Steel frame design. Opinions please

[MattH]

Do you run a CNCing business?

We've just started it in the last year. I run my own IT consultancy and we have some support contracts that pay the bills but doesn't take much of my time.
We started making and selling craft blanks/shapes and in the last 6 months it's gone mad and is turning into a full time business which will hopefully allow me to give up IT (been doing it for 20 years and it doesn't interest me anymore).
This year is the time to invest in the new venture.

 

[MattH]

Here's a rough design using only RHS and cross bracing in all directions:

 

[Screwdriver]

If it were me I'd be looking out for a big old machine base. They come up on eBay all the time, though you might have to buy a whole machine (for peanuts).

Can't beat a big old cast iron base for rigidity.

S.

 

[Spark plug]

Here's a rough design using only RHS and cross bracing in all directions:

View attachment 78052 View attachment 78053 View attachment 78054

This design looks good although if it where me I would have the braces opposite so they the two braces in the front and rear elivatiin came together at the bottom of the central pillar.

 

[MattH]

This design looks good although if it where me I would have the braces opposite so they the two braces in the front and rear elivatiin came together at the bottom of the central pillar.

Was thinking about that. Which would be best though, top outside to middle bottom? Or top middle to outside bottom?

 

[Spark plug]

I would say top outside to middle bottom.

The more I'm thinking about it the key to it going to be getting the right balance bettwen weight and strength within the gantary. one thing to watch out for is its very easy to get greedy with your z axis clearance. Only make it just bigger than you expect to use, the closer you can keep the cutter to the X axis bearing the better.

 

[MattH]

I would say top outside to middle bottom.

The more I'm thinking about it the key to it going to be getting the right balance bettwen weight and strength within the gantry. one thing to watch out for is its very easy to get greedy with your z axis clearance. Only make it just bigger than you expect to use, the closer you can keep the cutter to the X axis bearing the better.

I agree on the Top outside to middle bottom.
The Z clearance is going to be dictated by the ATC spindle which needs enough hight to clear the top of a tool holder. I will only be cutting sheet goods so the the max material depth will only be about 60mm.
I have a piece of Bosch-Rexroth 180mmx90mm heavy extrusion for the gantry so it will be very stiff with the minimum of weight (the 900mm long piece is about 15Kgs).
The X rails will be bolted to the sides of the long box sections on the top of the frame, two upright boxes will be made from steel plate and the t slots in the bottom of gantry will be used to attach them. The servos for the x axis rack and pinion will be housed inside these boxes and will be low down, roughly level with the bearing blocks.

 

[MattH]

Right, changed the bracing on the model as per Spark Plugs suggestion.

I've also added a basic representation of the rails and the gantry (basic because you can't import external models into the free version of DesignSpark).
Bear in the mind that the gantry supports will be box structures and NOT flat plate.

 

[MattH]

And just to show this isn't a figment of my imagination here are my ISO20 tool holders and the new forks that turned up today.
Can't be a***d to go out in the cold workshop to take a pic of the very expensive spindle these fit in!

 

[MattH]

For comparison here is the model of my current machine, I think the new one will be a bit stiffer!

 

[Maker]

The new design does look very stiff, but the old one doesn't look lacking either.

 

[MattH]

The old one was built to go through a standard sized door on it's side. This makes the mid point a flex point and the only things tying the frame together in the x axis is the main x rails which mount the bearing rails. In practice with the gantry flying back and forth the frame bounces and flexes a lot.
It's amazing what 2 400w servos driving a 50Kgs weight can do!

 

[mike os]

Plenty stiff there I think

You need to think how you will level/flatten the top of the frame as it will distort from welding

 

[MattH]

Plenty stiff there I think

You need to think how you will level/flatten the top of the frame as it will distort from welding

I have been, trust me. A lot of people use a self levelling epoxy poured onto the machine. I'm not a fan of that method.
The critical areas are actually the two side rails that take the linear bearings, these need to flat, true and parallel.
My two options are:
1, The two rails are going to have 40mm.x15mm flat bar welded down their length as a mounting platform for the rails. This will be machined after welding (but before the box is welded to the frame). If I can keep the distortion to a minimum then this should work. The tops of the rails can also have a flat bar welded to them and then I can fit a temporary low speed head on the machine and mill the top flat. Once the table surface is fitted (probably Tufnol), then the Vacuum Plenum then the spoil board, all of which can be skimmed flat to the machine I should be fine.

2, Is to take the entire frame and get it stress relieved and them the sides of the rails and the tops all get machined.

 

[gaz1]

query here mattH how come you havent done a table within a table as wouldnt that give it more strength and weight to the table

also you can replace the inside table as it gets worn down by the plasma cutter

 

[MattH]

query here mattH how come you havent done a table within a table as wouldnt that give it more strength and weight to the table

also you can replace the inside table as it gets worn down by the plasma cutter

A table within a table? Not following, if you mean a fixed gantry and moving table then that takes up a lot more space.

No plasma cutter involved. this machine is for wooden sheet goods and will use a 3HP router spindle. My vacuum table uses a bleed through spoil board that gets re-skimmed as necessary until it is to thin. Then it gets skimmed off and a new spoil board glued down.
Here is a pic of removing my last spoil board last weekend:

 

[mike os]

1 is probably the best/easiest solution... may be better to bolt the box to the frame? (or bolt, adjust then tack) i thought about a similar approach and came to the conclusion that slipping a strip of 40x12 ish tapped for the rails inside the box would allow maximum adjustment and minimum messing with difficult machining

2 is probably a non starter i would guess by the time you are finished you could buy something

 

[Spark plug]

Just to throw a mother idea into the mix, build your steel frame table, then fit a sheet of 25mm or there about's flash ground steel plate. Then you will have a perfectly flat serf ace to mount the bearings and vac bed.

 

[gaz1]

seen as you mentioned that your considering making a bigger table latter why not half build it now

this is based on a 4x3 shape



the red insert is your top table I recognise it as a metal gantry walkway inserted into a premade frame below it as you can use the walkway as a welding table base or for a plasma cutter only renewing the walkway metal when needed

if you wanted to do you could split that design into 2 at the side point therefore able to insert a bigger base insert therefore extending it to an 8x4 sheet as long as your runners are at the side points as these can be made bigger for the inserted table

the cnc would run on the outer edges of the 4' so when extended the cnc frame would not need to be rebuilt only the runners would need replacing or l shaped so they overlapped the previous build

therefore the whole build is all flat with no raised internals like yours

it also becomes transportable if needed

 

[MattH]

I'll try to reply without sounding arrogant.

There are a million different ways of building something like this. This will be my 3rd CNC router of this size, my first was built over 10 years ago. I have spent thousands of hours on and in CNC forums and communities. I have a large number of outside factors that govern exactly what I can and can't build.
The frame is less than 10% of the overall cost of the build, compromising the frame build at this point to part satisfy a future upgrade would be folly.

There is no Plasma. This is a router cnc, there are cutting forces being transferred from the material into the frame. Making a two part system would introduce flexibility which is bad.
Putting the rails on the outside means that the gantry has to be wider, as I am constrained to a certain gantry width then your design would make the frame narrower. With my design the rails are where they need to be but the frame below has been widened as much as possible to increase stability.

From the moment my current CNC was built I have been considering this design (about 3 years), the purpose of this thread is to make sure I build all the strength I can into the frame.

If I failed in my first point then apologies but this is not my first Rodeo.