shop-talk
[Top] [All Lists]

RE: Engineering Question Regarding Splined Shafts

To: "shop-talk" <shop-talk@autox.team.net>
Subject: RE: Engineering Question Regarding Splined Shafts
From: "Gordie's Garage" <mg_garage@ameritech.net>
Date: Sat, 11 Oct 2003 19:15:45 -0400
Okay, gotta ask a question...doesnt' the amount of torque applied to the
splined shaft (either wheel or trans) make a difference?  Seems to me that
the larger splines would handle larger torque loads better than the smaller
splines.  An uneducated guess says that the tranny shaft would have to
handle more torque than the wheels.
Just a thought.

Safety Fast!
Gordie Bird
62 MGA
67 BGT



> -----Original Message-----
> From: shop-talk-owner@autox.team.net
> [mailto:shop-talk-owner@autox.team.net]On Behalf Of Randall Young
> Sent: Saturday, October 11, 2003 6:48 PM
> To: shop-talk
> Subject: RE: Engineering Question Regarding Splined Shafts
>
>
>
> > Maybe I don't know what spines are. I thought that thing-a-ma-bob on a
> > drive shaft was a spline. Most I've seen aren't very tight -- they slide
> > back and forth a lot. If wire wheel like splines are stronger why aren't
> > they used on a drive shaft?
>
> Basically because there's no need for more strength in that
> application, and
> it costs money to make splines.  Wire wheel splines are also more
> sensitive
> to the sort of wear that would result from constant movement like a
> driveshaft sees.
>
> > I assume one of the reasons a spline is used in any situation is the
> > connect/disconnect is easily done (wire wheels in a pit stop in 1909),
> > or there is a need for lengthwise movement as on a drive shaft. Seems to
> > me that the nature of a spline implies a loose sliding fit.
>
> What about wheel studs ?  On most cars, they're kept from rotating by a
> spline ...
>
> > I guess I'll
> > have to agree that if the splines on wire wheels fit tight enough to
> > have to be pressed on rather than loose enough to side on and off
> > easily, we probably would not have to worry about tightening the
> > knockoffs and the wear situation with them would not exist.
>
> Not my point at all.
>
> > Again, in my mind, a spline with 1000 Billion elements and which still
> > slides on and off easily is a bearing.
>
> There is a definite limit on the useful length of a spline
> 'finger', related
> to it's thickness.  For the narrow splines to be stronger, they
> must also be
> shorter.  For your "1000 billion" case, they would have to be
> very short ...
> so I guess you're right, to make it slide easily the clearance
> would have to
> be larger than the length of the splines, which means they would slip and
> form a bearing (although probably not a very good one).  But since it's
> silly to build a spline with a clearance larger than the length of the
> splines; and not all splined joints are designed to slide easily;
> I stand by
> my statement <g>
>
> > In bending or deflection it's true the material at the center is at rest
> > and the forces are at the surface or edges of what is bending. I'm not
> > so sure that is true in a shear situation. If it is, then we should have
> > lots more spline elements on a drive shaft than we do - say make each
> > one about as thick as a piece of aluminum foil. I don't think that would
> > be very strong, no matter how many there were.
>
> It would be very strong, _if_ the splines were square (ie only as high as
> the thickness of aluminum foil or about .001") and the clearance
> was only a
> few percent of that.  Wouldn't slide worth a darn, tho.
>
> > I am learning that software engineering and automotive engineering are
> > very different.
>
> They are very different in some ways, I'll agree.  But, there are
> tradeoffs
> to be made in both disciplines, and much of engineering is balancing the
> tradeoffs.  For mechanical engineering, the tradeoffs are usually about
> cost, weight, ease of repair and lifetime.  For software, we don't worry
> about lifetime usually (although if you believe it doesn't have one, I've
> got a copy of CPM I'd like to sell you <g>).  But to compensate
> for that, we
> have several different ways to measure weight (execution speed, plus
> requirements for RAM, ROM, disc, CPU, etc.) that can be optimized
> independently and are usually somewhat contradictory.  Cost and ease of
> repair are concerns for both ...
>
> > In software it works or it doesn't, and if it works once
> > it will work forever
>
> Can't say I see it that way.  Take for example, Windows 95.  When it was
> first released for production, it was rumored that there were some 10,000
> bugs already logged against it.  I've not seen every one myself, but I've
> seen enough to make that number believable.  And yet, Win95 met it's most
> important goals/specifications ... it made lots of money for Bill
> Gates and
> helped keep Microsoft in it's position as ruler of the desktop operating
> system.  So, does it work, or not work?  I certainly saw it work once ...
> but it sure doesn't work forever !  (You're doing good to get 6 months out
> of it without a major overhaul <g>)
>
> > Software doesn't really exist in a physical
> > sense... It's soft-ware... You know what I mean :-)
>
> Well, I think so ... but I work in the nether regions known as firmware or
> embedded systems.  Without software, that microwave oven or GPS
> receiver is
> just a pile of parts.  To me, information (the stuff software is
> made of) is
> just as real as electrons, even though I can't see either of them
> directly.
>
> >  In automotive
> > engineering most anything will work at least once,
>
> Hehe ... I know a lot of kit car builders that would disagree with that
> statement !
>
> > and even the best
> > wears out eventually -- it's just a question of when.
>
> Ok, _that_ I'll agree with !
>
> I should hasten to add that I'm not a mechanical engineer, nor do
> I play one
> on television.  All opinions expressed are my own, and do not necessarily
> reflect those of this station.
>
> Your Mileage May Vary.
> Randall





<Prev in Thread] Current Thread [Next in Thread>