Tube bending 304 stainless steel

[Wicsteve]

Our company has been asked to make an exhaust manifold assembly from 38.1 mm diameter(1.5 inch), 1.2 mm wall thickness, 304 stainless steel tubing. One of the require(!) bends is 123.5 degree bend on a 47.63mm centerline bend radius.

Our tooling engineers tell us they cannot make the tubing bends without increasing the wall thickness and going to a larger bend radius. We're bending tubes using hydraulic, NC controlled benders, we are using ball style mandrels, and we have the correct size dies to make the 47.63 mm radius bend.

I would have taken their word for their concerns had I not known that this manifold has already been prototyped, as is, and our customer will not deviate from the original designed wall thickness and bend radius.

1) Can the bend be made using our present tooling, equipment or process?
2) Are there process techniques or equipment changes that would enable us to make the bends (even if that requires investment on our part)?
3) Suggestions or comments?

 

[rb1957]

if you've seen it, then it can be done, no?

if you heat it that'd help.

if you fill the tube (oil, sand?) that would help support the wall.

 

[Wicsteve]

The bend absolutely can be made but I don't know if special processes were used to make the prototypes. However, many thousands of these manifold assemblies need to be made each year so the bending process has to be somewhat production friendly.

 

[MikeHalloran]

How about asking the guy who made the prototypes?



Mike Halloran
Pembroke Pines, FL, USA

 

[TJOrlowski]

Are your tooling engineers employed by your company, or the manufacturer(s) of your bending equipment.

Have you spoken to any of your bending operators to get their opinion on the difficulty of making the bend?

-TJ Orlowski

 

[ornerynorsk]

Are you using a wiper die? Fully annealed material?

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.

 

[racookpe1978]

how much failure can you tolerate? Frankly, if you are making only a few, and those few are the survivors of a larger batch, you may be able to charge enough to account for the waste.

But a long production run or tight budget - and who doesn't have those? - eliminates that option.

I've seen all of the above recommendations work in other cases. This is an exhaust (low pressure) application, not a part of a superheated steam or hydraulic pipe: How much "wrinkling" can your client tolerate on the inside of the pipe before he rejects the tube as "ugly"?

 

[Wicsteve]

Replies

First, our associated factory is in mainland China so direct communication is difficult. Engineering and factory responses are done through interpreters. We're evaluating, at this time to accept or not accept the job. I believe the engineering opinion is from our factory's engineering staff not an outside consultant or the equipment manufacturer.

Second, we need to manufacture about 20,000 manifolds over a four month period of time. We can't afford to have a high reject rate. The production method needs to be reliable and efficient.

We're working through our customer to see if the company who prototyped the parts will talk with us. Normally, we are NOT in competition with them. We've been told they don't want to make the parts because of the additional manifold components such as powder metal flanges, flex tubing, and/or tolerance requirements that we can provide. There is no indication at this time that the tube bending was the hang-up. There wasn't any wrinkling on the prototype parts although we did not have the opportunity to closely examine their parts.

Filling the tube with a substance is not a practical solution for the production volumes we need to make.

While we are using a wiper die, I don't have any idea if the bending machines are completely set up properly. I have to rely on the fact that our factory has good experience in tube bending but has limited experience in bending stainless. We'd walk away from this project except for the money and don't want to give up so easily.

We've asked the factory to experiment with the setup. I was hoping that someone out there might have had enough stainless bending experience to be able to offer some general advice or to rank how difficult this bend might be to make (difficult, but can be made with care; need special techniques or equipment; can be made but reject levels are likely to be high, or you'll need a different grade of stainless or thickness; etc). ...and yes, we've already guessed that the bend is difficult.

 

[ornerynorsk]

Bending carbon vs stainless is like shaping hard wax vs silly putty. Stainless is not really difficult, it's just different.

A properly sized and designed mandrel and its placement is going to be fairly critical on any material, not just stainless, on a tight, high degree bend like this. I'm guessing it will take a multi-ball close pitch mandrel.

The raw material is going to be just as critical as the tooling. The state of anneal is very important to get the material to flow properly. 304 has such a high elongation that any hardness or unusual metallurgy is going to affect the result noticeably.

If you have a spare sample of your customers' bend that you can afford to destroy, section it through the middle of the bend and you will see how radical the difference is in wall thickness between the outside of the bend and the inside.

Good luck. As you stated, it is a difficult bend but it should be achievable on a production level without too much scrap.

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.

 

[racookpe1978]

20,000 parts. Hmmmn.

What about using a slightly thicker wall thickness so the shrunk- wall (on the outside of the bedn) remains thick enough to stay in spec? It would make the crinkling on the ID potentially worse ... but would the payoff work?

Try 6-8 bends locally using 304 stainless and carbon steel at a local custom muffler or pipe shop. Get local advice for 400.00 - 800.00 dollars to avoid losses on a 200,000.00 - 400,000.00 "we warned you it wouldn't work" failure.

 

[ornerynorsk]

Muffler shop would never be able to do that tight of bend. Not even a crush or press bend. Most exhaust is being bent at 3 to 4 X tube dia for a CLR.

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.

 

[berkshire]

Thinking out side the box,
Would stamping the bend in two halves and welding the seams with automatic welding be feasable?
B.E.

The good engineer does not need to memorize every formula; he just needs to know where he can find them when he needs them. Old professor

 

[Wicsteve]

Thanks
I'd like to thank everyone. I've forwarded your responses to our tooling engineers in China with additional discussion. So far, we're still 'negotiating' with them.