[Texasbowhunter]

Well Peter As You may recall I wanted to build a steel framed base sometime last time we spoke...Some would have to be welded which would have applied to much heat to the frame then with the stress relieving I was told I would require I kind of put that on hold...I felt there was possibly something else out there that would do more with less outside involvement which I would have no control over...
So, after stumbling upon this DIY Concrete Version, I posted earlier I knew that If I created a Mold in which I could get done most accurately I would be ahead of the possibility of out of square or flat...
The router would be in the garage where there was no climate control for the temps and with the steel frame it would be subject to various temp changes and then the inherit rigidity associated with this...
It certainly appears that there's rigidity incorporated with XYZ_spec's build using this concept...and as I have seen there's a lot of folks that are adding the E/G to their units to add the rigidity to improve on the various aspects of their desire for the CNC to perform to their desired wishes and expectations...
So, I want rigidity foremost and a material that's not subject to change when the weather changes...I don't want to go back and add material in the hollows/cavity's trying to add the rigidity later because I under designed it early on or after having that design in a steel frame regret not figuring out a way of achieving the better base to begin with...
I looked and watch all od Piotr's build...OMG is that impressive, then there's RouterCNC on the Utube and has built quite an impressive unit...Over engineered perhaps but none the less the engineering that went into that was impressive...Then you have Bryan Howard who built a PrintNC and although it was very well built, in which he spent some time ensuring all was correct and square he's fabricating additional designs into his build to make it heavier to ensure he has the rigidity he desires...
I don't want to under design and then redesign later...
Hope this relay's well,
Paul

[peteeng]

Hi Paul - When building a machine you need to summaries your resources and needs. contract machining, heat treat, painting, welding, comfort level with various technologies, etc etc including $$$$. Then find the pathway thru that puzzle that realises the machine you want. There are still hurdles with a cold cast machine.eg how do I deal with inserts and flatness etc. mainly to do with how accurate does this machine have to be? If your aiming at +/-0.1mm then a plywood machine or sheet metal machine or extruded aluminium machine is totally doable./ If your aiming at +/-0.01mm then you have to step up in quality of parts, machine all the important bits on a big VMC or manual mill. I think a grout cast machine base is a great thing, Plus I'd cast the saddle and the gantry... Consolidate as many parts as possible into one.

Machine rigidity is a big topic - If you have access to FE on your CAD its very helpful.. You describe the machine as a router but do you want to cut aluminium? as this lifts the bar quite high. By the way grout has the same COE as steel so it will expand and contract like steel. So do you have an exemplar machine in mind? This helps as then your not starting from scratch, they have been thru the movie a few times so have sorted some basic issues. Keep at it. I think its time you got some CAD models rolling what CAD do you use? or no CAD? Peter

[peteeng]

Hi Paul - This is what I have learnt in building 5 routers:
1) The high rail design is considerably stiffer than a column design. It also contains dust much better and the mechanics are up out of the big dust. Highly recommend a high rail design
2) My next design will incorporate a dust enclosure. Its difficult to do this afterwards. I do suggest you consider it at design stage. Two things annoy after a while dust and noise. A water-cooled spindle is as quite as you can get
3) Sounds odd but the z axis does drive the footprint of the machine. Its worthwhile spending time on this before you get too serious with the rest of the machine, Under done Z axes show up immediately and its too hard to fix this. Once you have the z axis super uber stiff then work outward to fill out the machine. if you work from the frame to the Z axis you will run out of space to make an uber Z axis...
4) take advantage of height. make everything as high as possible as this adds stiffness without affecting the footprint. Make everything as big as possible. Don't shape things so they reduce in size, take advantage of the entire space envelope available for the part. This maximises rigidity
5) If using tubes make them thick. Light tubes may look great on paper but cnc machine parts are shear stiffness critical. Light tubes are great for global bending but poor for local bending and shear stiffness.
6) Hard to beat a steel machine if you have good welding and access to stress relief. Easiest & cheapest way to gain stiffness is to use steel
7) Do detailed CAD models explore possibilities don't buy anything until you have designed it twice and are happy. I still have things I got excited about and bought but turned out to be not right... if I had done the CAD work and contemplation it would have become clear
8) Keep at it and don't take the easy path if you feel something could be better. The easy path usually has consequences and its the design time that makes a great machine vs an average one. The design sets the DNA and once set its difficult to improve

Thats about it - good luck keep at it. Peter

[ardenum2]

CAD can't be underestimated, it took me 3 years of CAD just to know what type of a frame I'll use. I have found through a lot of pain how hard it is to do a stiff Z axis. Most of the restrictions come from the materials you're gonna use.

There are 2 main ways to do this type of a Z axis. One you make the saddle long and mount the carriages to the spindle box, which becomes short. Two you make the spindle box long and mount the rails to it, the saddle is short. I prefer the first approach because no matter the position, stiffness remains the same. The second one, the further you extend the less stiff it is, though it is capable of reaching further down.

[peteeng]

Hi Paul - How big is the machine? and what sort of work do you want it to do? Designing a general purpose machine is tough tying to cover so many bases but if it has a purpose it makes design easier...Peter

[Texasbowhunter]

Hello Peter Hello Ardendum...

Hi Paul - This is what I have learnt in building 5 routers:
1) The high rail design is considerably stiffer than a column design..It also contains dust much better and the mechanics are up out of the big dust. Highly recommend a high rail design...
AGREED and this is one of the reasons for considering this Concrete build...So with this said I'm considering a Rail sitting up 6" from the base...Thoughts? Have considered 8" as well

2) My next design will incorporate a dust enclosure. Its difficult to do this afterwards. I do suggest you consider it at design stage. Two things annoy after a while dust and noise. A water-cooled spindle is as quite as you can get
Once more Completely AGREED and have considered making the base in which the CNC Base is setting upon bidder to build an encloser around to contain the dust concerns

3) Sounds odd but the z axis does drive the footprint of the machine. Its worthwhile spending time on this before you get too serious with the rest of the machine, Under done Z axes show up immediately and its too hard to fix this. Once you have the z axis super uber stiff then work outward to fill out the machine. if you work from the frame to the Z axis you will run out of space to make an uber Z axis...
This is something I have noted as well and am looking at incorporating the lost area (WIDTH) of the "Z" into the footprint size...What width of the base of the "Z" would you consider? I was hoping for 6" but not wider then 8"

4) take advantage of height. make everything as high as possible as this adds stiffness without affecting the footprint. Make everything as big as possible. Don't shape things so they reduce in size, take advantage of the entire space envelope available for the part. This maximizes rigidity.
Can you elaborate a bit...I understood that there is a point of no return where going to high would induce lost rigidity in the "Z" ...example the walls for the rails and OVAL length/height of the "Z" being from the top surface of the base...If the Gantry is to high then the "Z" has to be longer to allow the cutting head to lower which would induce lost rigidity...

5) If using tubes make them thick. Light tubes may look great on paper but cnc machine parts are shear stiffness critical. Light tubes are great for global bending but poor for local bending and shear stiffness.
Yes my thoughts as well and this was my intention whether I build a steel frame or building a concrete base was the frame inside was going to be heavy gauge being at least 11g however I would like to do 3/16" wall on everything but Definity 3/16g where any fasteners were going to go internally for supporting components due to threading and holding capacity...

6) Hard to beat a steel machine if you have good welding and access to stress relief. Easiest & cheapest way to gain stiffness is to use steel...
This is still an option as I have not totally put it aside at this point...Its the stress relief that has me concerned as to where to get it done then what factors changes after the stress relief if done?

7) Do detailed CAD models explore possibilities don't buy anything until you have designed it twice and are happy. I still have things I got excited about and bought but turned out to be not right... if I had done the CAD work and contemplation it would have become clear.
CAD Not doable personally at this point in my life, all my design in putting down on paper with a pencil coming from my head and after I have it on paper then I may put together on wood as a mock up and if all is right then move on, This will allow for any changes to transpire 1st before the final build...However, I sure would like to find somebody locally that does CAD and would ask them while putting a bit of coin in their purse to help with the design...

8) Keep at it and don't take the easy path if you feel something could be better. The easy path usually has consequences and its the design time that makes a great machine vs an average one. The design sets the DNA and once set its difficult to improve.
Agreed

So as we conversed in the past I still desire to build a machine that has a 48" x 36" cutting area
As the old saying is Strong enough for a man but made for a women so what I would like to see this machine doing, is, when the occasion calls to do aluminum it performs without remorse when its called upon and able to do wood without effort...

Paul

[peteeng]

Hi Paul - 3/16" in my world is 5mm say. By thick I mean 1/2" min 1/4" or 5/16" So what is your gantry going to look like? I think you have to think through how your going to machine things accurately and you need to find a heat treater and a machinist. Your ganty is going to be 1100-1200mm wide 47". I think its time to warm up your drawing board get out your 2B pencils and start detailing, you need a good general arrangement so you can work thru some logistics, then its time for a HB pencil. .... Peter

re wall height mine are 450mm 18" but my z travel is 11". The wall is as high as it needs to be so the gantry has no column and the z travel is what you require

re Z width - depends on your spindle size and clamp width. get drawing....

[peteeng]

Hi Paul - What you describe can be realised with a plywood machine. I highly recommend ply as a first machine. The attached machine will cut timber, plastics, etc and aluminium. Timber is easy to modify, easy to drill holes in when you need a wire to go "there" easy to attach things to etc etc. Your first machine is never right. You can research and discuss for years and never get closer to the actual machine. Timber allows you to build it with simple tools quickly. You learn heaps and refine your work. The second machine can be built by the first machine, in timber or aluminium.

Concrete once cast is it. If you decide you need a hole its difficult etc. Or a solid aluminium extrusion machine. More of a mechanno set approach. Especially without modern CAD. You will spend years drawing things and have a huge pile of rubber on the floor. I spent many years working as a draughtsman while studying and rubber dust was big.. I even had an electric eraser at one place. Peter

[peteeng]

Hi Paul - To illustrate this point. Here's a calculation for the machine base. If your grout was say 3" thick then a plywood base 100mm thick would be the same stiffness yet be 39% the weight. To take it further if you laminate 3mm aluminium top and bottom it would be hugely stiffer and stable. No moulds easy to stack up ply using epoxy or even titebond and you have a monster stiffness base. Peter

[peteeng]

Hi Paul - So to look at the plywood aluminium sandwich solution. Compare a 75mm slab to a sandwich of 1/4" 6061-T6 plate 3x1" plywood 1/4" al plate. The sandwich is considerably lighter then the concrete slab and its 41% stiffer. Easy to make in your garage no moulds direct to part. The plywood/aluminium laminate solution has a lot going for it for a one off machine. My interest in grout is because I want to make one machine per week. Which means casting is a good solution vs making relatively complex multistage products. But the laminate solution is a good solution. Plus you can make the laminate lighter by having holes in the internal ply bits... does not need to be "solid" to achieve the same stiffness. I'm always trying to make a light machine at uber stiffness... Peter

By the way I don't think you need a 75mm thick base for your purposes. A 1/4"al - 2x1" ply 1/4" al would over do it... Peter

[peteeng]

Hi Paul - If you can get access to sanding thicknessers, large belt sanders, jointers like most cabinet makers have you can make large slabs of accurate laminated timber blocks then epoxy on aluminium skins. This would be accurate for your purposes. Then you take 1/2" plate to a machinist to have it drilled and threaded accurately then you bed that on top of the walls with epoxy putty. Use good levels to establish flatness and level and it will be a great machine base. Epoxy coat all exposed timber or cover with aluminium entirely. If you use flush trimmers you will think its solid aluminium I keep coming back to this as the numbers are so good. The only way to beat this is laminated steel... but my trimmer doesn't like steel.

mass timber construction is now the biggest thing replacing concrete in construction. Cross laminated timber is really big also. Combine that with aluminium and it's a great material. Peter

I'm designing a full sheet size router at present and going through this grout, laminated timber, steel, other material machinations cycle at present. If I'm doing parts myself the ply laminate wins, If I'm contracting parts then billet aluminium wins... or laser cut steel. Pick your poison

[peteeng]

Hi Paul - its difficult to get your head around how stiff sandwich's are so I converted the [6mmAL/75mmply/6mmAL] to its equivalent thickness in solid aluminium. Turns out its 70mm thick. So are you happy with a 70mm thick Al machine base? I would be for what you are describing... Cheers Peter

[Texasbowhunter]

Pete Thank You for the continuing education with the formulas you have supplied to this conversation...I some cases Its out of my realm to comprehend in all fairness, but, these last couple drawings after studying them seems to be more comprehensible...
So after I designed the base lay-out footprint using 12' sides it's going to be cost prohibited if I understand the drawing in which I interpreted them to mean...
1st- Plywood...the plywood down here is very undesirable and in all likelihood is of very poor quality, so I feel that getting away from the builders grade used in the housing industry would not be a good choice, so I would choose to upgrade in quality and go with something in a veneer of the face of it...Last time I bought a sheet of this where there were 11-12 ply's in 3/4" at $150 per sheet verse what the big box hardware stores offered in 5-7 ply's was $65 a sheet, 4 months ago I bought the higher ply's and that was what the sheet cost...I would require 3 sheets for the base and the walls plus the adhesive to bond the plywood together...
2nd Aluminum...just checked on a sheet of aluminum 120 x 48 x 1/4" thick 6061...It appears 2 per side of the walls total 4 each and one for the top surface of the base however if one more is required for the bottom of the base then additional cost would be needed...sheet of 48 x 120 will cost $982 plus shipping would be relative......
$983 + $150 x 3= $450 would be $1433
I believe with this process its going to be cost prohibitive for me to pursue this thought and am going to have to rethink something more...
Paul

[peteeng]

Hi Paul - for a 48x120x1/4" aluminium sheet that weighs 60kg I pay around $750AUD (typical price about $11.50 per kg) for the sheet at a retail outlet. Seems you have expensive supplies in your area... Builders ply is definitely a no-no for a good machine build, too inconsistent. Furniture grade or good quality form ply is the best choice. I use hardwood formply and a 4x8x19 (3/4" thick) sheet costs $65AUD. Seems the grout is the go for you. How thick where you going to cast? Peter

[peteeng]

Hi Paul,
if I did it here in grout it would cost $729AUD if the base was 3" thick and the walls 4" thick. The walls are 4" because the way I do the drive is 100mm wide. But that design 100mm walls and 3" base is way over cooked for your router. It weighs 472kg, more of a mill then a router. Can be done at less than half that weight so halve the material cost. Use a honeycomb back or make it 50mm all round. Sorry about metric. I was brought up on imperial but we swapped to metric many decades ago and its much easier for the maths. Unfortunately the USA, Myanmar and Liberia are the only countries in the world that still use imperial. The USA will have to catch up one day.

A 75mm slab is equivalent to a 59mm thick aluminium plate. So way over the top. A 25mm plate would still be over for what you described. Typically I use a 3/4" hardwood ply top with 100mm webs creating a torsion box base. Or my commercial machines are built from 3mm pressed metal ribs and use a 3/4" formply top... Here's a happy chap building his machine in the workshop. It's a really stiff base. If you want it to be a wet machine then use aluminium plate base (6mm) and sheet metal sides (1.6mm laser cut steel). If you want a price on a base kit then PM me... I have one in stock.... Peter

[ardenum2]

if a U bed is too heavy make a slab and add legs to the gantry, should half that 450kg.

[peteeng]

Hi Ard - Then it's a column machine and the drives etc are in the muck. Thats not the point of a high rail design. The walls and base can be cast thinner with honeycomb backs and they will be much lighter and still quite stiff. I'd expect a light base design to be around the 100kg mark...Scoot-S3 as per the last image with happy Matt we picked up the assembled base and put in my van for delivery. That included the drives, rails and wiring etc. So was below 100kg....Paul hasn't decided how stiff he needs yet. Peter

[Texasbowhunter]

Hi Paul - for a 48x120x1/4" aluminium sheet that weighs 60kg I pay around $750AUD (typical price about $11.50 per kg) for the sheet at a retail outlet. Seems you have expensive supplies in your area... Builders ply is definitely a no-no for a good machine build, too inconsistent. Furniture grade or good quality form ply is the best choice. I use hardwood formply and a 4x8x19 (3/4" thick) sheet costs $65AUD. Seems the grout is the go for you. How thick where you going to cast? Peter

Pete Its seems like everything is Costly this time around...I expect it to improve soon...Not soon enough for this project unless I put it on hold another year...(which at my rate is possible)
The big box stores have a better grade of the veneered 5-7 layer laminated plywood. I went down to Lowes today and just out of curiosity I looked at their veneered wood in 3/4" and a sheet was Hovering at $70...I've used their wood on a few projects but soon realized it had an awful lot of voids in it where I made the various cuts...Not realizing there was better product out there until I bought the premium stuff from a real lumber store who obtained some of their product from overseas that had the 11 ply's, did I realize there was no voids to speak of and the wood seemed more uniform...
However, $150 a sheet, kind of put's the question mark as to whether I need to go down that route...Or find something more reasonable...

So just curious as to what kind of adhesive you would use for the plywood being bonded...large base would you vacuum it to pull as much of a uniform pressure on the wood or weights and hope for uniformity?
Now, what would you use for the adhesive to bond the aluminum to the wood?
As far as the thickness of the uprights (walls) and the base I was considering (playing with) 4" thick with 6" high walls but after your suggestion I need to double that so 12" high walls for the rail will be factored in presently......It seems Like I'm going to have 94 sq. feet to fill...
There I'll factor in a steel frame inside some of it to allow the exterior components at have a place to attach to which will already be drilled and taped for that component...
Paul

[peteeng]

Hi Paul - Timber to timber Titebond is fine, easier to use a single component glue. Timber to metal use epoxy. If you have vacuum that's a great way to go otherwise you and your Neighbours have to sit on it for a few hrs and drink beer. Wide girth visitors are useful as well... Peter

[peteeng]

Hi Paul - You can also add bean bag foam balls to the concrete mix. 50:50 by volume. This halves the volume but maintains the big geometry. Peter