "too accurate". In some occasions when I made an mistake or error in dimensions, even 3mm error is not visible.

with a taller gantry beam, youneed to modify 10.20.451~2 and also 10.20.456

The common solution to the chatters at curve are solved in 2 way,
1) Belt-reduction
2) Sand paper :D

[Axel1966]

Yep, sanding means also errors, especialy on curves, but it's unavoidable, anyway.
I guess I must also be very acurate on the build to minimize cuting defaults.
I'm realy affraid about any modification of the laser parts. :(
Any error conducts to problems and/or leads to new expensive laser parts.
As you said, I'm not brave enough ! :D

In this post : What is the max Gantry width allowable ? (http://www.mechmate.com/forums/showthread.php?t=2083)
Gerald says that he don't see big problems about 96" (3m) Y gantry.
Mine should be less than 3 meters (2940, to be confirmed by official plans, plans I must buy next month...).
For a 3,5m gantry, he says the static flex of a 4mm thick 3.5m tube is less than the static flex of a 2mm thick 2m tube, but he was thinking about low frequencies vibrations.

According my MM must have a belt reduction (to be calculated later), what about filling the Y beams to reduce vibrations ?
Does it have been tested yet ?
Polyurethane foam may limit the resonance but won't add significant weight.
Filling beams with concrete will, and reduce considerabily vibrations,
through, even with light weight concrete (styrofoam and nylon fiber charged) the weight can be too much important for motors and other parts.
The gantry manipulation will need at last 4 people ! :D

Well I don't know if it's a real good idea.
I would rather using the right shape with the right thickness...

[KenC]

Well... you can always build one & experiment with the filling. No one know to what degree vibration will start to affect your cut results... so it is really a shot in the dark.

remember to factor in the weight of the filling. never know if it will increase static deflection.

My philosophy... If you are not brave enough to modify the laser files, you should not waste time in customising your own & stick to the "standard" design, or build a smaller machine. or get a bigger work shop to house your machine.

[Axel1966]

I respect your philosophy, Ken.
I would love to have a huge workshop, it's still a goal...
Maybe in the future if the MM project is successfull and brings money back.

Modifying the laser parts path is not impossible. Technicaly, it's even easy.
But any change at this point might induce stunt changes.
I'll do it as a last resort.

I would like to make this project possible with the specifications I have to deal with.
I've read a lot the MM forum. My MM won't be se single that can be dismountable.
I've read also that large Y beams are possible.

I worked a lot on the table at this point.
I have by my side a cutting list optimised for 6000mm steel bars.
Except for the Y beam vibration factor which must be still discussed,
it seems to me as strong as possible compared to the standard design.
Profiles matches at last standard specifications and this is triangulation on each sides.
I wish I did not make too much mistakes on it...
I do not have the approval of the MM community yet.

For the gantry, let's follow Gerald advice and plan on 100x50x4mm beams.
If this profile's ok, I can test the filling method for the community, if it's not stupid.
What is the max weight the motors/rail/rollers can stand ?

[Axel1966]

uh... Anyone ? :o

[smreish]

http://content.yudu.com/Library/A1scv9/CompleteDualVeeCatal/resources/index.htm?referrerUrl=http%3A%2F%2Ffree.yudu.com%2 Fitem%2Fdetails%2F338626%2FComplete-DualVee-Catalog-Rev.-05-11

BWC #3 Vee wheels are rated each at:
382 # Axial
1326# Radial
for loading

The bigger issue is Torque with relation to drawbar required to accelerate, maintain and decelerate with respect to the mass.

I know from experience, that a 77" long gantry from 4 x 2 x 1/4" Steel Mechanical Tube is just fine with 7.2 OM motors. Average top speed of 325ipm, nominal 180-225ipm for most work. This is with a 40# extended Z-slide in place and spindle. SO, my machine "load package" in the y-car is about 60# heavier than all the rest.

Does this help?

Sean

[bozona1]

Axel

Would you be able to load material over the rails, with your gantry parked at the end of your x? That way you would have a significantly less massive gantry?:rolleyes:
Just a thought...

[Axel1966]

Hi all, thanks a lot for your help.

Don, I thought about that, this could solve a lot of problems,
but the overall access to clamp parts and boards would be quickly unbearable...

Sean, thanks for the link and data. It seems the rollers can stand much more than they do usualy. But not the stepper motors.
you must have a very heavy gantry. I guess the cut is nice with this weight.
With 1/4" thick, your two 77" beams have a total weight of 106 pounds (1/4" = 6mm).
My Y beams must be a bit less than 116" long.
To stay in a close weight than yours, I should go to a 5/32" (4mm) thick tube.
It's the specifications Gerald gave in the topic about long gantrys.
So filling the tubes with concrete, is a bad idea because the motors will not stand that extra weight.
Even 5 or 6mm (16/64" or 1/4") thick tubes must be too heavy.
62kg (135 pounds) for the 5mm and 72kg (160 pounds) for the 6mm. Hum.
I wanted to order those. Hopefully, you warned me before.

Do you have a belt reduction Sean ?
Do you think a belt reduction can compensate an extra gantry weight ?

A new question for all, did you bought hot or cold rolled steel for your MM parts ?
I quess the best is hot-rolled steel but is the difference important for the MM ?

[smreish]

As I mentioned before, I am using the 7.2:1 Gearbox motors from Oriental Motors with 20T pinions. The highest reduction possible for the best resolution possible with most considered.

I am not a speed demon, but the machine performed well with this mass. I never pushed the machine beyond 350 on quick moves because of the added mass.

[Axel1966]

Thanks Sean, I did not get 7.2 was for the reduction, not for amperage.
Sorry, newbie inside.

Some people says "long" motors have more torque, is it right ?
At this point, I don't understand a lot about the steppers, are your "short" OM PK296A2A among the most powefull nema34 ?
Could other steppers be even better for heavy gantries ?
Otherhand, Gerald said that too big motors are not adapted to the MM...

Othewise, I searched on the forum and did not found if steel must be hot rolled or not...
I'll take Hot rolled steel anyway, it's more expensive but I guess it's better.

[Axel1966]

Hi all,
I've modified the table base according to comments above.
Can you tell me what's wrong or if it looks ok for your wise eyes ?
Thanks. ;)


Here's the rough :
http://giemastruc.free.fr/Forums/MM_02_Perspective.jpg
Front view :
http://giemastruc.free.fr/Forums/MM_02_Front.jpg
Right view :
http://giemastruc.free.fr/Forums/MM_02_Right.jpg

[danilom]

You solved nothing about racking when using indexer with that cross bearer. it should go straight to the ground to stabilize it.

[Axel1966]

Thank you for your message Danilo,
I get what you mean.

While 4th axis is not implemented, the Gantry will not have to go so far.
I'll fix the proximity sensors and the stopper near 1220mm for now.
I thought those cross bearers beeing here temporary.
And I had 1300mm left on my ⌀40 6m tube.
They might be of no use for now and for later... :o Hum, I'm sorry about that...
Maybe a plan will explain better than a simple view :
http://giemastruc.free.fr/Forums/MM_02_Top.jpg

[KenC]

I don't think using C-channel for the cross beam which support the cross-bearer is a good idea. I will strongly recommend 200x50x3mm RHS for the job.
Also, it will be better if you place the cross bearer on top of this 200x50x3mm RHS, it will be structurally more efficient & a lot easier to fabricate.
I sense that you want are building your machine for mainly storage cabinet because you removed all the critical supporting trusses from the table... You gotta make up you mind whether you want your table structure to be a storage cabinet or to support your MM mechanism. My philosophy says a multiple purpose equipment will never perform a single task perfectly because of all the compromises you design into it. Like all thing in life, you can't have it all...

[Axel1966]

Well your comment is full of sense ! Thank you.

I thougth storage in the table will add mass which may stabilise the basis.
But if there's some structural problems, I should reconsider.
To strenght the long table beam, I thought about a cable straightener (like in architecture), but I left it because it's not cheap and complicated to build.
The stock may be simply under the table, on a simple frame.

Aren't the 70x40mm UPN chanels in Gerald's original maps ? (I did not bought them yet, next month, I promiss :o)
Well with 8mm thick, UPN seems to me easy to tap and bolt.
Maybe It's possible to bolt the 200x50x3mm RHS, passing through them with long bolts.
I'll try to find if I can find some, and ask about their price.

Thank you for your help.

[Gerald D]

Axel, I am not going to spend time analysing your design in detail, but I can "feel" that it is quite weak. I estimate that if I sit on the front edge of your design, it will go down at least 2mm, maybe 5mm.

You can use your design, but you will have to reduce the cutting speeds to stop the system from vibrating/resonating and giving you a poor cut quality.

[KenC]

Axel, 3m is a long span any sort of structure beam...RHS will be lighter & more rigid. With this sort of extended span, I usually reduce the beam weight as far as I could to reduce sagging due to its own weight. You can add weight else where if you wish.
Cable suspension won't really help here as you load will be moving along the beam & the cable suspension will deform the beam differently when the car moves.
IF you wish to enjoy the easy work on UPN & have plenty of weight; Brude force method is use the same UPN which you use for your rail support to support your cross bearer. I can be very sure that will work.

[Axel1966]

I understand Gerald.
Thank you for your regard, the time spent, and your advice,
I must reconsider anyway. The table storage is a real bad idea.
And last, it's more expensive. Let's do simple and efficient.

[Axel1966]

Ok thank you ken, I'll take your advice under consideration.
Thanks.

[Metalsmith]

I can send you a link to a app that will print out wrap-a-rounds for round tube....it is metric and you put in size of tube and angle anf it will print a pattern .... You want stiffness and light, think The Eiffle Tower ....trianglate, make girders !

[Axel1966]

Well, I think actualy to oversize materials, but I love experiments,
I'm interested in your link. Thanks.

[Metalsmith]

http://www.ihpva.org/Projects/tools.htm tubemiter.exe ..... it's at the bottom of page, there are other apps that might be used

with this , a hacksaw and files, you can miter any tube !

[Axel1966]

Thank you very much, I will check it out. :)

[Axel1966]

Very useful for a dumb like me ! Thanks
http://giemastruc.free.fr/Forums/Tubemiter.bmp
A permalink for the soft from my server : Tubemiter.exe (http://giemastruc.free.fr/Forums/Tubemiter.exe)

[Metalsmith]

I made motorcycle frames with a 1" electrical conduit bender ( Same as 1.25" crew tube), die grinder, hacksaw and the miter program.... ; )

[KenC]

Tig?

[Metalsmith]

Tig? :confused:

[WTI]

He is asking you if you TIG welded your motorcycle frames. You know; "stacking dimes".

http://www.amafineartgates.com/

[parrulho]

Hi Axel, you did a good job on CAD software. What program are you using? The renders looks very real.

[Axel1966]

It's not some CAD softs, just graphic ones.
No precision on parts or dimensions, just to make up my mind with the MM building process.
I'm using Gmax with a Vray renderer. :)
Both freewares.

[Metalsmith]

He is asking you if you TIG welded your motorcycle frames. You know; "stacking dimes".

http://www.amafineartgates.com/

I tack with mig and if I am selling , I have it professionally welded, Tig is too time consuming .....I am playing with silicon-bronze, like the Brit frames are done...or where !

[Axel1966]

Hi all, new progresses (so far...)
Today the plans of my four belt reduction systems :
3 different parts, 4 identical for motor support, and 2 different suspension plates,
one (x3) for both X and Y axis, 1 for the Z axis.
I'll give the DXF to be lasercuted with the original MM parts.
No weldind needed, just some glue for the bearings,
and 6xM6 Allen Head Cap Screws with 6 steel tube bridges by gearbox.
I'll make only the axis by myself using my drill as a mill. :rolleyes:

http://giemastruc.free.fr/Forums/BeltReduction.png

These little things takes a lot of time !
At last this is out my "to do" list. :p

[Axel1966]

Because I'm lazy, I did not calculate reduction ratio by myself,
So I came to www.sdp-si.com (http://www.sdp-si.com) and entered the desired reduction in the fields :
http://giemastruc.free.fr/Forums/Poulies01.png
And eyes wide shut, I used the calculation produced by the table.
But if you enter 3 to 1 you get a 2:1 reduction :mad:
So beware if you use the calculation table of this website.
Choose the motor pulley (A), then multiply the number of grooves (A) in the charge pulley (B) by the desired reduction.


4 hours spent for nothing.
I should have calculated by myself.
Nice, I've got to start over... :(

[twistedfuse]

Try this program made by brecoflex (http://www.brecoflex.com/index.php?CATID=9) for working out belt lengths and pulley centres in reductions. Works with all standard types of belt/pulley profiles. Seems to always want the latest and greatest version of java installed in your browser but is a nice piece of software. Their prices aren't too bad either.

Daniel

[Axel1966]

Hi Daniel, thanks for your link, the program is much better, and have much more pulley types ! :)

[twistedfuse]

No worries at all, keep up the good work.

Daniel

[Axel1966]

So !
I've spent a loooot of time on belt reductions.
To balance corectly the weight around the M10 fixation,
two types of plates are needed (IMO) one kind for X & Y, one for Z.
I've checked all three time (my last try was pittyfull...),
from dimensions of each part (avalaible in the neighborhood), to belt tension...
I've made some 3D views to be sure all works well together, including boltery.

http://giemastruc.free.fr/Forums/BR3-1_001.png
http://giemastruc.free.fr/Forums/BR3-1_002.png
http://giemastruc.free.fr/Forums/BR3-1_003.png
http://giemastruc.free.fr/Forums/BR3-1_004.png
http://giemastruc.free.fr/Forums/BR3-1_005.png

I hope this is ok now. I've made some DXF files.
Parts are meant to be cut in the same time
than original MechMate laser parts.

I keep on recieving the tools needed, step by step.
This week : quality taps and files ! :)

[baseball43v3r]

hey axel, Mike sells plate reduction kits that take the guesswork out of it. But I do like what you have built here, what are the silver parts sticking up to the right where the motor mounts?

[JamesJ]

Belt tensioning device?

[Kobus_Joubert]

French flair... nice curves on the plates.

[Axel1966]

Hi all, thank you for your comments.

@ John,
Well, I agree Mike's parts fits perfectly.
Through, shipping cost to France plus custom fees may at last double the part's cost.
Unfortunately... :(
Otherhand, it was fun to create my mine ! (and teachfull)

@ Jim,
You get it Jim, it's supposed to help the belt tensioning/positioning by pushing/pulling the motor.
Otherhand, it cross the whole box to avoid twist and add stiffness too.
I researched the simplest way, without the need of some bending/welding tools.
You just need glue, a saw for the tube bridges and a M6 tap.

@ Kobus,
Thank you, cultural distortion :o

When builded, if it works without any problem,
I'll post the DXF files in the "belt reduction" tread, if it's of any interest for someone,
who don't live in Mike's neighborhood.

[Pleyell]

Je viens d'acheter les plans et je t'avous qu'avec mon anglais, je suis paumé complet. Si tu peux m'aider pour déja savoir par quoi commencer je suis prêt a une aide financière pour te remercier. Merci
Pascal









Hi all,
I'm Axel, from France.
I was searching informations for a DIY Huge CNC, for 2,5D milling on various types of raw materials.
Just landed here, and see a lot of interesting informations.
Thanks for all. And please excuse my bad english.
Cya.

[Axel1966]

Salut Pascal,

Inutile de poster sur tous les sujets et tous les forums ;)
Une aide t'a été proposée sur un autre forum si je ne m'abuse.

[Pleyell]

Salut
tu sais Axel, métabricoleur a l'air un peu complexe pour un gars comme moi qui sais pas faire marcher une montre a quartz. Je peux mm pas t'écrire parce que j'ai pas écris assez (sur Métabricoleur). Avous que ça a l'air un peu con cette histoire! pour un clown professionnel comme moi, c'est le" bout du bout de bois" de raymond devos. Je suis obligé de t'écrire ici. Comme le 9 a Asnières de fernand Raynaud; je cherche un mec qui m'aide et moyennant qu'il me dise comment faire , je donnerai qq émoluements comme disent les avocats. Merci de me répondre ,via les Usa.Pascal

[Fox]

http://www.ihpva.org/Projects/tools.htm tubemiter.exe ..... it's at the bottom of page, there are other apps that might be used

with this , a hacksaw and files, you can miter any tube !

I used this a lot in the past ( not MM related - mine is all square tubes), it's a fabulous basic program, just works. I print it out ( check 1:1 printing scale ! ) cut paper at edge tightly wrap around tube, use permanent marker to follow the cutting line, remove paper, take handheld grinder ( files and hacksaw are way too slow ) and with a bit of practice you get 'perfect' MIG joints in a matter of minutes, if you use the handheld grinder with a cutting blade and work your way ( cutting/grinding ) until no black marker is visible anymore. I do this freehand, tube in left cutter in right. Maybe not the safest way (please watch your fingers), but when you have a good feel ( practice a little ) it's dead easy + quick and relative safe.

[Axel1966]

Pascal,

Essaie de prendre quelques jours pour digérer les plans.
Beaucoup de choses prennent du sens en multipliant les lectures.
Si tu as des questions entre temps, n'hésite pas. ;)

[jayrdwein]

If you want to design a machine,this is not the forum for that. We are here to help you build a MechMate and the reason everyone does basically the same build has also been covered.

[Axel1966]

@ jayrwein
You're digging out a pretty old topic ! I wanted to build a MM, but the MM needs welding skills I still don't have. I builded a very different machine two years ago. So don't worry, I know what's the MM forum is all about. Through, even if I'm not a MM builder, I helped a lot french people to understand the MM, that I promoted. Appreciate the MM forum is obviously english, and if you're not fluent in technical english you may feel lost.

Otherhand, in every forum, every topic has been covered many times. But, for a new comer, informations are spared everywhere, and hidden inside a huge amount of subjects. You can't find what you need easily, even with the search engine. Newbies should be welcome to ask questions that was many times covered. IMHO.

Cya.

[thuanle]

Hi Alex,
I like your concept at post #73 (except the storage). Did you build MM with this concept?

[Axel1966]

Hi thuan,

I never builded any MechMate, because I can't weld properly. The design on post #73 may cause problems along the gantry. A gantry long as that should be stiffer than on MechMate standard recomanded profiles. I guess even the bended plates at the ends of the gantry should be recalculated too.

BUT, I built another kind of large CNC (weldless) with the same principle for a rotary 4th axis. But instead of having it on the long side I decided to put it on the short side. A 1300mm wide area is enough for many purpose. A smaller gantry is better and is cost effective. Remember this is CNC, so you've got a super power : repeatability. You can build any long piece out of shorter ones that will joint perfectly. So I believe that was a wise decision.

[thuanle]

Axel, With your concept in post #73, operator could easily load 48'x96' sheet size-way and easier to clamp. If the MM is put in a residential double garage, we don't need to open the garage door in order to load the sheet on.

Could you share the picture/concept of your built CNC? Thanks.