PDA

View Full Version : Scaling up project dimensions



QSIMDO
10-27-2018, 02:05 PM
I found a project I'd like to build (JRW sheet metal brake) but it's a bit lighter than I'd like.
Would scaling up the component sizes by a percentage factor be a successful way to do this?

Tony907
10-27-2018, 02:44 PM
I'm no engineer but I would think scaling to the force required for a bend would be a better way to ensure performance. Take for example, (it's an example...) if it took 1 ton to put a 90 bend in 1/8" mild steel, that means that 2 tons would be sufficient to do the same bend in 1/4 mild steel. The same consideration needs to be applied to large increases in length.

Like I said...not and engineer, and my advise is worth what you paid for it.

BCRider
10-27-2018, 03:19 PM
I think the idea is fine in principle. But when it comes to doing so it won't be as easy as you might think. First off material only comes in given sizes. Second is that if you alter those sizes you'll also need to alter the parts that connect to them to adjust for the size changes. So basically you either re-engineer it on the plan and make a whole new plan or you do the re-engineering on the fly in the shop. And finally if the goal is to size it up by, say, 50% and use 50% larger sizes, if they are available, then the other issue is that the metal does not flex in a scale like manner. If you enlarge the whole machine then it's quite possible that you need to increase the size of the material by more than the scale factor to achieve the proper stiffness for the middle of the span of the parts.

Tony, you posted above;


Take for example, (it's an example...) if it took 1 ton to put a 90 bend in 1/8" mild steel, that means that 2 tons would be sufficient to do the same bend in 1/4 mild steel.

But that is not the case. The 1/4" material is more than twice the stiffness and strength of the 1/8". The bending strength goes up by the square of the increase in thickness. So two pieces of materials of the same size but one twice as thick will see the thicker one being four times as strong. Put a piece of 1/8 x 1 in the vise and bend it. Then repeat with some 1/4 x 1. WAY more than double the effort is needed.

Paul Alciatore
10-27-2018, 04:02 PM
I'm not sure what you mean by "lighter".

If you mean it is not strong enough to do the work it should, then if you just scale up all the dimensions will only make it more flimsy. What is really needed is to "beef up" the parts at their existing size so they can do the work.

But if you mean that it does not have a large enough work envelope and you need to work on larger parts, then scaling up MAY work. But you need to consider that as size goes up, the need for stiffness increases. So if you scale up a 12" brake into a 36" brake, a 3X increase in the work envelope, you may need to increase some of the dimensions by more than 3X. This is called engineering. It can be done with a lot of math calculations or it can be "seat of the pants" engineering. "Cut and try." or whatever method you choose.

The common idea here is that if you just scale up all the dimensions by the same factor, the machine that you wind up with will be less likely to be able to do the intended work, even on the same gauge of metal. There is a fundamental principle at work here. The simplest example of it is a simple, horizontal beam that is supported at both ends. Assume you have an I beam that bridges a certain span. This could be a bridge or a roof or whatever. Now if you just scale up that I beam, some would expect it to get stronger. But if you include all three dimensions in that scaling, then the length also increases. The strength of that I beam will increase with the cross sectional area which will be the second power of the scale factor. But the weight of the beam will increase by the third power of the same scale factor. So the weight of the beam increases faster than it's strength. This is a case of diminishing returns and sooner or later you will reach a point where the beam will fail/sag under it's own weight. So there are limits on simple up-scaling.

You can probably use the general design of the original machine, but you should consider the needs of each part individually. Many of them will need to have their cross section increased.

LKeithR
10-27-2018, 04:04 PM
...But that is not the case. The 1/4" material is more than twice the stiffness and strength of the 1/8". The bending strength goes up by the square of the increase in thickness. So two pieces of materials of the same size but one twice as thick will see the thicker one being four times as strong. Put a piece of 1/8 x 1 in the vise and bend it. Then repeat with some 1/4 x 1. WAY more than double the effort is needed...

Exactly! What I'd be doing is looking at machines of the capacity you need and designing around that. Remember that virtually
all of the cheaper import brakes are built lighter than what you really need to do the job--they will flex and give more than is
acceptable unless you're working with really light material. Mass is important in these machines and a lot of the lighter ones
simply don't have it...

BCRider
10-27-2018, 04:23 PM
HERE IS A LINK (https://grabcad.com/library/jrw-sheet-metal-brake-1)to the brake I think you mean.

If you compare the design to the larger ones I think you'll agree that larger brakes have more stuff on them to enhance their stiffness. Like the truss rods that triangulate off the pieces to enhance their stiffness. So I think you'd want to do more than just scale up the JRW design. To handle larger work and particularly heavier gauge steel it would need a lot of major changes that would make it similar to a proper commercial brake of the size you're after.

QSIMDO
10-27-2018, 05:27 PM
Here's a PDF of the plans; http://www.homemetalshopclub.org/projects/jrw_sheet_metal_brake.pdf

What I'm thinking is if I add, say, 25% to every size called out it should be able to handle 18 gauge, full width, with
no problem.
As it is I think capacity is only 20 or 22 gauge.

BCRider
10-27-2018, 07:21 PM
What I'm seeing is all steel plate. And with the plates sitting "flat" at that. But the bigger units that are strong enough to do things like 18Ga for full width tend to use box and angle sections along with those V trusses to aid with stiffness. Like THIS ONE (https://www.ebay.com/itm/36-in-Metal-Brake-with-Stand-metal-bender-Bend-sheet-metal-mild-steel-and-stai-/253155673320)

In searching for examples I also found THIS PAGE (https://www.instructables.com/id/DIY-Sheet-Metal-Bender-Brake/) on a DIY metal brake. Haven't watched it yet but it may be more in tune with what you want.

J Tiers
10-27-2018, 08:58 PM
That scaling and stiffness works for machine parts also.

If you double the depth of a beam, without changing thickness, you get an increase of stiffness. If you do it proportionately in depth, width and thickness, you increase by a considerably larger factor.

So a spindle, for instance, that is increased proportionately to 3x the OD, would be something on the order of 27 times stiffer than the original, taking into account the increase in basic diameter, and the increase in material area. A cube factor.

So the bender example, if you just make it take wider material, with no change to any other dimension, you lose stiffness. If you make it proportionately larger in all dimensions, then you will get an increase in capacity to go along with the size.

JRouche
10-27-2018, 09:15 PM
Ahhh, I didnt read any of the other posts. Just yours and I would say when scaling up you have to scale in a logarithmic "pattern" and not linear.

That means you have to know the exact weight and needs at every stage of the up scaling. If that makes any sense.

Figure out what you need and build that. You could prolly incorporate all the details of the brake you want too. Dont make it flimsy. JR

QSIMDO
10-27-2018, 11:29 PM
What I'm seeing is all steel plate. And with the plates sitting "flat" at that. But the bigger units that are strong enough to do things like 18Ga for full width tend to use box and angle sections along with those V trusses to aid with stiffness. Like THIS ONE (https://www.ebay.com/itm/36-in-Metal-Brake-with-Stand-metal-bender-Bend-sheet-metal-mild-steel-and-stai-/253155673320)

In searching for examples I also found THIS PAGE (https://www.instructables.com/id/DIY-Sheet-Metal-Bender-Brake/) on a DIY metal brake. Haven't watched it yet but it may be more in tune with what you want.

Yeah, I've searched for hours and turned up as many ads and plans as can be found on the web.
Most are making some really outrageous claims for capacity.
Of course, when I wasn't interested, Craig's List was lousy with them...decent ones too.
Now?
Broken, cracked, missing fingers, etc.

QSIMDO
10-27-2018, 11:31 PM
If you make it proportionately larger in all dimensions, then you will get an increase in capacity to go along with the size.

That's what I'm looking for.
Thanks.

Lee Cordochorea
10-28-2018, 02:06 AM
I don't mean to pick on Tony. I just happen to have researched this stuff recently.

Rigidity is proportional to the cube of the thickness. A 1/4" plate will need 8x as much force to bend as will a 1/8" sheet of the same material.

Longer levers to get more force mean the levers have to be thicker - deflection (bending) of the lever is proportional to an exponent of the length.

Another route... assuming the brake (and your project) can handle contact with higher temperature, just heat the sheet or plate to incandescence prior to bending.

RichR
10-28-2018, 08:27 AM
... Another route... assuming the brake (and your project) can handle contact with higher temperature, just heat the sheet or plate to incandescence prior to bending.

But then your pencil mark that shows where to make the bend may no longer be visible.:)

Mcgyver
10-28-2018, 10:38 AM
Another route... assuming the brake (and your project) can handle contact with higher temperature, just heat the sheet or plate to incandescence prior to bending.

have you tried that?



What I'm thinking is if I add, say, 25% to every size called out it should be able to handle 18 gauge, full width, with
no problem.
.

If the original could handle 22g and you want 18g.....that's a 60% increase in force you're going to need to apply - assuming the 22g was realistic in the first place. If you want the brake to have the same stiffness as the prototype would at 22, you going need the parts it be several times the size not 25% because of the exponential nature of stiffness, assuming the same bend length.

What are you wanting to bend? Bending metal takes a lot of force, just look at the size and weight of commercial brake (not that BS often offered up on offshore hobby stuff) ratings. imo whats practical is a very light folding style brake (I make the finger brake below) or a hydraulic press with dies. Unless it is a constant requirement, I'd consider outsourcing the odd job that went outside of the work envelope of those two

https://i.imgur.com/HRlDTZQ.jpg

Sparky_NY
10-28-2018, 10:38 AM
But then your pencil mark that shows where to make the bend may no longer be visible.:)

That is why soapstone is very common in marking metal.

QSIMDO
10-28-2018, 11:01 AM
have you tried that?



If the original could handle 22g and you want 18g.....that's a 60% increase in force you're going to need to apply - assuming the 22g was realistic in the first place. If you want the brake to have the same stiffness as the prototype would at 22, you going need the parts it be several times the size not 25% because of the exponential nature of stiffness, assuming the same bend length.

What are you wanting to bend? Bending metal takes a lot of force, just look at the size and weight of commercial brake (not that BS often offered up on offshore hobby stuff) ratings. imo whats practical is a very light folding style brake (I make the finger brake below) or a hydraulic press with dies. Unless it is a constant requirement, I'd consider outsourcing the odd job that went outside of the work envelope of those two



Always a catch, isn't there.
I'm looking for maximum 18 gauge capacity, 24" wide.
Anything heavier I have a rudimentary press brake but it's such a PITA to use on quick and light jobs.

Nice looking machine you've got there.
What's it handle?

Mr Fixit
10-28-2018, 12:21 PM
Mcgyver,

That is a very robust looking pan brake and well executed, really nice work! I too have been researching a DIY pan brake to do maybe 18ga steel and aluminum sheet. Is yours a design to handle that ga? Of course the big question is do you, or would you share the design with others? Is there possibly a build post here that we could follow? I know lots of questions, but this has come up before and your design looks like it would fill the need of many of us HSM types.

TX
Mr fixit for the family
Chris :)

Mcgyver
10-28-2018, 01:04 PM
thanks for the kind remarks guys



What's it handle?

i call it brake :) not 100% sure what your'e asking?


Of course the big question is do you, or would you share the design with others?

It was done as a an article in our benefactor's magazine, Home Shop Machinist. As its available as back issues and there are discussions on a possible book collecting a few articles, I'd like to encourage support for HSM so haven't shared it outside of that. The idea came from the Joe Williams design, with credit given, (thats the JRW of the brake mentioned above).

QSIMDO
10-28-2018, 01:18 PM
Wise guy, eh? ;)
nyuk nyuk
That wasn't supposed to be "what's".

I'll rephrase;
My, that's a sturdy looking device you've constructed there. What are the maximum thicknesses of mild steel, aluminum and SS it can bend (given you possess sufficient strength) across the entire width and still produce a satisfactory radius?
I'm sorry the radius question is vague but the correct terminology to result in a quantifiable answer escapes me at the moment.
Take a stab at it.

DR
10-28-2018, 01:29 PM
To expand a bit on what others have said. Deflection or bending is usually proportional to the cube of the width (length).

For example, if you have a cantilever beam of 1 foot long with a certain deflection from a weight on the end and you want it to be 2 feet long with the same deflection under the same load you need the beam to be eight times the stiffness. 2 x 2 x 2 = 8.

There's a free download "engineering power tools" that you can use to size components of structures. For the above example you could determine the moment of inertia of the 1 foot beam, then try various sizes and orientations to find one that's 8 times as stiff.

Mcgyver
10-28-2018, 02:10 PM
Wise guy, eh? ;)
nyuk nyuk


I called it a 20 gauge brake but have never bent 20 gauge full width, can't specifically remember, but might have picked 20 because that what Joe's claimed his does

J Tiers
10-28-2018, 03:43 PM
Point being that if you proportion up a 20" to be a 40", evenly in all dims, it will have 8x the stiffness, and should do the same thickness material at the new width.

If you want to do THICKER material, then scale it up in thickness or whatever at the 20" size to what will handle that new material, then scale THAT to the 40"..... Then it will be able to do the 40" at the new size.

Lee Cordochorea
10-28-2018, 09:07 PM
But then your pencil mark that shows where to make the bend may no longer be visible.:)

That's why we have chisels. ;)

JRouche
10-29-2018, 01:40 AM
That's why we have chisels. ;)

Yes. We break ships the same way in Indonesia...

Chisels are the original tool. JR