# Bending 2" ID, 8ft Long "C" Channels From 0.040" 2024-T3

### Help Support HomeBuiltAirplanes.com:

#### TAProwler

##### Member
Hello Everyone,

I am trying to bend "C" shaped channels that are 8ft long and are made from 40 thou. thick 2024-T3. The flanges are 3/4" and the channel has a 2" I.D. Here is a CAD drawing:

So, I start with a straight strip of material that is 3.6" wide and 8ft long. I put the piece into the press brake and bend the 1st flange to 90deg. Here is a section view of the press brake "air gap" bending process that I am using:

When I take the (now) L shaped part out of the press brake - it has a very uniform, gentle bow the entire length of the piece. Here's an exagerated (for clarity), simulated CAD drawing:

If I place a long straight edge along (and tangent to) the side opposite the flange and then measure the distance from the straight edge to the far corners it is just a hair less than 1/8" on both sides. This is over 8 ft, so it is not a lot of bow, but it is defintely there.

My 1st thought was that my press brake was causing this. Then, after a lot of "thinking chair" time - I decided that this a function/property of the metal (material) and not the press brake. I have three main reasons why I think it is the metal causing this bow and not the brake, and they are:

1. I got similar results when trying to use a sheet metal (leaf type) brake. Only it was worse with that brake because I could not get a consistent angle of bend along the entire flange. The press brake will give me a consistent angle of bend if I use a "crowning" technigue to essentially overbend the middle more than the ends.

2. The bow in the material comes out OPPOSITE what you'd think, by using the crowning technique that I mentioned above. I flex the beam that holds the upper dia down into the lower die farther in the middle than on the ends (by about 0.015" - give or take). If anything, this should cause the bow to be opposite of what I am actually getting.

3. It is a property of all metals to thin down when stretched (also known as "necking" - I think). Since the material is being stretched somewhat along the entire radius of the flange, it must be shrinking a little bit every unit length (couple thous per inch for example).

So, I think that what is happening is that the side with the 1st flange on it now has "shrunk" buy a small amount and has caused this nice gentle bow in the piece. Also, the bow is so even and uniform, I don't think the press brake could cause this bowing phenomenon. Seems to me that if it was the press brake, the bow would come out more accentuated in some areas and less in another - like, it would be more crooked than bowed.

What say the experts out there? Have I analyzed this correclty or am I totally out to lunch and doing this all wrong?

Thanks for the help and insight.

R/Todd

#### Jay Kempf

##### Curmudgeon in Training (CIT)
Hello Everyone,

I am trying to bend "C" shaped channels that are 8ft long and are made from 40 thou. thick 2024-T3. The flanges are 3/4" and the channel has a 2" I.D. Here is a CAD drawing:

View attachment 16073

So, I start with a straight strip of material that is 3.6" wide and 8ft long. I put the piece into the press brake and bend the 1st flange to 90deg. Here is a section view of the press brake "air gap" bending process that I am using:
View attachment 16072

When I take the (now) L shaped part out of the press brake - it has a very uniform, gentle bow the entire length of the piece. Here's an exagerated (for clarity), simulated CAD drawing:

View attachment 16071

If I place a long straight edge along (and tangent to) the side opposite the flange and then measure the distance from the straight edge to the far corners it is just a hair less than 1/8" on both sides. This is over 8 ft, so it is not a lot of bow, but it is defintely there.

My 1st thought was that my press brake was causing this. Then, after a lot of "thinking chair" time - I decided that this a function/property of the metal (material) and not the press brake. I have three main reasons why I think it is the metal causing this bow and not the brake, and they are:

1. I got similar results when trying to use a sheet metal (leaf type) brake. Only it was worse with that brake because I could not get a consistent angle of bend along the entire flange. The press brake will give me a consistent angle of bend if I use a "crowning" technigue to essentially overbend the middle more than the ends.

2. The bow in the material comes out OPPOSITE what you'd think, by using the crowning technique that I mentioned above. I flex the beam that holds the upper dia down into the lower die farther in the middle than on the ends (by about 0.015" - give or take). If anything, this should cause the bow to be opposite of what I am actually getting.

3. It is a property of all metals to thin down when stretched (also known as "necking" - I think). Since the material is being stretched somewhat along the entire radius of the flange, it must be shrinking a little bit every unit length (couple thous per inch for example).

So, I think that what is happening is that the side with the 1st flange on it now has "shrunk" buy a small amount and has caused this nice gentle bow in the piece. Also, the bow is so even and uniform, I don't think the press brake could cause this bowing phenomenon. Seems to me that if it was the press brake, the bow would come out more accentuated in some areas and less in another - like, it would be more crooked than bowed.

What say the experts out there? Have I analyzed this correclty or am I totally out to lunch and doing this all wrong?

Thanks for the help and insight.

R/Todd
Is there bow in both flanges. If so it is because you are stretching the area of the bend and so it is bowing up out of the bottom of the V in the weakest direction. Measuring the bow in the middle would tell you how much you are stretching it. If the top die is lower in the middle and higher at the ends this will happen. Also, the material may be grabbing on the bottom die at the ends or at one side of both ends. Check the radii at the bottom die and see if there is anything that would give the material traction on the bottom die and polish or lube to solve. Last but not least. Have you tried samples in both directions across the grain and with the grain?

HBA Supporter
Log Member

#### TAProwler

##### Member
Hey Jay, Thanks for the help.

Is there bow in both flanges. If so it is because you are stretching the area of the bend and so it is bowing up out of the bottom of the V in the weakest direction. Measuring the bow in the middle would tell you how much you are stretching it. If the top die is lower in the middle and higher at the ends this will happen.
I agree (now). It appears to have the bow in both flanges, similar to what is shown in the link that Hot Wings provided. I'm pretty sure that the camber phenomenon is what I am seeing. There doesn't seem to be much I can do about the bow happening, but I think I have found a way to deal with it so that I can get a straight flange on the other side. Which is all that I really need.

Also, the material may be grabbing on the bottom die at the ends or at one side of both ends. Check the radii at the bottom die and see if there is anything that would give the material traction on the bottom die and polish or lube to solve.
That is a great observation. I try to keep a thin layer of grease on the bottom die. Originally I also greased the top die. But, I found that the pieces slip a lot less (causing a non-uniform flange length) if I keep the top die mostly dry.

Last but not least. Have you tried samples in both directions across the grain and with the grain?
I've only done the long pieces with the grain. I don't have access to long lengths across the grain. However, I have noticed the camber effect either with or across the grain.

#### TAProwler

##### Member
Great link Hot Wings. This adresses exactly what I am seeing. Thank you for posting this. As I mentioned above, now that I know that it more a property of the material, I think I have found a way to work around the bow and still get a straight flange on the other side. This was a great help.
Thank you.
R/Todd

#### Hot Wings

##### Grumpy Cynic
HBA Supporter
Log Member
This was a great help.
Thank you.
R/Todd

Glad a bit of triva from my memory cells was of some help.

#### orion

##### R.I.P.

There is commonly another contributor to this, that being the spring-back. For instance, if you try to get a perfect 90 deg. angle you will do so only on the edges - towards the middle of the part the bend ends up something less than 90. The result is a bowed, non-uniform flange. Sheet metal shops have design schedules for the brake blade where the center of the blade is set down a bit further than the ends, resulting in more of a bend towards the center of the part where the spring-back is more prevalent.

To mitigate some of this, most part designers also include a bend at the outer edge of the flange, thus stabilizing the free edge. It's not a total solution but it can help. The differential stretching on the two faces is a bit more difficult to deal with but a bit of experimentation can often get you close.

#### TAProwler

##### Member
There is commonly another contributor to this, that being the spring-back. For instance, if you try to get a perfect 90 deg. angle you will do so only on the edges - towards the middle of the part the bend ends up something less than 90. The result is a bowed, non-uniform flange.
Agreed. This "canoe-ing" effect you see is also (mostly) a function of the material. You get LESS spring-back at the ends and MORE spring-back in the center (pretty uniformly - I might add). The net affect is a flange that is "canoe" shaped. This I had previously identified and is the reason why I abandoned using a leaf style sheet metal brake and scratch-built my own press brake for bending these channels (and a few other parts). The leaf brake I had would bend mild steel just great, but the spring-back (tensile nature) of 2024-T3 is just too much for it. This greater spring-back in the middle cannot be effectively dealt with when using a leaf type brake. Which leads to your next comment.

Sheet metal shops have design schedules for the brake blade where the center of the blade is set down a bit further than the ends, resulting in more of a bend towards the center of the part where the spring-back is more prevalent.
Before building my press brake, I studied up on them for a while. The feature you are referring to is what the press brake industry calls "crowning". In the new CNC press brake designs this schedule you mentioned is programmed into the controller and it automatically compensates for beam deflection and this differential spring-back effect with something they call "Auto-crowning." There is litterally a seperate huge ram that will flex the middle of the die beam to produce this crowning. Pretty nifty, with a very nifty price tag to boot. In my original post you will see that I am doing a form of this crowing in my bends - and it does overcome the spring-back issue. But as Jay mentioned, I am also stretching the material more in the middle and causing more camber. It's all a compromise/trade-off, I guess.

To mitigate some of this, most part designers also include a bend at the outer edge of the flange, thus stabilizing the free edge. It's not a total solution but it can help.
Good idea. I'll make a note of this. The idea I keep coming back to (to deal with this phenomenon) is to somehow grab the ends of the flange and stretch the entire radius the longitudnal (long 8 ft) way. The risk here would be snapping the material if stretched too far. I could also try to shrink the very wide flange in several places to straighten the piece before putting it back into the press brake to bend the 2nd flange. So far I've opted to just use bending blocks in 3 places (middle and ends) between the 1st flange and the upper die to force the ends of the piece in tight to the die. Like this:

This ensures uniform channel width, but causes the wide flange (before the 2nd flange) to buckle slightly. However, once the 2nd flange bend starts, the stretching from that offsets the buckles - somewhat. I get a few small buckles in the 2nd flange of the channel - but I can live with that. If they are really noticable I can put a quick, light squeeze on that area with a shrinker and it will pretty much straighten it out.

The differential stretching on the two faces is a bit more difficult to deal with but a bit of experimentation can often get you close.
I've been doing a lot of that - and I think I am getting close to acceptable results.

Thanks for sharing.
R/Todd

Last edited:

#### TAProwler

##### Member
Press brake success!

I just wanted to post a follow-up on the flange bending. I was really using these 2" ID channels as practice on, and to calibrate my purpose built press brake. After learning more about the metal/material properties in this thread, I bent up one last 2" channel and it came out pretty decent.

With that accomplished, I finally got brave enough and put the two pieces that I really needed bent into the press brake. All the practice on the 2" ID channel must have paid off, becasue my main spar outboard channels came out very nice. They're not perfect - but the press brake does a much nicer bend than I could ever have gotten with a sheet metal leaf brake. Here's a pic:

With the wider flanges on these pieces, the camber effect is a lot less (almost not noticable) than on the 3/4" flanges of the 2" ID channels (which makes sense). The most important thing is that the bends are straight and uniform. The crowning technique (slightly overbending the middle more than the ends) works very well. The flanges are a little wavy, but the bends placed up against a straight edge are very straight. I'm really pleased.

I also went on to use the press brake to bend the foreward skin flanges for the main spar center section. Here's a pic:

All the pieces came out with nice uniform 90deg bends. Since they had 3/4" flanges - I did get a lot more of the camber bowing, but a VERY light application of the shrinker to the flanges in several spots helped to straighten them perfectly.

Thanks for the help and information needed for me to figure out what was going on with my bends (what was being caused by the metal and what was being done by the machine). My parts came out great in the end (with many more to make). I really appreciate everyone's input and insight.
R/Todd

#### Edokus

##### Member
Hi again.
My previous reply may or may not have gotten through.

I'm also required to fabricate C channels for my current project and the quote I received from a local commercial shop ia beyond my means.

Ed Dokus

edokus@comcast.net

#### TAProwler

##### Member
Hi Ed,
Sure. I don't have any kind of "plans," but I do have some pictures of the fabrication.
We made it from a bunch of "I" beams I had laying around.
It's basically an "H" frame design. The top beam is stationary and holds the upper die.
The middle beam slides up and down and holds the lower die.
The lower beam(s) serve as the base and is what the hydraulic jacks push against.
Here are some pics, starting with the base triple beam:

Here is the sliding (middle) beam setup:

Here is the initial construction showing side beams welded to the top beam (upside down):

And, finally the initial basic construction completed:

Machining the upper die (punch die):

Just after fabrication of the lower die and installed.
Lower die is made from 1/4" X 2" angle iron with 1/2" spacer bolted between the angles.
The upper edges are machined with a smooth 1/8" radius (polished with scotchbrite rol-lock disk):

Installed back-gauge on back side of lower die (to quickly set flange length):

Installed cheap HFT digital calipers (modified) to measure plunge depth accurately:

Installed a system that I call the "deflecto-meter."
It has a 3/16" x 4" flat strap rigidly attached to the ends of the sliding beam.

The middle floats up/down with the ends of the sliding beam.
The middle pushes on the dial indicator and indicates beam deflection.
It's probably somewhat crude, but it is effective and it works:

The picture that is second one above, also shows the hydrualics system installed.

R/Todd