Material Selection for Splined Shaft - Maraging, 300M, 4340?

[JCorsico]

Hello!

I could use some advice on selecting a material for a splined shaft. This is a newly designed prototype.

The shaft is small (0.707" OD, 0.375" ID, and about 24" long). It connects an internal combustion engine (prime mover) to a hydraulic pump. At peak design load of the hydraulic pump (max flow at max RPM), I compute the stress in the shaft to be only 20.5 KSI.

However, the shaft is expected to see very significant shock loading, as the prime mover itself is subjected to very significant shock loading and resulting RPM fluctuations. This shock loading could lead to some load reversals on the shaft, but I don't expect these load reversals to be significant, since the pump has a fairly limited rotational mass.

When I take this shock loading into account, and use the spline formulas provided by Dudley (

https://www.eng-tips.com/viewthread.cfm?qid=511244),

I get the result that the shaft material should be able to handle at least 286KSI. This seems like overkill to me (maybe Dudley is very conservative?), but I am not an expert at fatigue or toughness engineering. Thus, I need help!

What materials are recommended? Please bear in mind that this is a prototype and we're only making one shaft at this time. 300M seems like an obvious choice, but it also seems like a hassle given the tricky heat treating requirements and likely post-heat treat machining required to correct for distortion. It may be cheaper and quicker to use a more expensive maraging steel that doesn't distort while being age hardened?

Thank you!!!
Jon

 

[EdStainless]

Don't use 300M. Trust me.
What is the spline style, involute?
We used 4135/4140 for these commonly.
When we needed more strength or fatigue life we went to 4335M.
Our favorite temper was 800-825F.
In 4335M this reliably gets you in the 220ksi range with great toughness and fatigue resistance.
We would have the bars Q&T, then we would machine length and spline, re-temper, and hone the spline.
There are other high strength alloys, but I have never had good luck with them in this type of service.

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

 

[JCorsico]

Thanks Ed.

It is an involute spline. It's a very outdated Air Force Navy Aeronautical standard, published in 1953 and then canceled in 1993. But it's what we've got to use because it's what the pump uses, and we can't change the pump. 20/30 pitch, 30 degree pressure angle, flat root, side fit.

Why the dislike of 300M?

When you have an application like this, with high shock loads that are well above steady state loads, should I be focused on absolute strength, toughness or fatigue? Or all three?

Thank you again!
Jon

 

[mfgenggear]

jcorsico

please take into account the cost and manufacturing, for a 24 in shaft that size it would be a pita, and there is significant dimensional change during heat treat. draw up a preliminary drawing send it out for quote , it may be no one will want to touch this part, no less one part.

 

[EdStainless]

We had serious fatigue (or impact) issues with 300M.
Your shaft is hollow?
too bad I am not still with a former employer or I could send you a few pieces of 0.75" OD X 0.340" ID in 4140 and 4335M.

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

 

[JCorsico]

Thanks again guys.

What can I do to make the part easier to manufacture? It has to span a 24" long gap. I can use a larger OD (1"?) if that makes manufacturing easier. But the splined ends will still have the smaller OD (the minor diameter of the tooth radius is 0.707"), and I can't change the splines.

The shaft is currently designed to be hollow for weight savings. But I can keep it solid for manufacturing, if that helps.

I suppose I could make a composite shaft, with a much larger center section (1.5" OD?) welded onto two short stub-shafts that are splined. The stub shafts should be easier to make, because they are so short. But the welding seems tricky, and I'd be concerned about keeping the splines aligned with the rest of the shaft.

 

[EdStainless]

A solid 1" shaft might make sense.
The center is really a non-issue. A hole doesn't actually remove much weight, and it makes very little contribution to the strength and stiffness.
I have friction welded stubs to the main shaft and them machined the entire assembly.
I don't think that stubs would help your case.

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

 

[Tmoose]

Internal combustion with pistons and cylinders ? The torque variation without an immense flywheel can be more than significant.

 

[JCorsico]

Correct, internal combustion with pistons and cylinders. I suspect this is why the Dudley spline analysis is saying I should select a material with 286ksi strength, even though the calculated steady state load is only 20.5ksi.

 

[mfgenggear]

Jcorsica
Sae 4340 modified is what i recomend.
The issue with shafts is preventing it from becoming a pretzel during heat treat.

 

[WKTaylor]

In aerospace for high shock loading and toughness and high strength... You might need something like forged...

MP35N or MP159

or perhaps Ferrium M54

... forged shaft and forged splines...

I suspect that if You could find a torque dampener for shock attenuation, that would allow substantially lower strength high toughness steels.

Regards, Wil Taylor
o Trust - But Verify!
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", HBA forum]
o Only fools and charlatans know everything and understand everything." -Anton Chekhov

 

[mfgenggear]

Hi Wil
With all due respect.
Those materials are a pita, are expensive, and have lead time upto 6 months or more.

 

[WKTaylor]

MFE... then to regain toughness, the conventional shaft-steel heat-treat HAS TO be reduced. A high quality grade of low alloy steel should work fine ['aircraft quality'] without going to exotics.

A few issues are bothering me.

Shaft between X engine Y pump has to be very rigidly aligned by mounts... or the male splines have to have lengthwise curvature to allow for miss-alignment or there has to be some mechanical device for torque thru misalignment... U-joint, stacked rubber discs, etc... such as use in tail-rotor drive shafts. This is especially important for long thin shafts running at high speeds/torque.

Soooooo why is the pump 'remote' from the engine, such that it requires a PTO-shaft? I am curious as to 'why' the pump isn't mounted [bolted] directly to the engine drive-shaft flange and engine case... possibly with an adapter... unless of course this is a 'country engineering project'.

for giggles...

MIL-HDBK-776 Shafts, Elastic Torsional Stress Analysis of

ALSO, there is a lot of design literature on agriculture PTO drive shafts/systems... doesn't seem like this should be such a big deal.

Regards, Wil Taylor
o Trust - But Verify!
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", HBA forum]
o Only fools and charlatans know everything and understand everything." -Anton Chekhov

 

[JCorsico]

WKT - the pump will not fit in any other location. Believe me, if we could have put the pump next to the engine and powered it via gearbox bolted to the engine, we would have done so. It would have avoided this shaft and also been more simple overall. But it just wouldn't fit.

Also, if we could have used a larger spline (and thus a larger shaft), we would have done that too. But we can't. The pump is the pump, and a newly designed pump wasn't in the cards. We had to use what Eaton/Vickers already designed and already had available off the shelf. So we're stuck with this small spline that has a 0.707" root radius.

Regarding spline alignment - you are correct that the design entails two machined plates, bolted to rigid substructure (which is mostly the engine block), to ensure alignment between the two ends of the shaft. Ball bearings are installed in the machined plates to guide the ends of the shaft into the hydraulic pump (on one end) and a pulley (on the other end). That said, your suggestion of possibly adding a flexible coupling is a good one, and we'll look into it.

Thank you again!
Jon

 

[TugboatEng]

Have you considered that such a long, slender shaft may act as enough or a spring to reduce the shock loading some? These are sometimes referred to as quill shafts.

 

[EdStainless]

What people have suggested is to reduce the shock loading on the spline by turning the center of the shaft smaller than the spline root.
Allow a longer smooth section to take the shock and spare the teeth of the spline.
This is rather common in impact tools.

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