PDA

View Full Version : Force Vector Theory Question - ??

uute
07-17-2009, 03:22 AM
Or some such jargon. :D

Set up:

A flatted drive shaft drives a load basically at 2 points like a flat screw driver. The shaft is driven in turn by a gear keyed to it by single woodruff key.

Here's the question:

Is there a prefered/advantagous position for the key relative to the drive flats? Supose the keyway was moved 90* around shaft, any difference?

Example that inspired question:

1" shaft running under 10 rpm in 2 needle bearings, drive gear between bearings. The flats form a "dog" to drive ice auger on end of shaft. Keyway lines up w/ center of one flat.

The only possible difference I could see would be tiny gear movement due to clearence fit, which would be there regardless of position. The example shaft shows no wear there.

CT2

philbur
07-17-2009, 04:51 AM
The drive coupling has a varying ability to transmit a bending moment in different planes. An ice auger is going generate forces in all directions.

Phil:)

Or some such jargon. :D

Set up:

A flatted drive shaft drives a load basically at 2 points like a flat screw driver. The shaft is driven in turn by a gear keyed to it by single woodruff key.

Here's the question:

Is there a prefered/advantagous position for the key relative to the drive flats? Supose the keyway was moved 90* around shaft, any difference?

Example that inspired question:

1" shaft running under 10 rpm in 2 needle bearings, drive gear between bearings. The flats form a "dog" to drive ice auger on end of shaft. Keyway lines up w/ center of one flat.

The only possible difference I could see would be tiny gear movement due to clearence fit, which would be there regardless of position. The example shaft shows no wear there.

CT2

agrip
07-17-2009, 08:53 AM
How far apart, shaft axis wise, are the load points?

What is the actual running torque?

What kind of impact, if any, exists?

Ag

A.K. Boomer
07-17-2009, 09:03 AM
You should be good no matter where the key is placed,

Two drive flats, Reminds me of the old volkwagon transaxle drives --- God what an archaic drive system, I guess you could call it independent rear sus. but who cares if your camber is floating +/- all over the place around every turn or bump...

Then came the pulsating U-joints

Then finally the C.V. joint - thank god - we've come along way baby...

BMW had a unique way of transmitting power in their propeller shaft,
A rubber doughnut with 6 steel insert holes --- all holes were internally wrapped to each other with nylon cord and then the entire unit was vulcanized --- three of the holes went to the lead side of the drive and the other three went to the aft.

I blew a few up launching my 1600-2 off the line just about every time I took off, the thing is - is they were all used to begin with so actually they held up surprisingly well and they always gave you a warning and never left you stranded ---- the cords would rip on one side and then the shaft would pulsate while under load but you could still limp it home --- there was a bronze center pivot bearing where the two shafts met and that would keep them aligned and from going off center.

This system was limited to very small angles --- it was used on a rigid mounted rear diff (with independent rear drives) and a rubber mounted trans. and engine.

boslab
07-17-2009, 01:30 PM
the only thing that i can remember is,..draw v diagram, direction is reverse closing line.
mark
[remember bows notation]

Carld
07-17-2009, 01:40 PM
I don't see where it would make any difference where you put the key as long as it's under the gear.

Why would you think it will make a difference??:confused:

uute
07-17-2009, 02:42 PM
Phil - yes, the bending moment was one thing I was trying to get my mind around. Seems that in a rotating system it would cause a similar eccentric where ever the key was located, correct?

agrip - in the example: ~ 3" center of key to stare of engagement on dog, bearings very close to gear above & below (vert. shaft), ~ 1" between bearing & start of dog engagement (overhang?).

I don't know the torque, the drive train produces a LOT, it will twist off 1" shaft!

Shouldn't be any real impact, though load may vary slightly.

Carld wrote: Why would you think it will make a difference??

The question came to me when I drew out the shaft for possibe future replacement (no longer available), noticed I drew key in different quadrant.

Wondered if there were any esoteric engineering reason for key placement. (ease of manufacture prolly had more influence)

Besides, I wanted to start a friendly argument. :D :D

CT2

philbur
07-17-2009, 02:54 PM
My thought was that a none symetrical drive coupling would not transmit a bending moment back to the drive shaft uniformly around 360 degrees. However it's difficult to give a more detailed opinion without a clearer understanding of the arrangement. A photo might help.

Phil:)

PS: the mending moment induced by the auger would also vary depending on the relative position of the auger blade along its shaft.

Phil - yes, the bending moment was one thing I was trying to get my mind around. Seems that in a rotating system it would cause a similar eccentric where ever the key was located, correct?

uute
07-17-2009, 03:33 PM
I meant this as a more general discussion, but since I provided an example:

How bout x-ploded drawings. :D :D

Forgot to resize them, so here are the links:

Shaft:
http://smg.photobucket.com/albums/v300/uute/?actionaction=view&current=OutputShft-IceMac.jpg

Drive Train:
http://smg.photobucket.com/albums/v300/uute/?action=view&current=IceMac-Drive.jpg

http://smg.photobucket.com/albums/v300/uute/?action=view&current=IceMac-Freez.jpg

Sorry about the cut & Paste
I can take some photos if youlike.

CT2

PS, Darned Photobucket pics aren't showing up in IE for me - work fine in Firefox????????????

ckelloug
07-17-2009, 04:19 PM
My instinct says it won't matter but if it's critical, do a finite element analysis.

None of the important net forces will change but some very small deflections (second order and higher) in the immediate area of the keyway and flats could change due to the way the shaft winds up when torqued.

By an off the cuff application of St. Venant's Theorem, there is sure to be no difference outside the immediate region of the flats and the keyway.

So, I'd say that unless there are concerns about either manufacturing or installing the shaft with a particular keyway orientation, I wouldn't expect problems regardless.

agrip
07-17-2009, 05:43 PM
Nearly 3 diameters separation flat to key - - - should not be a problem, put key where convenient.

You did not mention a snap ring groove in the torque zone.

TILT

MAKE SURE the groove bottom is fully radiused.
That sucker WILL be the point of failure. Particularly if the groove bottom has sharp corners. Something like a 10x stress riser.

Ag

uute
07-17-2009, 06:53 PM
Thanks Ag, hadn't even thought about the groove, mostly because the one that did twist off broke where the drive end engaged auger.

The factory parts have a flat bottom groove w/ very small radius in corners. Not discounting what you've said, just telling what I see.

And of course it is a hard carbon steel shaft, I think I put SS on my drawing.

Thanks All for the discussion.

If I attempt to make one, I'll post results. The heat treatment adds a grinding operation that I was hoping to avoid. I haven't got a good tool post grinding setup.
CT2

uute
07-17-2009, 07:46 PM
Photos, if you're really bored:

Shaft & Auger:
http://smg.photobucket.com/albums/v300/uute/?action=view&current=IMG_0443.jpg

20 yrs wear & tear:
http://smg.photobucket.com/albums/v300/uute/?action=view&current=IMG_0449.jpg

http://smg.photobucket.com/albums/v300/uute/?action=view&current=IMG_0442.jpg

CT2

Carld
07-17-2009, 11:24 PM
As I see it the parts are in a straight line.

With that, the flattened sides driven by the the auger and the key driving the pulley would be an inline transfer of power. It would not matter where the keyway is in relation to the flats. It is not a universal joint but even on a universal joint it would not matter where the key is in relation to the U joint.

The shaft is or should be a snug fit in the bore of the pulley and the power transfer would be straight line at that point. The only thing the key is doing is keeping the shaft from spinning in the pulley.

EDIT: Ice auger, opps, the pulley does the driving, ok.

A.K. Boomer
07-17-2009, 11:59 PM
That's not a poor lube failure, note all the straight cut lines going directly to center, big drag poor lube the lines would be spiral,,, What you got there is a case of lots of use over a long long period of time -- the stress is greatest right where the shaft meets the auger - its where flex meets no-flex,,, common place for failure and although its up against just ice - ice is very rigid and when it gives way there's nothing, lots of crunching lots of vibe/frequencies +/- impulses ----- that's what got that shaft end.

uute
07-18-2009, 12:30 AM
Thank you Boomer, that makes sense. It is hard to tell from the drawings, but the auger forces flaked ice scraped off inside of cylinder through tapered (conical) holes in cylinder head, extruding mix into "pellets". I assume this adds additional stress right there.

Stub shaft appears to be separate bit pressed into auger. broke off right at the shoulder.

CT2

A.K. Boomer
07-18-2009, 07:12 AM
Its always nice to know you didn't "jimmy" it up yourself,
another indicator that it took lots of time and wasn't a lube failure is the darkened ring around the outer parameter -- This thing was stress fractured for a long time even before it totally broke.