Bending of steel plates - how it is done and calculated
 

My question is ..... are the etched marks on the laser cut parts the center of the bends?? Now that I've seen how a hydraulic bender works, I think it has to be. Thanks J.R.

Yes, those marks indicate the center of the "blade" for bending into a "V-block" illustrations

But, lots of things can go wrong in that process . . . . . .

- The blade's tip can be too sharp and leave a sharp dent

- The blade's tip can be worn in the center of the bending press from mostly bending narrow stuff and then when you do a wide plate the bend angle is not consistent along the whole length.

- The blade's tip can be damaged because someone else (of course) tried to bend square bar or round bar.

- The big blunt blade meant for thick plate is used for thin plate. It is a schlep to change blades and the rush jobs . . . .

In the same lazy sense, the wrong bottom V-block is used. The width of the V needs to be about 6 times the plate thickness.

- If the V-block is too wide then the bend radius gets too big.

- If the V-block is too narrow then the bending forces get very high and the tooling leaves marks on the job. Plus there is a risk of the material forming cracks.

Why am I telling you all this? Well, they are all lessons I have learnt in dealing with various companies doing bending. You have to pick a company that bends metal in the range of thickness that you are working with and those companies must have the right tooling. That is why I prefer to use one-stop lasering and bending companies and to stay away from the guys bending 25mm [1 inch] plates all day.

to make one think

You guys might have noticed that the bending drawings do not specify the bending radius. If I had done that, the price would have doubled and we would have solved a problem that no one has reported yet. . . . . .

The marks are placed on the plates at about an average bend radius. Take the spider plate for example - it would not fit inside the y-car if the bender person had used a big radius (wide V-block) and lined it up on the lasered marks. Most people should be finding a millimeter or 1/16" clearance when dropping the spiderplate into the y-car because that is what I allowed for when positioning the lasered mark. So, how does one find the "average" position. . . . . . .

Let's say we want a 3mm [1/8"] thick plate bent into a channel-shape with the outside dimension to be 100mm [4"]. How far apart should the bending lines be? Well, most of us simply subtract the plate thickness - in this case 2 ends go up so the distance between the lines becomes 94mm [3.75"]. And it mostly works out using this simple rule.

Here is the "official" way to calculate the bend line positions:
http://www.sheetmetaldesign.com/Whit...dAllowance.pdf

Suggest you mark the bending lines onto the plates, on the inside of the bend only. Also write the angle of bend onto the plate. Reduces time & mistakes.

Gerald

That is an excellent idea! Thanks

Paul

Gerald,
Great idea. It sounds like you have been there before? :eek::)

Imported as a part the DXF, removed all the anotations and made a flat feature, then added the bend lines and left SolidWorks do its magic (sketched bends)

Here is the result, questions below:

http://img528.imageshack.us/img528/1...estion1pm3.jpg

Keep in mind that I made a direct 1:1 import from the DXF

When I compare the measured dimensions in SW with the ones in the bend drawings PDF, there are small discrepancies as you can see in the image. There is even a 0.5mm discrepance in the piece length (147 vs 147.5mm)

Is this critical, or 1mm more/less is not relevant? I ask because I don't want to model all the parts and later find they don't mate as expected.

Thanks!


Pablo

Practical experience, since before the days of CAD, says that 95% of the time when you bend a channel on two lines, the distance between the lines gives you the inside dimension of the channel. It is as simple as that, and it very seldom goes wrong.

Your CAD program is probably assuming that the neutral fiber is in the center of the thickness, which gives results that don't agree to practice. Tell your CAD program to move the neutral fiber to the inside of the bend, as follows:

Bingo Gerald!

In SW this is called K-Factor, changed it to 0.2732 and now measurements are perfect, I assume this factor does not change with sheet thickness right?
http://img150.imageshack.us/img150/9...estion3ol9.jpg

Thanks again (Im sure I still have a lot of thanks to give :) )


Pablo

Gerald never cease to amaze me with his knowledge and experience.

...and that is why some people have engineering degree's and others just copy great work. :)

May be a good reference for future questions on this subject. Any input is very much appreciated. :cool:

http://www.ciri.org.nz/bendworks/bending.pdf