How to eliminate plate warpage

[JCC]

How can I eliminate or significantly reduce warpage of large/long parts? Is the answer to use a different process or buy a laser cutter?

For example, with a 10 Ga flat pattern part that is ~48 long and 12 inches wide and has some 7/16" holes in various locations, such as putting a single row of slots every 4 inches along along the far left and right sides. The part comes out with a very slight warp if you look down the edge. I then go to bend 1.5" flanges down the long edges of the part on both sides to form a U shaped part. What usually happens is the finished part is curved like a banana down the long direction. This makes fabrication difficult, as the parts are no longer straight.

I have similar issues with 3/16" and 1/4" plate. I have optimized the settings to have minimal dross, and if I speed it up any faster, I'll get a lot of dross on the parts. I have been using a 40A (0.7) nozzle on all of these thicknesses, which is the largest I can use on the small table, but I can go to 1.1 on the 90A table. I haven't really spent much time testing the larger table, but 2 years ago I cut some ~10ft long, narrow parts, which warped significantly.

I have been using a 40 A Komatsu Rasor 4 x 8 table, which is a down draft table, but I can also (but haven't spent much time) use a 5x10 90A komatsu Rasor table.

Do I need to mirror the dxf to make sure the burning is on the opposite face as the bend points? Or do I need look into shearing or a turret punch? Or get a laser cutter? Or do something different with the plasma I have now?

[djreiswig]

If the cutting of the holes is causing the warpage, make sure you move around to spread the heat out. Cut a hole or 2 on one end, then go to the middle and cut some, then go to the opposite end, then back to the start. If the perimeter cutting is causing it, then maybe leaving some tabs might help by holding the part to the skeleton while cutting. Then you could also move around and cut it in sections. A shear would give you a straight edge, but the holes might not be located correctly when you're finished if the part is warped from the holes.

[Deezl Smoke]

I do quite a bit of this same sort of cutting and bending. 16ga to 1.25". ArcMax 6'x12' table. I run liquid in my table and keep it pretty close to the plate. Makes a mess, but it helps reduce warpage a lot. One thing that DOES make a difference is the metal quality, and the condition of it before you put it on the table. Try flipping the plate over and see if it cuts with less warp. The seller's system is "supposed" to use state of the art straightening/flattening tooling to make the plate flat as it comes off of the coil. But I have been having issues with certain sellers that prioritize throughput over quality. I do specify "cosmetic" when ordering for whatever good it does.

As djreiswig mentions above, tabs are an option, but will slow production some and create a bit more post cutting work.

The thing that concerns me about your situation is your mention that post bending you still have a canoe. This is the brake set up. Even a warped piece of 10ga, should come off the brake straight as an arrow. 4' long by 1' wide is definitely enough to need crowning attention. For 10ga, you should be using about a 1.5" die. Even a larger radius with say a 2" die will still need crowning adjustments. But the end result is even a warped piece post cutting, should come off the brake arrow straight.

Couple things I would try so as to gain new insight are, cut 1 piece from the plate, then flip the plate and cut a piece. See if the warp reverses with the plate. This will tell you a lot about the plate quality and what role it plays in your issue. 2,on a scrap piece, brake one lip, then flip the plate and brake an opposing lip. In other words, instead of a U, form a Z. Take note and compare to the U.

[JCC]

I appreciate the replies. What I'm experiencing is not canoeing. For say a 4ft long part, the bend angle from one end, the middle, and far end vary within .5 to 1 degree, such as 89.5 to 90.5.

What I'm experiencing is more of a banana shape, such as from this reddit post:


Now that I have tried to clarify, do you still think this is a crowning issue? Or flanges being too short (1.5 in flange for 10 ga). This press does have a manual compensator bed... will making any leveling adjustments do anything to prevent the banana bow shape?

I have been using a 1 inch die, which was chosen from a suggested table I found. Are you suggesting I should use a 1.5 in die?

[JCC]

I should also mention that I am using hot rolled A36. Should I be using CRS?

[weldguy]

Forgive if this has been mentioned, i haven’t read the entire topic. How close are the slots to where you bend? Any pics?

[JCC]

https://photos.app.goo.gl/mDZPugoHQLDaTFRr8

Slots are generally 3 or 4 inches apart and in the middle of a 1.5 in flange.

[weldguy]

Interesting, wish I had the solution for you. If you were to bend that same material without any profiles plasma cut into it would it not "banana" like that?

[JCC]

Like are you saying to bend a blank piece cut via a shear? Or are you also saying to punch the holes as well and try to bend it?

[weldguy]

I was referring to bending a blank piece without all the heating and hardening from the plasma.

[ROKCRLER]

Is the part banana'd before bending the flanges?
How are the flanges bent?

[JCC]

No. The part appears floppy and will lay flat against whatever surface it's on.

The part is bent on a 12 ft brake press using a 1 inch die. The edges are bent 1.5 in long each. The press has a compensator on it, but I can't say if it's dialed in right. I can say that the actual bent angle from one end of the part varies within 1 degree. Say, 89.5 to 90.5.

[ROKCRLER]

Reason I asked, I was having similar issues bending on my break I noticed the center would under bend due to alignment causing the banana effect.
After adjusting it was good.
Yours might be having the same issues if it lays flat after cutting.

[SegoMan DeSigns]

When was the last time you put a straight edge on your punch & die? I would still do a bend on some blank material for verification.

[B Holmes]

Your issues are most likely caused by the heat affected zone of the plate. I deal with this all the time on hot rolled sheet.
The material actually heat shrinks along the cut edge. It isn’t usually noticed until forming short flanges.
I have never tried using cold rolled sheet, might be worth a try. I would assume this would happen on a laser as well, just maybe not as bad. Maybe different cutting gasses could also make a difference.

[JCC]

Your issues are most likely caused by the heat affected zone of the plate. I deal with this all the time on hot rolled sheet.
The material actually heat shrinks along the cut edge. It isn’t usually noticed until forming short flanges.
I have never tried using cold rolled sheet, might be worth a try. I would assume this would happen on a laser as well, just maybe not as bad. Maybe different cutting gasses could also make a difference.

What works best for you?

[JCC]

When was the last time you put a straight edge on your punch & die? I would still do a bend on some blank material for verification.

I'll give this a try as I have time.

[B Holmes]

Your issues are most likely caused by the heat affected zone of the plate. I deal with this all the time on hot rolled sheet.
The material actually heat shrinks along the cut edge. It isn’t usually noticed until forming short flanges.
I have never tried using cold rolled sheet, might be worth a try. I would assume this would happen on a laser as well, just maybe not as bad. Maybe different cutting gasses could also make a difference.

What works best for you?

I don’t do large production runs so usually I just straighten the part by hand if it is necessary. I also take this into account during the design phase of the parts, making longer flanges where possible. This can be a problem if someone else is designing the parts.
If you have a shear available, I would shear a blank and then index the blank on the plasma table and cut the interior details. If it forms straight then at least you have identified what the problem is. If it still forms with a curve, I would shear and form a blank with no plasma cutting to prove that it is, or isn’t part of the equation.

[JCC]

When was the last time you put a straight edge on your punch & die? I would still do a bend on some blank material for verification.

So I tried bending a piece of 3/16" that was about 6'-2" long. I bent it into a channel profile that was 3.5" wide, 1-1/4" tall on one leg, and 1-1/2" tall on the other leg. The piece was cut via shearing, rather than the plasma table. After bending both legs, this piece seemed similarly bowed/looking like a potato chip.

Does this point to a crowning issue?

[SegoMan DeSigns]

Does this point to a crowning issue?

That would be my guess without knowing the specifics of the press brake's style or condition. Like everything else they need a once over to check for wearing pars.