Stell frame cnc 1500X1000. Adjustable table your opinion please

[JAZZCNC]

Any of these are strong enough but the design can be simplified regards building and be just has strong. No time at moment to get into full debate about why etc but here's frame I've drawn that's similar.

Main points are the sides and Bed are welded flat has separate units then joined together with end braces etc. The Bed being a separate unit means the outer frame pieces support the bed material at the edges, the way yours is drawn the edges will be unsupported and could bend or flex under cutting conditions.
Lower diagonal braces help keep machine frame square and stiff, remember it's not just strength you have to consider there's resonances to deal with and bracing helps reduce resonance.

Welding major parts in sections on a bench then either welding or bolting together is much easier than struggling trying to weld individual pieces together and keep every thing sqaure and flat, aligned etc.

One other note.!! On this design the top rail extends.? This is to make full use of the Bed without wasting material having the bed/frame longer than needed which can't be used.

[EddyCurrent]

One other note.!! On this design the top rail extends.? This is to make full use of the Bed without wasting material having the bed/frame longer than needed which can't be used.

Good idea and after all these frame discussions I think I'm now at a stage to start cutting metal this weekend.

[Boyan Silyavski]

Deans design is very good, after all he is far more experienced at building and dealing with vibrations. I like the Z at the baze, that would definitely help with resonance. I like the extra bed frame, i have thought it but never included it in my designs, due to extra weight.


Dont get me wrong on what i will say, but i am fan of simplicity. I will say the same what i said to Tony at his thread

Look at that design below and adjust it to your needs if you like it. Just don't change anything radically, cause in its deceiving simplicity there is a lot of reason. In fact 2-3 months of reasoning. There are no random distances there. Tony changed it a bit and you can see his great looking machine. He made it from 80x80, mine is 100x100x3
If you look at my thread i have a frame soldered in my garage with the same design and it has deflection under 100kg of load which is irrelevant.

Now what Dean says about resonance is true. Its not the same as stiffness. In fact in some cases the stiffness will help the resonance, especially in our case where we build machines from thin walled profiles.The great thing about my design is that all the forces are working only at the upper part of the structure, its stiffness does not depend at all of lower elements. The legs are just legs / i am talking about 100x100 or short 80x80/.

Now there are hundreds of points why resonance will not happen at all in the machine i am building, nor twist or bending, even if i were able to route deep aluminum and why not steel / which will not happen cause i limited it with 0.8kw spindle/ . This points are too long to explain, but be careful, cause these are particular decisions, like that the machine i am building will have 100mm thick MDF bead, underneath plastic sheet or aluminum sheet and etc...

Long story short, just take a look and study it. Use it if you like it. It implement Deans idea of the shorter table. Also implements subtle ways of dealing with resonance, like the beams of the bed gradually don't coincide with the short vertical ones. In fact in my next build everything will be slightly offset here and there, to kill the harmonics.
The best thing is if like Dean says the bed beams flex, at a later time can be added ribs bolted to them vertically and if there is a twist , you can add the same Z diagonals like in Deans drawing of the lower part just drilling from below and bolting. Add if when you need, why overbuild now? Not to speak of that with 100x100 profile is way overbuild as my measurements show on my build.

[EddyCurrent]

The best thing is if like Dean says the bed beams flex, at a later time can be added ribs bolted to them vertically and if there is a twist , you can add the same Z diagonals like in Deans drawing of the lower part just drilling from below and bolting. Add if when you need, why overbuild now?

Exactly my plan, if there are problems later the frame can always be upgraded.

[JAZZCNC]

Now there are hundreds of points why resonance will not happen at all in the machine i am building, nor twist or bending, even if i were able to route deep aluminum and why not steel / which will not happen cause i limited it with 0.8kw spindle/ . This points are too long to explain, but be careful, cause these are particular decisions, like that the machine i am building will have 100mm thick MDF bead, underneath plastic sheet or aluminum sheet and etc...

Silyavski think Your fooling your self if you think you won't get any resonance, Even machines built from Cast iron get resonance so you have zero chance of getting None with hollow Steel structure.!! . . . . . Yes it will be less than machines built with thinner material or of lesser construction but I guarantee you'll get resonance which shows at the tool when cutting slightly deep in aluminium and definitely in steel.

All the frames shown will be more than enough for DIY use right upto cutting aluminium and massive overkill for most softer materials. But it would be foolish to think just using larger dimension material means you'll get no resonance, no matter how you arrange the joints or welds etc.

Also what's the point of going to all the trouble of building machine that doesn't deflect more than 0.01mm only to use a bed material that will deflect more than an elastic band.??? . . . . Waste of time doing all your doing if your going to use MDF in any part of the bed other than has a spoil board.!!
Even then for any kind of acceptable accuracy you'll have to surface it for every Job.!!

One more point regards adding stiffeners at a later date.?? . . . .WHY . . . when it can be done at the beginning. When ever you weld you have risk of heat distortion so it's not a good idea to be doing this at later date.
Also has most who have built a machine will know and back me up on ounce you have started using the machine you'll never stop to do even simple things let alone major upgrades like welding in stiffeners. . . . . It's known fact if you don't do all the planned things before powering up machine they'll never get done ounce it's working. . .Lol

Ba99274 any of the frames shown including your own designs will be good enough so don't worry you won't go wrong.

[Boyan Silyavski]

Hi Dean,
First i want to thank you for the help with the machine we talk about that i am building for a friend which wouldn't be possible without your kind help.

I absolutely agree with what you say. I wouldn't want to mislead somebody. I am always speaking in relative terms here, meaning having in mind DIY machines, with best possible price functionality relation, designed to be multifunctional, mainly for wood and for the occasional aluminum job.

That is why i said it will be good if the design is not copied but adjusted to the particular needs. So let me clear some points about the specific design, not to you i mean , cause i kn ow you know it, but to those who read:

-Its meant for wood and plastics. Its meant also for aluminum but only when additional bed is fixed to raise the job
-its meant for deep 3d jobs mainly and Z axis travel -170-200mm
-Its meant for a trunion table to fit for aditional axis for 3d jobs
-its meant if need arises for a removable plasma water bed to be fitted for occasional jobs.

In other words its meant to be all type of machine, so certain sacrifices has to be accepted.

Silyavski think Your fooling your self if you think you won't get any resonance, Even machines built from Cast iron get resonance so you have zero chance of getting None with hollow Steel structure.!! . . . . . Yes it will be less than machines built with thinner material or of lesser construction but I guarantee you'll get resonance which shows at the tool when cutting slightly deep in aluminium and definitely in steel.

That is exactly why the frame is overbuild and a small spindle is used

Also what's the point of going to all the trouble of building machine that doesn't deflect more than 0.01mm only to use a bed material that will deflect more than an elastic band.??? . . . . Waste of time doing all your doing if your going to use MDF in any part of the bed other than has a spoil board.!!
Even then for any kind of acceptable accuracy you'll have to surface it for every Job.!!

I told that to the friend i am building the machine for. In any case, the machine will have 20mm aluminum sheet bed. Over the aluminum 10mm sacrifitial hard phenolic plastic sacrifitial layer. Over it when necessary will be fitted solid 100mm high block of MDF, ply wood or wood, and yes it will be resurfaced each time, i told him so. So basically nothing would flex there. you misunderstood me because of my English, the machine will have 20mm aluminum bed at least,ribbed where necessary, not MDF!

One more point regards adding stiffeners at a later date.?? . . . .WHY . . . when it can be done at the beginning. When ever you weld you have risk of heat distortion so it's not a good idea to be doing this at later date.
Also has most who have built a machine will know and back me up on ounce you have started using the machine you'll never stop to do even simple things let alone major upgrades like welding in stiffeners. . . . . It's known fact if you don't do all the planned things before powering up machine they'll never get done ounce it's working. . .Lol

You couldn't be more right. Thats the truth. My smalll cnc has 2 flimsy bearings and i cant find the time even to dismantle and tighten, as i dont stop it at all.
So I believe it would not be necessary at all. Just said it to remind that if a mistake is made with the design, can be fixed later when testing the machine.



So in conclusion/my conclusion i mean / , if maximum flexibility and at the same time rigidity for heavy duty aluminum jobs is wanted there are only 3 ways to go:
- Make a specific purpose machine from the beginning like Jonathan did for his friend, with removable bed
or if larger machine is desired:
-make a machine similar to how Dean does them, with removable adjustable bed and bottom frame structure
-make a machine the way i propose and use additional bolted beams to raise the bed or lower it, look at the picture below
In these both cases, at the end the material we use is roughly the same. main point in my design being limited space, so the legs can be removed without compromising the integrity of the design and having to redraw it again.



Of course that design can be further developed with cost savings in mind. For example the middle 2 beams from the lower bed can be made dis-mountable, so they can be used for the upper bed and so on... Look picture below.

Thats what i like about the constructive discussion. makes me think and develop. The idea of the removable 2 beams just made me realize what savings that would be on my next 1250x2500 design

[ba99297]

The Bed being a separate unit means the outer frame pieces support the bed material at the edges, the way yours is drawn the edges will be unsupported and could bend or flex under cutting conditions.

Excuse me Dean but i don’t understand why my cutting table has unsupported edges. At every plan the cutting table beams are welded at the side pieces. Could you please make it more clear?

Ba99274 any of the frames shown including your own designs will be good enough so don't worry you won't go wrong.

Thank for your good words. The truth is that i don’t have time pressure. When i make a step i want to be sure about what i am going to do
I am searching over the internet for this construction about a year
The last few months although i thought i knew very much about the subject i find out that mycncuk exist
The answers that i take from the members make me change many of my thoughts
For example i didn’t know that the ball screws have critical speed. My first plan was to connect the motors straight to the ballscrews. Imagine how i would fell when i broke a ballscrew 1500 long running at 3000 rpm. I also thought that I beam is the strongest type of beam..... ( May be for vertical forces )
So every day someone in here give me an advice or correct me, i fell happy about avoiding another mistake

Thanks everybody so much for your help
P.S Sorry for my English

[Jonathan]

Since it's already been said that most of the frame designs posted here would be more than adequate, I'll just highlight some general points which may be useful.

I think the frame design silyavski posted in post #30 is well thought out since, among other things, it makes relatively efficient use of material. I'd be inclined to swap the two 100x100mm beams and 80mm spacers with a single piece, perhaps 200x100mm possibly with a higher wall thickness than the rest (or scaled down accordingly if making the rest of the frame from smaller sections). This will obtain a similar stiffness to the original, but reduces the number of pieces to cut by 12 and the number of welded joints is reduced by 20. This would save a lot of time and reduce residual stresses due to there being less welding. Clearly it may not be as cost effective as making the whole frame from the same size material.

Something else to bear in mind is that increasing the size of the steel you use can sometimes make the structure strong enough, with a lot less work than adding lots of small supporting pieces and without necessarily increasing the cost. For example, if you have a piece of 60mm box section with 3mm wall thickness supported at both ends, then in general increasing it to a 80mm beam of the same wall thickness will reduce the deflection to less than 2.4 times the original.

Also, it's better to increase the size of the beam (within reason), than to increase the wall thickness. Continuing with the previous example, the 80mm beam will cost about 33% (assuming it's priced by mass) more than the 60mm beam, so suppose you instead invested that 33% extra in getting a 60mm box section with 3mm thickness to match the weight of the 80mm beam. Both sections will cost about the same, however the 60mm beam is still only just over half as strong as the 80mm, as you've only made it about 33% stronger (not 240% as above). The reason for this is that increasing the wall thickness only gets a linear gain (y=k*x) in strength, compared to a quartic (y=k*x^4) relationship from increasing the size. So in general, you only use a greater wall thickness when you haven't got space to fit a bigger section.

With regards to resonance, MDF is actually a great material to use for the machine bed as it has such good damping properties. I had a lot less problems with resonance on my machine with the MDF bed than I do now with aluminium. If I'm machining an aluminium part for which the finish is more important than the accuracy, I'll sometimes put a piece of 18mm MDF between the part and the aluminium bed as the MDF damps the vibrations, making the 'sweet spot' to get a good finish a bigger spot.

[JAZZCNC]

Something else to bear in mind is that increasing the size of the steel you use can sometimes make the structure strong enough, with a lot less work than adding lots of small supporting pieces and without necessarily increasing the cost. For example, if you have a piece of 60mm box section with 3mm wall thickness supported at both ends, then in general increasing it to a 80mm beam of the same wall thickness will reduce the deflection to less than 2.4 times the original.

Also, it's better to increase the size of the beam (within reason), than to increase the wall thickness. Continuing with the previous example, the 80mm beam will cost about 33% (assuming it's priced by mass) more than the 60mm beam, so suppose you instead invested that 33% extra in getting a 60mm box section with 3mm thickness to match the weight of the 80mm beam. Both sections will cost about the same, however the 60mm beam is still only just over half as strong as the 80mm, as you've only made it about 33% stronger (not 240% as above). The reason for this is that increasing the wall thickness only gets a linear gain (y=k*x) in strength, compared to a quartic (y=k*x^4) relationship from increasing the size. So in general, you only use a greater wall thickness when you haven't got space to fit a bigger section.

Well ounce again I'm going to have to disagree.!! . . . Not on the strength basis has that's true but on the fact it's better to use larger.? It's only better if the application warrants it and in this case it doesn't. The thicker wall thickness will be more beneficial because it will be less resonant, going Larger on the tube size just increases resonance. For this application then 60mm tube will be more than strong enough so strength isn't a problem so any extra efforts to improve would be better focused on reducing resonance and going thicker on the wall would help here.

With regards to resonance, MDF is actually a great material to use for the machine bed as it has such good damping properties. I had a lot less problems with resonance on my machine with the MDF bed than I do now with aluminium. If I'm machining an aluminium part for which the finish is more important than the accuracy, I'll sometimes put a piece of 18mm MDF between the part and the aluminium bed as the MDF damps the vibrations, making the 'sweet spot' to get a good finish a bigger spot.

Oh dear again I'm disagreeing. . Lol . . I don't have any problems with my aluminium bed and differing finish.? I get the same finish regardless of whether I'm working direct on bed or on any other sacrificial surface. If your bed is affecting finish then reckon you've under built it or your clamping methods are poor.!!
If anything I get more trouble when using MDF than I do when working direct on bed because material can slide if cutting deep, direct on the bed I get much better hold down.
Also Poor finish comes mostly from the tool chattering so if your struggling with finish then I'd be looking at the machine strength and resonance, hence why I put more importance on material Thickness than Size when it comes to steel. (obviously size has to be fit for application)

BA 99297 I'll say it again to be clear "Any of the designs will be strong enough" so if any more focus is needed then I'd direct it to reducing resonance rather than strength.

Oh when I said about unsupported edges I meant you don't have any members to support the Bed base material at the edges. On your designs the edges between table supports won't support the material or any sacrificial material. It may seem a small detail but it can make a difference and for the little extra material/work it's not worth not doing IMO.!