Making A Beam More Rigid.

looking up from head height.

cnc jim

The design is flawed in that you cannot equalize the force applied to each screw clamp. An extra ¼ turn on any clamp loosens the rest. Four matched pneumatic or hydraulic cylinders fed from a common manifold will apply a constant force along the whole length and automatically compensate for and tolerance errors in the assembly. It also has much quicker setting times.

Are you sure it is the top that is bending? From your sketch it is much more likely to be the 50x100x? base angle that is deflecting. Weld a vertical 100x10 underneath the angle immediately below the points where the clamps are acting. Tack welds 1" every 6", every other weld on alternate sides of the bar. Make sure that the bar is firmly clamped to the angle along the whole length then start tacking at the center and work outwards alternating towards each end to keep distortion to a minimum. Replacing the top box sections with a two 100x10 vertical bars (with the clamps between) will improve the top stiffness but fix the base first and the top should not need changing. If you do change the top, bolting with spacers rather than welding the two bars together is a better option because if you do want to automate with pneumatic / hydraulic cylinders later, it will be much easier to make the change. Design with the future in mind.

HI Jhassociates

If the top beam was more rigid, the screw clamp loosening each other wouldn't be such a problem, they don't have to have equal pressure, just pressure, but I agree that pneumatic or hydraulics would solve the problem, but how to do it easily...,

The angle is 8mm thick, and fastened against the corner of a wall, so i doubt it is flexing, but i will check.

Looking at other responses as well, i will see what clamping force i need per clamp, that would be useful info.

And a fabricated beam, bolted through at intervals, that's good. In wood they are known as soldiers, because they remain standing straight.

Well, putting a weighing scale under one of the screws and winding 'till it feels about right, 13kg.

It's definitely the top structure that is bending flexing, not the angle, and of course the tabs connecting it to the base angle splay outwards slightly.

just a thought, what about the 80x40 aluminium machine-build type section as a replacement for the 2 40x40's ?

cnc jim

Ally has about 3 times less stiffness than mild steel, I'm only at Chesterfield if you need any bits

well, not just aluminium box, profiles like this:

You need to know the modulus of elasticity and the moment of interior to work out deflection, this might help to give you some idea

Engineering Tool Box, Deflection of Beams

Inertia, not interior.

Yes i tried to edit it but was too late :-)

So if 40*40 ally section has a Ix of 9.0cm4 would putting 2 back to back have an Ix of 18?

One of the options is to fill the box up with concrete.

A beefy section of angle iron over one or both seams will be better than a flat bar for eliminating flex. (if they will be out of the way of the sliding carriage)

Channel iron would also do the job with less total protrusion.

-Those are some good sized tack welds on the side. I'd double them up before I covered them. I'm always a fan of fully welding a seam.

if it were only flats on the side you are able to add consider going beyond 80 to the 100 (or greater) and protruding from the backside like a stiff fin..

All of these toggle clamps work great.If you have more than one popular thickness to clamp? Consider additional clamps alongside the others to avoid the labor of always changing the popular clamp thickness. I've used the pneumatic type for wood working and they are a true gift from above.

Ok, JE, thanks for the suggestions.

There is no room on the carriage side, so any extra will have to go on the other side; good idea to make it 100, that won't be in the way of anything, and although I did look at the toggle clamps, discounted them because the range of grip required was too great to use without adjusting.

Never though of having two.

cnc jim

In your sketch, the 100 x 50 angle is turned the wrong way, and is a poor choice anyway. Your 100 x 40 box tube idea was way better. And get a new blacksmith.

Hi Tornado

The large flat of the angle is the bed of the device, which the cutter sits on, and there are grips on the bottom of the cutter, whose base is 100 also. It seemed a good idea to go that way, as the device sits round an external corner of a wall, and can be well supported.

Material is passed between this bed and the cutter and scored. Because the device is on an external corner of a wall, I can snap the material against the sharp edge of the angle by bending it beyond where the wall would be.

I chose the angle because it wraps round the wall, and box section corners are too round and cause bends rather than snaps.

The blacksmith is sound, I just shouldn't have let myself be pursuaded. I was looking for the best solution from the forum, which if it involves welding, I could take to him and say 'right, lets put this on here'

If you think 100x40 would do the trick, that might be the best way to go. Is there a way to calculate?

thanks for response.

cnc jim

A key variable for beam deflection is the depth of the beam in the direction that it will deflect, together with the cross-sectional areas of flanges especially, and then webs. (I'm not going to try to describe moment of inertia here.) You could build this out of two pieces of flat bar, if you account for bending. There are too many options to describe in a brief post.

Correct me if i'm wrong but if i remember right 80mm diam round is stiffer than 80mm sq. ( with the same wall thickness. )

Also an ideal beam is thicker in the middle. A la Sydney Harbour bridge.

Jim

Wrong ! The square is MORE rigid because more material is away from center.

I am sure you are aware that a boring bar made from mild steel flexes and causes chatter. A boring bar made from carbide is stiffer. Similarly, machines are made from cast iron and not fabricated from steel, to get stiffness. Your clamp bar is elastic and has flex. That said, you could put a lump ( 40x40 for example ) in the middle and stretch a strip of flat bar over it. Weld one end of the flat bar to the end of your 40 x 40 clamp bar and prestress your clamp bar until the flex is gone then weld the other end. This should remove most of the flex from the bar by bending it like a strung bow. I don't know your clamp pressure but you need to know that by doing this you will be close to the yield point and so a lot of push near the ends could make the bar crimp in the middle.

The ideal clamp bar is cast iron with a hump back.

Jim

Cast iron use is preferred because of its higher internal damping property. Steel has a lower hysteresis.

The other reason is that cast iron has a lower viscosity than steel making the casting better and allowing thinner walls.

Sure, cast iron will flow better than cast steel BUT i was comparing a cast machine with a fabricated from hot rolled steel machine. Bugatti made his early engines from fabricated steel and had no end of problems.

Regular steel that the OP can readily get his hands on will be more elastic than a cast iron bar that he can also readily get made.

Just sayin'

Jim

Wouldn't totally weld it will end up as a banana, hit 50mm miss 250mm should be ample welding wise but stagger the welds so they aren't opposite each other.

Yes I would weld a flat bar to the side or both sides, you could even use 120x6mm flat bar and let it overhang (it's the width that will increase stiffness more than the thickness would) or start again and use 100*40 RHS

Oh yeah - great big arc when done with the welding, for sure.

Your problem is the stiffness of the screws!

I will explain it since at first look it seems ununderstandable.

Which ever the beam stiffness will be, when you tighten one of the crews it will deform and will displace the previously tightened screws up so that their preload will decrease. If you want I can give you the equation with which you can compute the beam deformation at every point as function of the force in one of the screws.

There are 2 solutions for your problem:

a- you use flat cylinders (from Enerpac for instance which I know very well) but it will be a quite important investment since you need as well a pump and connecting hoses and couplings which are not low cost.

b- you tighten via Belleville disc springs which number in the stack and loading capability have to be determined according to the beam stiffness and the clamping force tolerance you can afford. The stiffness of the stacks should be low enough in order to obtain the right result.

I suggest Belleville springs because of the limited space you have.

The stiffness of the beam alone will NOT solve your problem.

Thanks for the idea, will look into it.

cnc jim