Bending 1/4" aluminum plate

[SRT Mike]

I want to make a run of motorcycle license plate frames. A little different than whats out there but you probably know the kind of thing I'm talking about.

I want to take some 1/4" plate and then mill some features (bolt holes, grooves for wires for tag light) then I want to bend it to fit the curvature of the rear fender of the bike. Then I will take a 2nd piece and mill the center out to make it into a trim ring, then bend it to the same shape so that the two pieces will fit together and hold the tag plate in them.

The preferred material would be 6061-T6 aluminum. I'm guessing the radius of the bend would be 6 to 10 inches depending on application. The size of the aluminum would be approximately 6" x 4" x 0.25".

Questions...

Will 6061-T6 bend to this spec without cracking?

Any tips or advice on doing the bends?

What size press would I need to do a bend like this? I'm thinking I'd probably pick up a used hydraulic press for this job, but depending on the capacity maybe use one of those hand-operated hydraulic ones (I've seen them up to 12 ton, but not sure if that is enough).

I was planning on making a die to get the right bend. Would something like 303 stainless be acceptable as the die for bending 6061? Or do I need something harder? Would 7075 AL work?

Finally, when I see presses, they often just have a table that squeezes down on a lower table, or a table with a column that comes down. Are there special types of presses designed for this job or do you just drill/tap holes in the actual press to mount your die?

I have a CNC mill so I can cut whatever die I need, although due to the curve I'd rather do it out of something easier to machine than 316.

Thanks for any info!

[HuFlungDung]

Aluminum needs to be handled with kid gloves or will scratch up. This is because the material wants to drag over the top of the female die as the bend is being made.

Might I suggest UHMW for your die block? It is a breeze to machine, and is tough enough for something like what you are doing. Just make it a bit overwidth, so that you can radius the top corners of the female half, as they do the brunt of the work and will deform after repeated use if left sharp and pointy.

Or, you could make a metal die set out of cheap steel, and overlay the surface with some 1/8" UHMW pads to protect your part. Suitable allowance for the altered radius can be built in to your machining plan.

There will be some springback of the material. I do not know how to predict that, other than to take a wild guess after an experiment or two with a crude die involving welding a couple of round parallel bars to a backplate and pressing a sample part in between to a depth stop.

The pressure required to make the material take a new set is fairly high. While a 12 ton would do the rough bend, the material nearest the edges of the part, closest to the supporting edges of the female die, takes lots of pressure to get that last little bit of roundness to 'take'. This high pressure requirement can be reduced if you do the center bend, then rotate the part to reapply the pressure near the edges, with the edge more centered in the die. I think it could be done with a 20 ton hand press, if you just have a few to do. You can never have a press too big when bending stuff

[DareBee]

Hu's points are all very good and I agree with all them.
You will have a very hard time getting the edges to form a radius, period.
The easiest and least expesive method would be to roll form the FB.
Get a smallish hand crank roller and do the rolling on lengths that will create about 3/4 of an entire circle the dia you want. The last 4" or so at each end will not be curved. Cut your segments out of the aluminum ring and machine them afterwards.
You can put some duct tape on the forming rolls to keep the marking on the FB to a minimum and replace the tape as it wears or gets crap embedded in it.
Whether pressing or rolling you will likely be surprised at how much you need to overbend the material to get your desired radius.
6061 is not a good material for forming applications but your radius is easily within it's capabilities.

[Geof]

You are making two pieces that must fit together after bending. This might be difficult if they are bent separately. You could consider a combination of what has been suggested so far: make your two pieces, fit them together sandwiched between 1/4" (or thereabout) sheets of UHMW and then pass them through a roll former. If you put a spacer piece on either side you might be able to get complete curvature on your part with the not curved ends Darebee refers to being the spacers. It will depend on the thickness of the UHMW. One advantage with roll forming is that you can vary the bend radius whereas with dies and a press you have to make new dies.

[SRT Mike]

Thanks for the info so far. I'm having a hard time figuring out the best way to make this. See the attached picture - I am trying to achieve something somewhat like this, although with a little thicker material. I want to create a bottom plate like that with tapped holes, then a top plate that will sit on top and be screwed into the bottom plate, holding the license plate between.

The accuracy of the curve would be somewhat important, because it is going to have to match a factory fender, and while the tolerances would be pretty lax since there would be a gasket, it would still need to be pretty close.

I can definitely envision the problems of using a press with a die to form this and the edges not 'taking the curve' correctly. How would the piece shown be made?

The thing I like about the press idea is that it seems more consistent once I get it worked out. I make the die, then I press the metal and that's that. If I roll form, it would need adjustments from time to time to remain accurate, I'd think.

It seems that if I could determine the amount of spring-back (probably through trial and error) I could make a die out of UHWM as Hu suggested and just punch 'em out all day long without worrying about adjusting anything. It's just that getting it right the first time would be the trick

Also, if I rolled it wouldn't it be easier to do it as a long channel and then slice it into segments - and if so, then I'd lose my nice beveled edge I put on it (and rolling the cut-out piece would be harder). I have a 3-axis VMC so milling after forming would be a real PITA.

[lerman]

I would think that the "best way" to make this would be to die cast it. But unless you need thousands of them, it probably wouldn't be cost effective.

Ken

[Geof]

After seeing the shape I take back the rolling suggestion; I don't think it will work on something not symmetric.

There is a possible method to avoid spring back. There is a heat treating procedure that 6061 can be put through which finishes with the alloy at 0 temper at a temperature of 0 Fahrenheit, -18 Celsius. Provided it is kept at this temperature it stays at 0 temper but once it has warmed up to room temperature it goes back to something like T6 in a few hours. You need T6 for nice machining so possibly you could make the parts, heat treat them to T0, bend them and let them go back to T6. Maybe after that a polishing process.

I do not know the actual heat treating procedure; it was described to me by a local company when I was enquiring about get some 6061 tube bent. They do the process but would not tell me the actual temperatures and times. You should be able to get some information via Google.

[SRT Mike]

Originally Posted by lerman I would think that the "best way" to make this would be to die cast it. But unless you need thousands of them, it probably wouldn't be cost effective. Ken

I don't think that would work, even with high volume. The part would need to be polished and/or chromed after production, and my experience with cast parts and trying to polish/chrome them has been... uh... not great It seems you always get pits in the smooth surface and as soon as you sand through them you expose more. You're right - I also dont need thousands (at least, not at once). I'm guessing I'd need 1000-3000 a year, but no way would I want to do them all at once, I'd love to be able to do runs of 100 or so.

[SRT Mike]

Originally Posted by Geof After seeing the shape I take back the rolling suggestion; I don't think it will work on something not symmetric. There is a possible method to avoid spring back. There is a heat treating procedure that 6061 can be put through which finishes with the alloy at 0 temper at a temperature of 0 Fahrenheit, -18 Celsius. Provided it is kept at this temperature it stays at 0 temper but once it has warmed up to room temperature it goes back to something like T6 in a few hours. You need T6 for nice machining so possibly you could make the parts, heat treat them to T0, bend them and let them go back to T6. Maybe after that a polishing process. I do not know the actual heat treating procedure; it was described to me by a local company when I was enquiring about get some 6061 tube bent. They do the process but would not tell me the actual temperatures and times. You should be able to get some information via Google.

Thanks for the followup. I was familiar with AL going back to a higher temper at room temp but I didnt know you could special order it at T0. The problem with the above would be first I'm sure my metal supplier would charge a fair bit more for T0 (or anything other than run of the mill T6). Also, I have no real way to keep it at that temp around the shop. I was hoping to buy 1/4" plate and slice it up into 4"x6" sections and just press it into the shape I need, when I need it.

Do you think it would be impossible to make a die that overbends the right amount so it springs back to the shape I need, or do you think that it would be impossible to do that? I'd gladly undergo the cost/hassle of having to dicker around with various die shapes rather than worrying about special ordering AL in special tempers that must be kept at freezing temps before use.

Also, on the picture - thats not the actual part I want to make, just an example. If you look at the bottom plate (shown at the top of the picture), it would not have that tab part on it. It would be symmetrical along its centerline. The top plate (shown at the bottom of the above picture) would look pretty close to the way it does there except I intend to use a multitude of screws around the trim ring to provide a more 'industrial' look.

Thanks - still havent made any firm decisions, gathering all the info I can - its much appreciated.

[snakebit95]

Probably the only way you will maintain such a shape without spring-back, is to build a constraint fixture and anneal it while in the fully-contrained condition (fastened in the fixture). This will eliminate/reduce the stresses in the material and it will hold it's shape when you remove it from the fixture. I worked on a contract that required this exact operation for large radial-shaped aluminum panels that had a VERY TIGHT tolerance on the contour. Hope this helps!!

[SRT Mike]

Originally Posted by snakebit95 Probably the only way you will maintain such a shape without spring-back, is to build a constraint fixture and anneal it while in the fully-contrained condition (fastened in the fixture). This will eliminate/reduce the stresses in the material and it will hold it's shape when you remove it from the fixture. I worked on a contract that required this exact operation for large radial-shaped aluminum panels that had a VERY TIGHT tolerance on the contour. Hope this helps!!

Thanks for your input - regarding springback, does the metal continue to spring for a period after its removed from the mold? I figured it would spring back a given amount and stay that way - but will it continue to spring back for a few hours or days or something? If so, that would be a problem. I don't mind doing trial and error to determine the right amount of spring back to figure in so that it works out OK.

My tolerances are not all that demanding, but I wouldn't want to have to worry about it continuing to change shape for hours after it was bent.

Thanks!

[snakebit95]

For what you are doing......If you compensate for the spring-back at initial forming, I can not imagine it moving that much after you apply a "set" to the material. We form a lot of metal parts at my current job, and do no additional processing other than the forming operation. I think it mostly depends on what level of tolerance you are working to. From the looks of it, you have (4) FHCS to pull it into shape when you install it. That should be enough to address any slight deviation in contour between parts.