Visual Inspection of Bent Sheet Metal? 2

[aetherTap]

Hi All,

We are dealing a vendor that is insisting on using VERY tight bend radii on aluminum sheet metal pieces and we basically have no choice but to visually inspect their parts and be done with it. It is far from an ideal situation, but I've been asked to find a standard that addresses visual inspection of Aluminum alloy bent sheet metal.

I see standards from NASA and AIAA discussing active Nondestructive evaluations, but nothing about purely passive visual inspection. Is that because this is frowned upon without the use of active ND testing?

I'm sure there are specs I'm just missing or maybe you all have some helpful trade knowledge about changing surface roughness, elongated grains, discoloration?... Thanks!

 

[ornerynorsk]

"a vendor that is insisting"

It can be a very slippery slope once you capitulate your apparent or perceived authority as The Customer.

If your vendor is insisting, perhaps it is they that should inspect and cull out the bad ones per your requirements, although "per your requirements" has, at this point, lost its flavor.

It is better to have enough ideas for some of them to be wrong, than to be always right by having no ideas at all.

 

[aetherTap]

I'm a new addition to this project, so I have limited understanding of the politics involved. Fact of the matter is that I've been tasked to find acceptable visual inspection criteria for these 'suspect' bends.

The closest thing I've found (but have not purchased) is ASM Handbook 17. Anyone have experience with this?

 

[MikeHalloran]

Visual inspection is expensive, and its subjective nature catalyzes arguments that get more expensive as they drag on.

If you are concerned about the bend radius, add a minimum radius to your drawing or your contract, to which you can inspect without ambiguity.



Mike Halloran
Pembroke Pines, FL, USA

 

[ornerynorsk]

Understood, I've been in positions where company politics/nepotism/handshake-agreement-of-3-decades-ago has forced less than desirable components and processes upon us. The question usually arises "who eats the scrap"? Using an unvalidated process that is prone to failure is not inexpensive, and is very frustrating to deal with. It sounds like you may have already explored all rational avenues to change the situation, though.

Do you have some typical failure samples from which to build an inspection protocol? What equipment do you have at your disposal? What kind of production volume are you dealing with? Would a cheaper/better solution be to just buy the correct tooling for them?

Just a couple of thoughts. Good luck.

It is better to have enough ideas for some of them to be wrong, than to be always right by having no ideas at all.

 

[aetherTap]

All valid points... An adequate bend radius (brought upon by new tooling) would require design changes to components around the bent pieces which are not going to be considered at this point in time. After this production run, a new vendor will be found, designs will be changes, etc.

Right now, it's unfortunately been decided that 'we will let them implement their own manufacturing techniques and visually inspect them. they will eat the cost of failed inspections".

Thus, my problem... I don't see any visual inspection standards for this ANYWHERE. ugh....

 

[KENAT]

Maybe I'm missing something.

What drawing (or otherwise contractually stated) requirement is it that you want to visually inspect?

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[aetherTap]

Kenat, vendor was given freedom to use bend radii they deemed adequate. Parts were reported back as having cracks along sheet metal bends, some cases even cracked BEFORE being welded onto a larger sheet metal structure! Company is trying to save face, but has its hands tied in terms of 'true' remedies. Only immediate solution is to make sure that these parts are inspected for visible cracks. The engineers here would love to have a more complete inspection process (without needing much training or added tools), ie maybe there is discoloration or surface roughness in the bend that indicates (although no visible cracking) that these parts are too weak, etc.

I have walked into a nightmare. Trust me, I know how ridiculous this all is.

 

[KENAT]

OK, so I assume you've ruled out something like die pen inspection?

One type of anodize, I want to say chromic, will accentuate certain types of defects but I forget the details and it doesn't sound like an option for you as you are welding after forming.

However about requiring the visual inspection be conducted under a certain magnification or similar, while still quite subjective I've seen this kind of thing done before e.g.

INTERNAL SURFACE OF ALL BEND RADII SHALL BE VISUALLY INSPECTED UNDER 2X MAGNIFICATION IN WELL LIT AREA. ANY VISIBLE CRACKS SHALL BE CAUSE FOR REJECTION.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[MikeHalloran]

How do you know there aren't undetected cracks in what you've used, and shipped, already?
You can't. You have to assume they're all cracked.
Does your end product function OK with completely cracked bends?

If not, you shouldn't ship squat until you find a new vendor, fix your documentation, and fire the MBAs.





Mike Halloran
Pembroke Pines, FL, USA

 

[aetherTap]

They do function with cracked bends actually, they are not structurally critical parts. This is the main reason that the 'true' fix is being pushed farther down schedule.

And yes I agree that we have to assume they are all cracked because there is no control in place (vendor or company) that determined when the cracks were formed. Some were before welding, some may have been after shipment, some could've been in the field. Who knows!

 

[KENAT]

Do your drawings even say that cracks are unacceptable?

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[dvd]

Is the production run finished? Can you apply some thought to what would help to prevent cracking without changing tooling or process? Do you stipulate the rolling direction of material perpendicular to the bend line? (Doesn't help much if there are bends on orthogonal sides.) Is the receiving inspection of the aluminum sheet showing any problems, i.e. is the material to spec? If bends are cracking, why is the material condition specified as it is - does it need to be T6 (or whatever)? Do you have a blank pattern drawing defintion that is not allowing the formed part to be produced to its drawing definition without the tight bends? Seems like inspecting the quality in at this stage is not a win-win situation. Either work with the vendor to solve it, or scrap the lot. If the parts are subject to vibration you may end up with lots of fatigue failures in the field.

 

[ajack1]

I might be missing something here, but if other parts would need modifying if the bend radius was increased, then isn’t it the design that is driving this, not the vendor?

Other than possible heat treatments like annealing or changing the grain direction on the part isn’t the real problem the design requires a radius that is not achievable on a regular basis in this material?

 

[aetherTap]

Yes, it is my opinion that the design is faulty. Companies should not, in theory, get to this point where their hands are tied. Unfortunately what I gather from the responses here is that no one is familiar with a visual inspection scheme. I'm going to make my own and look forward to the time when I can clean this project up.

From vendor: "we orientated the grain in such a way that it allowed us to fit as many pieces into one sheet".... haha

 

[MadMango]

I've never seen a published "standard" to cover this type of visual inspection, if that is what you are looking for. I've seen plenty of home grown drawing requirements though:
1. Under 10x magnification, no more than 25% of any bend length shall be cracked.
2. From a distance of 18" no cracks shall be visible prior to paint.
3. No cracks permissible thru material.

"Art without engineering is dreaming; Engineering without art is calculating."

Have you read faq731-376 to make the best use of these Forums?

 

[ajack1]

Again I might be missing something but if you add a new level of inspection or change the criteria of how the part is measured, would they not just put in a new price?

I realise it is a very grey area we are talking about here as minor cracks may not be acceptable and it could be argued that is a given but as there will always be some kind of cracking or at least thinning on any bend in sheet metal, it is a grey area.

If you changed the tolerance on a hole that then required a different process it would not be unreasonable to expect to pay more for the part. I don’t see this as much different.

Why not talk to the vendor and ask what would the cost implications be to solve this problem, it could be you only get 99 parts from a sheet not 100 with a very small cost increase to change the grain direction for example. There could be a very cheap and simple solution out there if both parties look for it.

If both parties just dig their heels in and think right I will show them who is boss, you will probably arrive at the situation where they think right they will now reject 50% of the parts so we will double the cost.

 

[aetherTap]

Thank you MadMango! I like that, specifying the distance. I will probably touch on the work lighting as well. Does it make sense to have a "touch" inspection, or is that pretty much worthless?


KENAT - no, they didn't mention this type of quality control, hence the reason the company is in such a mess. I only hope it's a lesson learned for them.
ajack1 - no, you're not missing something. we have had phone conferences, which has driven the company to this temporary "solution". sometimes reasonable solutions DON'T get arrived at, at which time one of the two companies normally makes some type of corrective action in regard to the other company.

 

[KENAT]

"Does it make sense to have a "touch" inspection"

Could be a recipe for lots more entry's in the accident log and higher use of band aids from the company first aid kit depending what you mean and how big the cracks are.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[MikeHalloran]

The normal way to control bend cracking includes:
- specify a minimum inside radius
- specify a grain direction
- use a softer temper of material
- or combinations of the above,
... and that's it.

Sorting for cracks by visual inspection or by touch is not a common or accepted technique, which is why you can't find a standard for doing it.


Judging by the little you have revealed here, I can only infer that whatever drawing exists is utter crap. You might wish to start fixing that, just in case your company manages to survive this particular clusterf*ck.




Mike Halloran
Pembroke Pines, FL, USA