MadModder
The Craftmans Shop => General Crafts => Topic started by: jcb121 on July 21, 2015, 05:09:33 PM

Hey, long time lurker here, know a few of you from the mycncuk forum,
So I have a 30" sheet metal folder and folding 1.5mm aluminium is pretty easy.
Is there some sort of method or equation for producing accurate bends? I'm not so much talking about the angle, but more the expansion, crush and radius of the material.
If any of you have some tips, I'd greatly appreciate it! Thanks
JCB.

Hi, and welcome  have some reading material :

:bow:
That's just what I needed! I knew there was a bit more Math behind it.
Just a question though, If I calculate the Kfactor of a 90 degree bend, Is that the same Kfactor for a 45 degree bend?
Put better:
Is the Kfactor a constant or is it per bend?
Thanks again.

JCB,
Have a look here....
https://www.oemmfginc.com/sheetmetalbendingbendallowanceandkfactor/
Hope this helps...

It depends on how accurate you have to be. When I owned and ran a fabricating shop where supper accuracy was not that important I generally just used inside dimensions. For example if you wanted a 1" leg on 1/8" thick stock the bend line was at 7/8" and 15/16" on 1/16 thick stock. Using the type of bending equipment available in most home shops that is probably as good as you need. +/ .010 means nothing when you are eye balling the bend line on a hand brake.

So I tried figuring out the Kfactor, but my calculations never came out right!
So I think the Kfactor can go and get bent :whip:
I just score a line with a 3mm offset and it seems to give me OK results!
My Alu Cases are coming along ok, I need to find an aluminium welder tomorrow as I'm leaving Sunday morning! £80 per hour seems a bit steep to me, that's like 1/5th the welder cost.....I know there's more to it then that but still.
Thanks again all! :nrocks:

JCB,
The kfactor is one aspect making up the Bend Allowance (BA). The problem with the kfactor is that it is taken out of context. The kfactor is the distance (usually towards the IML Bend face) that the neutral axis shifts during a bend divided by the thickness of the metal. The neutral axis radius (from the center of the bend) is reasonably well approximated by: rho = t/{ln((r + t)/r)  where rho is the neutral axis radius, t is the thickness of the piece being bent, and r is the inside bend radius of the part.
The neutral axis distance from the IML Bend face is t  rho = r  t/{ln((r + t)/r}. [Now, mind you, the value of rho is a reasonably accurate approximation as the actual path taken by the neutral axis is a rather obnoxious (mathematically) elliptical arc  I spent much of an academic year calculating the equations of this elliptical arc back in the mid1970's. The "error" in rho is rarely greater than 0.05%.] The neutral axis shift component of the Bend allowance if readily calculated as: Theta X rho  where Theta is your bend angle measured in Radians.
The other "factors" in a Bend allowance are: (1) the overbend allowance required to yield the material to hold the bend; (2) the crushing of the material required to yield it to hold the bend; and/or (3) the other tooling conditions required to make the bend. You would add (or subtract) those to the neutral axis distance to complete your Bend allowance. Items 2 and 3 in the above list are rather complex Herztian contact stresses and associated Poisson's ratio "contractions" that are most commonly used as teaching material in advanced stress/strain analysis courses.
So, unless you are involved in high production rate/close tolerance sheetmetal work, using the kfactor value as defined above will give you a result closer to your desired result than anything other than experimental measurement evaluation.

JCB,
The kfactor is one aspect making up the Bend Allowance (BA). The problem with the kfactor is that it is taken out of context. The kfactor is the distance (usually towards the IML Bend face) that the neutral axis shifts during a bend divided by the thickness of the metal. The neutral axis radius (from the center of the bend) is reasonably well approximated by: rho = t/{ln((r + t)/r)  where rho is the neutral axis radius, t is the thickness of the piece being bent, and r is the inside bend radius of the part.
The neutral axis distance from the IML Bend face is t  rho = r  t/{ln((r + t)/r}. [Now, mind you, the value of rho is a reasonably accurate approximation as the actual path taken by the neutral axis is a rather obnoxious (mathematically) elliptical arc  I spent much of an academic year calculating the equations of this elliptical arc back in the mid1970's. The "error" in rho is rarely greater than 0.05%.] The neutral axis shift component of the Bend allowance if readily calculated as: Theta X rho  where Theta is your bend angle measured in Radians.
The other "factors" in a Bend allowance are: (1) the overbend allowance required to yield the material to hold the bend; (2) the crushing of the material required to yield it to hold the bend; and/or (3) the other tooling conditions required to make the bend. You would add (or subtract) those to the neutral axis distance to complete your Bend allowance. Items 2 and 3 in the above list are rather complex Herztian contact stresses and associated Poisson's ratio "contractions" that are most commonly used as teaching material in advanced stress/strain analysis courses.
So, unless you are involved in high production rate/close tolerance sheetmetal work, using the kfactor value as defined above will give you a result closer to your desired result than anything other than experimental measurement evaluation.
I bow to your knowledge... but you lost me after the first sentence; I have usually pencil marked the bend, clamped in the vice, sometimes using two pieces of angle iron,
& persuaded the object with a potential to kinetic energy convertor to move it to the required position... unless it's plastic whereas I persuade it with a heat gun... so far so good! after all I am not a professional / BSc qualified sheet metal fabricator, just some one who enjoys model engineering as a hobby with the adage... K.I.S.S!... no disrespect of course.

A bit of a dead thread, but since it has been resurrected I shall submit the following...
Way back when I first started learning how to bend sheet metal, I was taught a somewhat "rule of thumb" method of using set backs. The person who taught me this, although a real bastard to deal with, knew his fabrication skills very well. I credit him for some of the skills I retain to this day. I am also plenty happy to be rid of him. :) But there I go digressing again....
Attached is a chart that I derived beginning with his memorized values for setback in bending steel sheet. From those, I was able to come up with a simple formula that gets you very close without the complexity of the Kfactor method. The formula used in the chart is:
SETBACK = (2 * THICKNESS)  SHRINKAGE
Where SHRINKAGE = 1/3 * THICKNESS
It has been quite a while since I used this, as I just use SolidWORKS now to develop a blank for bending, but if I recall correctly the blank length is made up by unfolding all of your true lengths then subtracting ONE SETBACK for each bend from the total bend length. In other words, you SET BACK each bend by the amount in the chart.
Attached image and spreadsheet versions of the aforementioned chart.