Spring Return Mechanism

[Cockroach]

I have a Clapper which is held by a 5/32 inch axial to a Seat, a check valve arrangement. The Clapper weights 4 1/2 lbf, the centre of mass is 3 1/16 inches from the centre line of the axial. I need the Clapper to be spring return, 95 degree rotation counter clockwise.

Using a torsion spring, I would need a moment of 14 inch pounds at 5 degrees of rotation. My intended FOS is 1.25, so I would like about 17.0 in lbf at 5 degrees. This means the torsion spring would need to be 1/8 wire, have an OD around 1.0 inch and have a minimum axial space of 1.438 inches. This will not work for me!

I need something that will provide me with about 20 in lbf return at 5 degrees and have an additional travel of 90 degrees. I only have 1.0 inches of axial space, inner core is 5/32 inch slip fit, maximum OD is under 1/2 inch. Clearly the torsion spring idea is dead.

What else can I do? Torsion Bar? Watch Spring? I'm right out of ideas at the moment!

Kenneth J Hueston, PEng
Principal
Sturni-Hueston Engineering Inc
Edmonton, Alberta Canada

 

[israelkk]

If it gives 20 in lbf return at 5 degrees and have an additional travel of 90 degrees what is the desired torque at the extra 90 degrees travel?

 

[MadMango]

Have you considered a polyester-based elastomer, like a urethane bumper acting like a traditional spring?

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


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

 

[TheTick]

You don't mention anything about preload on the spring. Can you go with higher preload but lower spring rate?

 

[MikeHalloran]

If it's in a lubricating medium, you might get away with a toggle or a barrel cam, and a linear spring. ... but I don't like the idea much.



Mike Halloran
Pembroke Pines, FL, USA

 

[Cockroach]

The torque at 90 degrees can be anything less than that responsible for material yield. I would like to stay around a FOS = 1.25, IsraelKk.

The preload would be just that Tick, five degree deflection holds the weight of the Clapper given the moment arm displacement, axial to CM of 3 1/16 inches. I just need the Clapper to remain firm in the seat, then swing open 90 degrees and have good spring return within elastic material limits. Cycle load lifetime, 50,000 or so, nominal temperatures, 100F or so.

There may or may not be a lubricating medium. In an oil well, yes the liquid acts as a lubricating medium. But if the well is a natural gas formation, then she's dry.

I haven't considered a bumper or otherwise, played a touch with a Torsion Bar application but discounted it early since angle of twist is 95 degrees and the shaft is only 5/32 inches in diameter. Not much meat there to make it happen.

Kenneth J Hueston, PEng
Principal
Sturni-Hueston Engineering Inc
Edmonton, Alberta Canada

 

[gerhardl]

Sketch? (For us foreigners to get the details properly ;-) )

 

[Cockroach]

Thanks guys. Attached is the model of my Check Valve application, shown from the vertical top position, so looking downwards.

The representative Torsion Spring, which doesn't work, is clearly shown. The axial is of diameter 5/32 inches, body length of the Torsion Spring is 1.0 inches. The left hand arm of the spring is 0.88 inches and is anchored into the body of the Seat. The second arm runs vertically down along the backside of the Clapper for 4 3/4 inches. As you can see, the maximum allowable OD of the Torsion Spring would be no greater than 5/8 inches.

In the relaxed position, clock position 09:35 HRS, the Torsion Spring must deflect 5 degrees CCW. This is the position given in the model attachment, clock position 09:30 HRS. Clearly the Torsion Spring must hold the weight of the Clapper, repeating, 4 1/2 lbf with centre of mass location 3 1/16 inches from the centre line of the axial. I need the Clapper to open an additional 90 degrees CCW, so that clock position would be 09:15 HRS.

I've tried 3/32 inch and 1/8 inch round wire diameter to no avail. Perhaps square wire would be better, although I'm not sure of manufacture availability of rectangular spring steel. I need 50,000 cycles for an average lifetime, thermal application no greater than 100F. The environment may be raw bitumen or natural gas. Both would be considered relatively clean and free of particulate matter.

I must admit, this problem has got me stumped. Is there a Torsion Spring suited to this application or must I consider an alternate system for Clapper return to the closed position?

Kenneth J Hueston, PEng
Principal
Sturni-Hueston Engineering Inc
Edmonton, Alberta Canada

 

[Cockroach]

Suppose I do something like this, attached file for a Torsion Spring, fully machinable from a steel sleeve. How can I compute the torsion spring rate in (in lbf/deg)? Do the closed form equations for Torsion Springs also hold for this device?

If I treat it like a torsion bar, then the angle of twist equals the torque times length divided by the product of polar moment of inertia and torsional rigidity. But what of stress?

Does anyone have a reference source for this type of Torsion Spring? Cross section is more rectangular in nature, I can approximate that but stress at the free ends theoretically increases dramatically, does it not?

Kenneth J Hueston, PEng
Principal
Sturni-Hueston Engineering Inc
Edmonton, Alberta Canada

 

[hydtools]

How about a constant force spring or one like that in a steel tape measure?

http://www.ametekhunterspring.com/products/constant.cfm

Ted

 

[MikeHalloran]

I'm not convinced that you can treat the entire spring like a torsion bar.

Compression/extension springs are stressed in torsion.

Torsion springs are stressed in bending.

The tangential tangs of a classic torsion spring don't produce any reaction torque until the end loop contacts the axle. Then the force applied to the tang and the reaction force at the axle constitute a couple, and the coil sort of approximates a long beam with a moment applied at each end, until some intermediate portion of the coil contacts the axle. Then things get messy.



Mike Halloran
Pembroke Pines, FL, USA

 

[zekeman]

Making the torsion spring rectangular will not work either. Since you have 17 in-lb at 5 deg you need about 300 in-lb at 90 deg, you would need a monster torsion spring with a wire diameter of at least 1/4 inch; rectangular cross section doesn't buy much.

You need a nonlinear spring, maybe like a cam against a linear spring or a springy post like you see on the carburetor choke shaft, without the detents.

 

[MikeHalloran]

Where did the 5 degrees come from?
I.e., if you preloaded it such that the flapper was seated at 90 degrees, or 95 degrees, you might have a more practical spring design. ... at the expense of more difficulty in assembly, winding up the spring to get it into the assembly.



Mike Halloran
Pembroke Pines, FL, USA

 

[Cockroach]

The 5 degrees came from the necessity to hold the Clapper tight against the Seat in the closed position. Usually Torsion Springs have a two to four percent full range preload. Since CLOSE/OPEN is ninety degrees, we're talking 3.6 degrees, so I rounded up to 5 degrees. This guarantees I hold the weight of the Clapper tight to the Seat as I run the system downhole.

Kenneth J Hueston, PEng
Principal
Sturni-Hueston Engineering Inc
Edmonton, Alberta Canada

 

[israelkk]

I am afraid the space for the torsion spring is too small as I hinted in my previous post and as zekeman specifically said it. I do not see another simple option instead of re-design the whole system and make the needed space for the spring.

To my best experience this is the most frequent mistake by designers and developers. They dismiss the design of a spring (bolts and other "so called" simple parts) as an important issue and leave it to the very end of the design, while it has to be designed and analysed even before modeling and/or drawing starts. The spring has to be designed and analysed in the very first steps of the design process where a rough sketch, forces and masses of the system are roughly known. Otherwise, you end up with a nice 3D modeled assembly or even production drawing of all the parts just to find that no space was left for a spring to do the job.

 

[Cockroach]

Exactly my fear, IsrealKK, the spring ended up being the design nemesis. I'm thinking as you are that the expectations for this spring will make it too large for the room provided.

Probably a spring and cam arm arrangement will be used, or at least that's the way I'm thinking as of late. Thanks everyone for the input!

Kenneth J Hueston, PEng
Principal
Sturni-Hueston Engineering Inc
Edmonton, Alberta Canada