Accurate Strong Gantry

[Ross77]

The spindle on a vmc has to be like that so the table can move. But the column is fixed and only the z axis moves. Don't get your designs mixed up, you have the opportunity to build some thing specific to your requirements so best look to optimise it.

If you want to get technical then you don't have to put the cutter directly in line with the bearing but you need to balance the y axis and put its centre of gravity in line with the rails as that will produce the smallest moment loading in the bearings and hence deflection.

I wasn't meaning to use a H beam but rather trying to explain the position of the beam relative to the vertical bearings, ie keep the beam 200mm deep but increase the bearing spacing To 400-500mm. Might need to draw something up.

[vargai]

Might need to draw something up.

Tried to catch the principle based on sketches but this is more complex and one detail can change the others.
Here are some slightly more detailed plans -the C/G is near the C/L of the horizontal ball screw so hope this setup works.

That would naturally reduce the racking and take the pressure off the screws and rams.

I have increased the bearing spacing-350 mm now.I do not think it is worse than a moving gantry since it has double Z screw. Forces on screws mainly come from the vertical feed force -assuming it about 200 N(??) while I would like to reach 1000 N machining force in the horizontal plane.
If something might be considered to apply one more shorter Z rail and +1 each carriage on both side-I do not know if it worth
Drilling only means center drilling in this machine and max 6-8 mm holes. The machining that requires Z force will be arranged in the middle zone of the table where the load is almost symmetrical on screws

Regarding to the profile I started to change them to bend opened ones from 6-8 mm thickness. All of them havv to be designed according to their function

By the way when I am taking about balancing I am thinking of the Z motor too. I want it to get relieved form beam mass. It cannot be balanced perfectly since the spindle is moving but I want to remove the dead load expect the spindle at least.

This vertical in-between railing makes me worry a little.
The implementation will not be too easy. At the moment I am planing to put together the Z frame from three bolted parts

[Ross77]

Looks good. I would probably extend the gantry uprights down to the floor so the more area would be available for welding and increase rigidity. the vertical rails could then drop down and increase the bearing spacing a bit more.

I think golden ratio is 2:1 so if the y axis is 800mm wide then you need 400mm bearing spacing. as you pointed out one change can affect another and you will need to look at flex in the mounting plates and possible bracing,


Do you think the gantry uprights need bracing as well? looks quite tall

It may have already been mentioned but how to plan to keep all the surfaces perfectly aligned whilst welding and given the surface tolerance on steel c section?

[vargai]

Looks good. I would probably extend the gantry uprights down to the floor so the more area would be available for welding and increase rigidity. the vertical rails could then drop down and increase the bearing spacing a bit more.

I think golden ratio is 2:1 so if the y axis is 800 mm wide then you need 400 mm bearing spacing. as you pointed out one change can affect another and you will need to look at flex in the mounting plates and possible bracing,


Do you think the gantry uprights need bracing as well? looks quite tall

It may have already been mentioned but how to plan to keep all the surfaces perfectly aligned whilst welding and given the surface tolerance on steel c section?

Extending the gantry uprights down seems good idea -this node should be rethink with the bracing and reinforcement plates in profiles.

My aim is to design parts that can be done by both DIY and machining method.
machine bad is Ok -it has surfaces to be leveled on top -preferably positioned on the same level-that is way stand the gantry leg here at the moment

Y beam is the same -can be done on both way

Z beams will be separated and bolted so it is similar to Y -I do not want to weld a U form and do something with frame -it makes my life hard
rather to make strong flange nodes and the careful assembly and adjustment will give the accuracy (especially the two parallel surfaces)
Nodes will be oriented and fixed with parallel pins.
That is only may plan for now -God knows nothing about it.

[JAZZCNC]

That design will be a night mare to build and setup which isn't required and will have no gain over other simpler designs. C channel is bad choice due to it being bendy/twisty in relation to boxsection.
I don't think you have really thought about the complexitys and realitys of building this design.?

The C section inner surfaces won't be flat so will need milling for the rails. Then you have the problem of making sure the uprights are perfectly parallel in 2 planes and perfectly perpendicular in the other.!! . . 99.9999999% Nailed on that you won't be able to weld this up and keep parallel/perpendicular in all planes.
Any surfaces that where milled flat and true won't be flat and true after welding so this will be a waste of time and you have no way mill when welded up.
There is a no provision for adjustment and the design is such that there is no room for any error to be taken out.

Boxsection will work far better and be much easier to setup or deal with alignment errors after welding.
The slight difference from ideal bearing loading which Ross suggested won't make any differance to the machine in real world performance or longevity but it will to your mental health and sanity when setting up.

Also the Uprights need far more bracing as any deflection here will show at the cutter and finish quality. Personally for best strength and bearing loading I would have double uprights each side of gantry with 1 rail on each upright and the uprights would be braced in 2 directions.

I feel If you go with this design you'll regret it come setting up time.!

[vargai]

99.9999999% Nailed on that you won't be able to weld this up and keep parallel/perpendicular in all planes.

I do not want to weld a U form and do something with frame -it makes my life hard

First of all thank you for giving me a 0,0000001 chance.
This is a design evolution process with analysis, checking, listening to the pointers . I am not able to put a ready for production plan on the desk right now and I am patient and have time. Bad design leads to a better one. (in a good case)
This is still a study where I try to find the place for all the function.

Some detail explained in written form and never meant to weld this construction otherwise I would need medical treatment right now.
Yes, box section is better than C no question-even when equipped with bracing
For some reason I supposed that box section is more inclinable to resonate than a thicker opened one
The stronger the better so I am opened to the closed section
Just a sketch about how I understood double upright braced in two direction

[Boyan Silyavski]

Now you have got me with that constant changing of the design. Why don't you like one design, stick to it and further develop it to suit your needs. If you liked the before suggested design from the video in post #41, then just see what is good and what could be bettered. i would suggest even contacting the guy and asking him if he were to make it again what he would change and what he liked. That will save you a lot of unnecessary trouble and give you invaluable real life feedback. Thats why people stick with certain designs, cause they are good and there is feedback.

I understand that the overhang of his design is bothering you, and it should be. Cause it bothering me. Not only the spindle, but the hole gantry overhangs. That kind of kills the good points of that particular design.

One thing from a design point of view people should understand is that the fixed gantry design we discuss here has one really strong point. I am sure most people don't understand this point properly. Its not that the gantry is fixed. Its that the rectangle it forms is very big so its moment of inertia is incredibly big. So when changing that fixed gantry design/where it forms a portal with 2 legs/ in any way that compromises it main strength is not a good idea.

Another basic point to understand is that all rests on a couple of bearing blocks in all 3 different coordinate planes. So no matter how strong you make the machine frame, the bearing spacing, placement and forces there will form the real weak spots.

the purpose of the strong frame is to hold enough for the forces. next purpose is to minimize bend and twist which leads to minimizing vibration.

What people fail to understand, i see it here and in the other forums, is that the machine frame should be sturdy to a certain point and then the aim should be to be integral. No need to make 1 ton machine when same could be achieved with integral 200kg one.

Sturdy machine that is extremely rigid will ring like crazy if not integral. So thats when all make them heavy instead of making the design integral.

By integral i mean designed so that all forces spread equally, neutralize themselves and no weak spots and overbuilds in any direction possible.


Thats why there are some principle designs that all use, cause they are integral. Not that always people take full use of the possibilities with that design.

I assure you that a properly designed L gantry will beat any day the fixed gantry design if its not properly designed.

And IMHO after playing a lot and rethinking all designs and reinventing the wheel, my conclusion is that if all things taken into account we must not forget that we usually use only 2.2 -3kw spindle, using only 6-12mm bits, so we should not forget the strongest point of each design and choose it when needed, cause at the end they will all work if made properly.

[JoeHarris]

The C section inner surfaces won't be flat so will need milling for the rails. Then you have the problem of making sure the uprights are perfectly parallel in 2 planes and perfectly perpendicular in the other.!! . . 99.9999999% Nailed on that you won't be able to weld this up and keep parallel/perpendicular in all planes.

I feel If you go with this design you'll regret it come setting up time.!

This is so important when developing your design. I don't know how much experience you have welding? But things will distort and if you are just starting out you want to be able to weld in a flat position as much as possible. Building in adjustment is key - especially if building using limited equipment and skill. You can draw to crazy levels of accuracy but when it comes to drilling a hole or cutting to 0.1mm using the kit you have in your garage!???

With my cheap drill press, cut off saw and stick welder I have needed some tolerance building my frame and I have seriously taken my time!!!

You are doing the right thing sharing on here though you will get some v helpful advise.

[Boyan Silyavski]

Better read this Advice sought on new Mill build

What you need will be BT30 spindle+servo motor.+ATC attachment - all from China , Aliexpress


You should beef everything then considerably. This is a serious build for first one. It seems you still don't have it clear for what exactly will use the machine? Maybe an used mill /VMC/ will be best ?

[vargai]

You can draw to crazy levels of accuracy but when it comes to drilling a hole or cutting to 0.1mm using the kit you have in your garage!???
With my cheap drill press, cut off saw and stick welder I have needed some tolerance building my frame and I have seriously taken my time!!!
You are doing the right thing sharing on here though you will get some v helpful advise.

Thanks for the attention and care.
I have to share my inner motivations and means for the best understanding.
I.e. Silyavski thinks I bite off more than I can chew
I am amazed at the way you guys build very good machines.
As for me I want to apply machining where necessary.
Perhaps I am not the only one here with this. But personally I feel a pang of conscience because this is not really DIY method and do not know where is the border.