3000x1500mm table drawing stage

[reefy86]

rather have this as its own thread so i apologise if im bombarding the site.

frame is 3000x1500 so i can have a cutting area 2420x1220 and potential upgrade to a 4th axis in the future. just focusing on the frame for now and looking for any criticism. I am bolting it together to help with adjustments and hopefully make it easier to align all nice and square.

its all made with 100x100x4mm mild steel.

looking foward to any feedback

thanks

[JAZZCNC]

Hi Reefy,

I've replied to your emails and much of what I'm going to say now is just repeated for other's who may be thinking to go large.!

Regards bolting or welding then bolting the mainframe together at this size won't help much other than making it easier to move if need it to go through small doors etc.
Bolting the top rail is a good idea because it allows you to adjust the rail surfaces onto the same plane, however, this is where it gets difficult because you need some way to accurately measure that each surface is on the same plane.?
Also, and here is where it gets even more difficult when it gets long.? The steel you use to mount the rails onto must be perfectly flat and straight or you are wasting your time with shimming etc.
If the rail surface is up and down or twisted along its length then you are screwed and you'll never build an accurate machine. I can tell you with 100% certainty that the box section you buy, no matter how thick, will not be flat and will be bent or twisted to some degree.

This is why people use the Epoxy method on large machines, it self levels to a reasonably accurate degree and removes the twisted or bent steel problem.
So in this case bolting the frame and top tube doesn't really matter so much because the epoxy levels everything away.

If you don't want to use epoxy then the next best option is to just have the top rail tubes machined flat, then you just have to align and shim the surfaces onto the same plane to deal with.!

Or the best option is to weld the whole frame up and send it away for the top tubes machining flat and on the same plane. This will cost you but will give you an accurate machine.

If you try to do this without either epoxy or machining surfaces flat and just rely on the steel tube surfaces then you'll end up with a poor and low accuracy machine. If you just want to cut 2D shapes, like signs or paneling, etc then it will be perfectly fine, but if you want to do engraving or work which requires higher accuracy then it will fail badly.

People confuse strength and accuracy, they think building strong with thick tubes, etc gives the machine accuracy and it doesn't, it just makes it strong with less vibration. The accuracy comes from the frame alignment, surfaces, and quality of rails,ball-screws, etc.
If the rail surfaces are out of whack then doesn't matter if the machine was cast from a solid block of titanium it will still give rubbish accuracy. Likewise ball-screw alignment and quality drastically affect performance and accuracy so again if the surfaces these mount onto are not perfectly flat and parallel to the rails then you'll have set up issues which affect performance.

I know you are confident to give this a try but still make sure you fully understand what's required to build an accurate machine at this size that will do all you need.!! . . . I'll warn you again it's very very very difficult at this size to build an accurate machine and very very very very very very very very easy to f$ck up.!!

My best advice is that if you are determined to carry on at this size then weld it up solid and send it away for machining. It will be money well spent.

[cropwell]

I have the feeling that, unless you get the rigidity and bracing right, then sending it away for machining will only result in distortion of the frame whilst being handled for transport. Epoxy levelling seems to be a better way to get the machine bed level and far less costly than transport and machining. I would love to see any quotes you get for getting it machined flat as any company with adequate kit to do 3000 x 1500 is going to charge mega for a one-off.

[JAZZCNC]

I have the feeling that, unless you get the rigidity and bracing right, then sending it away for machining will only result in distortion of the frame whilst being handled for transport.

Yes, the frame needs to be braced and welded completely if being machined but that's true anyway for a good machine whether machined or not. If done properly then transporting should make no difference.

Epoxy levelling seems to be a better way to get the machine bed level and far less costly than transport and machining. I would love to see any quotes you get for getting it machined flat as any company with adequate kit to do 3000 x 1500 is going to charge mega for a one-off.

Epoxy is ok but at this size, it's not cheap and it's very easy to get wrong. Also, while the epoxy method is much better than no prep of the surfaces it's still only an intermediate level of accuracy and again at this size very easy to get wrong (can you see a theme here.?)

Having the frame machined isn't as expensive as you might think, but yes it's not £150-£200 like epoxy but it's also not thousands as you might imagine it to do be.!
To be honest, I was just about to hit the button on buying/building a large machine for surfacing frames but our landlord recently sold the industrial estate so we may need to move premises so I'm holding off until it's decided as it won't be an easy or cheap machine to move, plus we have a large Hurco mill that weighs 9 ton and other smaller mills/equipment which will take some shifting. However, when we do then I will be building and offering the machining service and steel frames for sale.

[cropwell]

My experience of small engineering firms is a bit 'sharp intake of breath' when price is mentioned. I had some surface grinding done 10 or 15 years ago and my first comment was I wasn't asking to buy the machine.

[JAZZCNC]

My experience of small engineering firms is a bit 'sharp intake of breath' when price is mentioned. I had some surface grinding done 10 or 15 years ago and my first comment was I wasn't asking to buy the machine.

Maybe so in that case but you can't tar all firms with the same brush. I use an engineering firm that is very fair so they are not all the same.!

[mekanik]

If i had to attempt something this size my plan of attack would be weld the main frame, make the pads on your uprights a good thickness, after welding spot face the pads so you have a flat surface for taking shim sizes, on your top rails have some good thick pads to mate with the ones on the frame, send your two top rails away for machining of top suface and pads plus the sides for ballscrew alignment, align the top rails and measure for shims.
just my opinion
Hope it helps.
Regards
Mike

[JAZZCNC]

If i had to attempt something this size my plan of attack would be weld the main frame, make the pads on your uprights a good thickness, after welding spot face the pads so you have a flat surface for taking shim sizes, on your top rails have some good thick pads to mate with the ones on the frame, send your two top rails away for machining of top suface and pads plus the sides for ballscrew alignment, align the top rails and measure for shims.

That's a good approach and one I mentioned before but it actually works out more for machining 2 rails than having the whole frame machined. Then you still have to deal with aligning the rail surfaces onto the same plane.!

The reason it costs more is due to setup time, the frame is just one setup, with 2 separate rails it's two setups. You do save in transport because rails will fit in a smaller vehicle.

[mekanik]

Aligning the rails should be doable what you need are two taps on each of the pads frame and rail, tabs on the frame get tapped with a fine thread so you can adjust the rail for roll & pitch/align and shim.
Was looking for a picture of Boyan's big machine yesterday but couldn't find it, also i believe Boyan has a couple of his machines in CAD if reefy86 was to get in touch, if i remember correctly Boyan stitch welded his frame to keep down distortion.
All ideas to ponder.
Regards
Mike

[JAZZCNC]

Aligning the rails should be doable what you need are two taps on each of the pads frame and rail, tabs on the frame get tapped with a fine thread so you can adjust the rail for roll & pitch/align and shim.

Yes, it's doable, I've done it many many times. I've also tried your suggestion and it doesn't work quite as well as you'd imagine on large rail surfaces. It takes an age to get right and can defeat the point of having the rails machined flat because it's so easy to introduce error if the shims are not spot on when tightened down.

The best method I've found other than sending the whole frame away for machining is a mixture of using the pads with set screws for fine adjustment like you say but also to raise the pads leaving a gap of about 3-5mm when tightened down solid. Then instead of shims use an Epoxy metal putty sandwiched between the pads with a release agent so the top rail can be removed if needed.
This method fills any voids or distortion from welding the pads and gives a very solid surface, the epoxy also acts as a vibration dampener.

Was looking for a picture of Boyan's big machine yesterday but couldn't find it, also i believe Boyan has a couple of his machines in CAD if reefy86 was to get in touch, if i remember correctly Boyan stitch welded his frame to keep down distortion.

Boyan's machine is well OTT IMO and still doesn't get away from the epoxy method which is only suitable in accuracy terms for an average machine. It also took him a few attempts and a lot of wasted money in epoxy.!

End of the day it depends on the usage of the machine and what accuracy levels your looking for or expecting. If mostly cutting woods and V carving or panel type work then usually if done correctly any of these lower expense DIY methods work great.

It's often the smaller details that show up the difference in build quality, like the example of one of my customers which I gave Reefy in an email. They produce bespoke large highly detailed doors with 2.5D scenes cut into them and for a long time struggled to produce the quality of door they wanted. They first blamed the machine because it was a merchant dice machine, then they went the DIY route and that didn't end well, so they bought a Chinese machine but still, they couldn't get the quality right.

The problem wasn't the highly detailed 2.5D scenes which were quite complex models, it was the little details like chamfers on the door edges or Grooves around the edges of panels etc, they couldn't get a consistent width along the length of doors or large panles because the rail top surface wasn't flat or on the same plane and anyone who's battled with chamfers nows, they look a mess when not right.

Now you can imagine the frustration after cutting a 12hr 2.5D scene into the door (twice) only for the last details to mess it up.