Fracture Free Forming

[Greg Mercurio]

Good morning: I have a problem that requires expertise not available locally. We are experiencing moderate to severe fracturing of 0.250 inch 5052-H32 plate during forming. We are using a CNC press brake with a .375R forming die. We have successfully formed this plate hundreds of times in the past. In discussing with our plate supplier we are told that they cannot guarantee fracture free forming because there are no N values in the ASTM B209-14 Table 2. That seems disingenuous and a free way to absolve themselves of liability. Can someone familiar with that particular Specification explain what the absence of a recommended Bend Diameter Factor N really means? Does it truly mean that these materials cannot be formed?

Thanks in advance.

Greg M

 

[metengr]

Your best position in a situation like this is to have a 3rd party (independent), metallurgical lab perform a proper failure analysis to determine the cause for forming cracks. I can think of several factors so you need to have a proper analysis to eliminate certain possibilities and provide a position for discussion with your supplier.

 

[Greg Mercurio]

Appreciated. We will try to do that but here in the wilds of NW Montana, met-labs are few and far between.

 

[EdStainless]

The initial investigation is not that difficult, and I am sure that FedEx can deliver the samples. I would start by looking at tensile properties, grain size, preferred grain orientation, and general microtructural cleanliness.
Either they changed a process in order to save money so they don't want to go back, or something changed that is not under good control and they can't go back.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[Greg Mercurio]

We are going to see if we can get the Montana State University Materials Department interested in a cooperative project. Two of our engineers are grads, still have contacts, so we might be able to help the students with some real world situations, while we get some definitive answers. Win-Win.

Would still like some guidance on the B209-14 spec though.

 

[IRstuff]

One should be careful to do a proper root-cause analysis. While the likelihood is high that your supplier has somehow gummed up their material, there's also the possibility that someone in your shop "improved" the forming process to the breaking point.

I was once in a company where someone "improved" the life of some furnace tubes by turning down the temperature and increasing the processing time to compensate, but that someone told no one about it, and it went undetected for about 3 years, since that part of the process was only affecting a portion of our typical product, which already had lousy yield. We then got a product where most of product required that process and we got abysmal yields. Big panic ensued, as the product was for a corporate sister division, Paymentec (the credit card chip guys). Months of tiger-teams followed, but it was an obscure article in an IBM Systems Journal from more than 15 yrs prior that had the data showing that a mere 50C drop from 1100C could totally and completely discombobulate the process.

In any case, you need to do your due diligence:
> try to find old material that can be formed to make sure your internal process hasn't gone funky
> try to find out if someone decided to make the process "better" without telling anyone, or worse yet, it improves the process for other products, but not yours.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529

 

[mrfailure]

I would still recommend you sending out to a metallurgical lab that does industrial failure analysis routinely as opposed to going to a university - you will have their real-world experience from similar customers. I acknowledge my own bias as a failure analyst working at such a lab, but as noted you do not need proximity to the lab and modern communications will allow direct input and feedback throughout the investigation.

 

[Greg Mercurio]

IRstuff and mrfailure: For reference, we are a small fabrication facility, and tooling for the press brake is limited, as are the opportunities for "process improvement". Having 40+ years of manufacturing behind me including 10+ building aerospace structures for solid rockets give me a fairly broad understanding of your "improvement" statement.

We have made arrangements for Montana State University Materials Science faculty to perform the investigation for us. As much as I'd love to be able to use an independent lab for this, we don't have the budget so we'll take the free help.

I left this part out of the original post, but the plate's origin is China. Given my long experience with Chinese manufacturers and their quality groups, I suspect that the material certs supplied with the plate are worth just about the cost of the paper. We have a sample of US plate and we're waiting for the brake operator to return from vacation. I did say we're small.

I'm still looking for an interpretation of B209-14. Anyone?

 

[metengr]

Greg Mercurio
I have the latest copy of ASME SB 209 material specification. The section on bend testing is 12.0

Sheet and plate shall be capable of being bent
cold through an angle of 180° around a pin having a diameter
equal to N times the thickness of the sheet or plate without
cracking, the value of N being as prescribed in Table 2 and
Table 3 for the different alloys, tempers, and thicknesses. The
test need not be conducted unless specified on the purchase
order.

The ... for missing N values in Table 2 means the value is not specified. It does not mean it prohibits bending. You will need to determine the pin diameter based on trial and error or agreement between the supplier and Purchaser.

 

[btrueblood]

Greg, do you need the few ksi of tensile yield strength that the H32 buys you over H0? If not, you could stress-relieve the plate fairly easily (provided you have an oven big enough, and your wife isn't home to see you use it ) - 1 hour at 220C is the typical treatment for that thickness.

FWIW, the standard data sheet from Alcoa for 5052 shows a min. bend radius of 1.5t for 1/4" thicknesses, i.e. you should be able to make a crack free 90 deg. bend.

 

[Greg Mercurio]

metengr and btrueblood: The parts are not structural to the extent that the H32 is required. We could easily accommodate the H0 temper. Thanks for the tip. As to the pin dia and missing data, we're being told by the supplier that the lack of N data means it can't be formed. I laughed out loud at that.

We are using a 1.5T die, no problems for more than a year then boom. Obviously we did something magical for a year and the heat lot had nothing to do with our magic disappearing, as all magic does eventually...

Thanks all! It'll be interesting to see what the materials folks find.

 

[WKTaylor]

Greg Mercurio...

You never mentioned the total bend angle You are forming to.

An [aerospace] formability chart that I have for various aluminum alloys indicates that the min bend radius [BR] for 5052-H34 0.190"-thick sheet [no data for -H32] is BR0.50". This implies that a much larger BR for -H34 0.250-thick plate is required. Extrapolating 0.25 thick leads me to believe a safe BR for this material should probably NLT BR0.66"... far from BR0.375 [BR0.38] You have been using.

On the-other-hand, the chart indicates that min BR for 5052-O sheet 0.190" thick is only BR0.19"... hence would extrapolates to a min BR0.25 [+] for 0.250-thick -O stock plate.

Quite frankly I simply cannot conceive of how You have been getting away with such a tight BR for -H32 plate.

NOTE. Three concepts to be very aware-of when bending thick stock to extreme tight BR are:

(a) ALWAYS bend ACROSS the sheet/plate grain [rolling direction] whenever possible... NEVER bend along the grain. Maximum material ductility is almost always in the 'L' direction... not the 'LT' [across the plate] orientation. Yes, I stated that correctly...

(b) Ensure that the plate stock is well supported during the bend process. One side must be rigidly/flat-clamped; and the other side must be free to move/slide as the 'break-tool' rotates that free edge in a sweeping motion. Applying light lubrication to the under-side of the [unclamped] plate edge being bent [sweeping/sliding] MAY help it slide smoothly [minimize friction] during the sweeping motion around the BR and against the break face. A light lube on the side formed-around the brake radius might also help.

(c) 'Hot-bend-forming' is also a way to further reduce BR and make a successful bend. I'm unsure what temperature/time -H32/0.250"-thick would require to substantially tighten the BR. this is tricky and could involve high temp SFLs, etc.

NOTE.
IF you see 'Luders-lines' [Luders-Bands, etc] form along [parallel to] the outer [tensile strain] side of the bend, You can guarantee You're right at the formability limit for that particular raw stock.

https://en.wikipedia.org/wiki/Lüders_band

https://www.google.com/search?q=lud...X&ved=0ahUKEwjwoKCs_-HPAhUm6oMKHYhaDEAQsAQIQg

Regards, Wil Taylor

o Trust - But Verify!
o We believe to be true what we prefer to be true. [Unknown]
o For those who believe, no proof is required; for those who cannot believe, no proof is possible. [variation,Stuart Chase]
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion", Homebuiltairplanes.com forum]

 

[EdStainless]

The "L" vs "T" could be the issue here. Turn a piece 90deg and try bending it.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[Greg Mercurio]

All: The bends are all 90 degrees, we have successfully bent both with and across the grain as our nesting program for the router ignores grain for maximum part density. We are bending some domestic plate today in both directions for comparison. The radius I have stated thus far in this post is the nose die, the lower die has a large throat, apprx 1" by 1" deep and the plate is unrestrained and allowed to take whatever finished radius it pleases. Most outside radii measure approx. 0.5" - 0.6"