Ball-passing test on bended pipes 2

[bernardirafael]

Dear all,

We are currently manufacturing an equipment for a client that is composed of many bended 48 mm pipes. The client requires a "ball-passing test" on all curves, which consists in passing a tolerated sphere inside the pipe: if it goes through, the bend is approved, otherwise it's not.

However, I can't find any standard for this test. The client also doesn't knows what standard contains the test.

Does anyone knows it?

Thank you in advance.

 

[MintJulep]

I vaguely remember that the American Association of Railroads has dinner like this for train air puppies.

 

[LittleInch]

ISO 15590 applies to hot and cold bend fabrication for pipeline.

What bend radius are you talking about here?
Hot formed or cold formed?
Internal former?

It's a common requirement in bend fabrication specs.



Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[SWComposites]

why is a test standard needed? either the ball goes thru (passes) or gets stuck (fails).

 

[bernardirafael]

@SWComposites

The standard would indicate the ball diameter for the given pipe, I think.

I could just pick any diameter and work with it upon agreement with the client... but that is not ideal in my opinion as I could choose a diameter which I'm 100% sure will approve all bends, or the client would insist on a diameter which wouldn't approve the bended pipes.

I'm feeling that on the contrary of what I've been told, there's no standard for such test.

 

[rb1957]

no, ball diameter is specified by design ... no?

but IMHO expecting a pipe to be undistorted around a bend could be "challenging". Obviously what they want to avoid is significant change in section, but that shouldn't rule out minor ellipsing of the round tube.

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[bernardirafael]

Yeah a minor ellipsing is okay. The ball diameter should be smaller to pass with an "acceptable" distortion.

I usually consider a maximum distortion of 12% (max diameter vs min diameter in a bend). I could make a sphere that passes through a 12% and below distorted bend. Maybe this could satisfy the client... but as it follows no standard, I think it's kinda pointless and I should straight consider < 12% distorted bends as approved.

 

[LittleInch]

A common feature of pipelines with bends is the requirement for a plate mounted on a pig to pass through based on 95 (or sometimes 97.5 %) of the minimum internal diameter when you apply all the tolerances for wall thickness (+ve) and ovality.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[jmec87]

Why does your customer want the ball-passing test?

Will they be sending balls through this pipe? Do they want to minimize flow disruptions? Did somebody just think that they should limit distortion but without much reasoning?

The reason why they want the ball-passing test should guide the agreed-upon ball diameter for the test (or dropping the test requirement if there isn't really a good reason for it).

 

[EdStainless]

In tubing ubends we use ASTM A688. You can take the tube tolerances (min diam and max wall) along with the maximum allowed ovality (measured on the OD) and come up with a ball size which should pass.
This test can be very dangerous, only use low pressure (10psi) air to push the ball through.
I saw someone accidently use 125psi shop air and he shot a 0.485" carbide ball through a cement block wall.
This test usually would only be applied to the few tightest bend radius', once they get larger it isn't an issue.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[chicopee]

My response is based on my past experience in tube bending which could be of value but may not rightly answer your OP. In the 1970's as a mechanical designer, I was to research and buy for my employer equipment to be used in the fabrication of circular polarized FM antennae. A tube bender with an appropriately sized mandrel and die to form 1-5/8 inch straight tubes into circular shapes without any deformation in the roundness of the tubes was commercially available and bought. The only deformation was in the stretching of the tube walls on the outside of the bends while the roundness was still maintained. The MANDREL was the key to keeping the roundness of the tubes. The bends were of the order of somewhere around 8 inches in the internal diameters of the circular pattern. There was another method to bend the tubes in question but not selected as it required the tubes to be filled with material such as sand, seal the ends prior to bending.

 

[JStephen]

One approach would be to form a section of pipe, see if everyone is in agreement that it is "good", and if, so then see what size ball will pass through.
Measuring the OD in both directions might be as good as the ball test.

I think I've seen videos of them hot-forming 90 degree wrought ells, and one of the steps was forcing one or more balls through it. That wasn't a tolerance thing, but part of the forming.

 

[EdStainless]

When you make tube bends (esp tighter ones) you have to balance the ovality in the bend with wall thinning.
The ubend tube specs have limits on both, and it is a trade-off because you need to have a bit of each.
Om tight bends we always used UT to verify the wall thinning as well as measuring the OD ovality.
The ball pass is sort of bogus anyway.
In a large tube bundle where there are many different bend radii the clearance is the most concern in the tighter bends. But they pose more of a flow restriction anyway and will see less flow naturally.
It isn't a bad practice, just usually not needed.
Making sure that the tubes meet the dimensional specification and measuring the outside of the bends tells you the same information.



= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[chicopee]

JStephen, you still need a mandrel to keep the roundness of the tube bends. Note the electrical metal conduit bends performed by electricians, a round ball would not pass thru the bends of theses electrical conduits.

 

[mfgenggear]

it's my humble opinion, to do a such a test is going to be a PITA. and many lose or scraped parts.
the tubing will probably needs to be in sections, reason, to make perfect bend or nearly perfect, there needs to be rubber or similar in the tubing during the bend.
each end does need to be supported. like I said PITA, the size of the ball has to be agreed with the factory, the customer, and engineering.
then the sections needs to be assembled and welded, in a fixture.

 

[JStephen]

FYI, here's a couple of videos using balls to presumably maintain roundness:

https://youtube.com/shorts/a5KQvMrlnYg?feature=share

https://youtube.com/shorts/wx-qpRvyt2w?feature=share

 

[rb1957]

"PITA" ... he, he ... had to look it up !

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[DBreyer]

EN 12952 (European standard for boilers) has guidance for the ball-test including permissible ball sizes