Aligning a threaded shaft connection 4

[smarte]

Hello,

I want to design a shaft connection using a threaded bore and threaded shaft end to transmit torque. I want to have a straight parallel interferrence fit (on the shafts) above and below the thread to accuractly align the shafts. Does any one have experience with this kind of connection or know of any availible standards.

Thanks,
Sean

 

[desertfox]

Hi smarte

Why not just use the interference fit into a standard coupling to transmit the torque?
I say this because depending how you apply the torque relative to the threaded shaft joint it might have a tendency to undo and hence the reason I assume you want an interference fit.

regards

desertfox

 

[smarte]

The shaft is only spinning in one direction so the thread will only tighten, never undo. I am not using the interference to prevent the thread from undoing. I am using the interference to align the shafts precisely as a typical thread has to be built with clearances. A NPT thread (I am assuming this is what you mean by standard coupling) is also imprecise.

 

[GregLocock]

Drill chucks on small Wolf drills used to be mounted via a blind thread. I don't know if the centreing was especially good - on a small drill it is probably not very important.

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[desertfox]

hi smarte

The coupling I mean is one you buy off the shelf.

http://www.lenze.co.uk/luk7.htm

Like the above link.

Can you give more info on your application?

regards desertfox

 

[MikeHalloran]

You couldn't just buy a longer shaft?



Mike Halloran
Pembroke Pines, FL, USA

 

[unclesyd]

If you need that type of precision you might want to try a Polygon Coupling. The first two sites are information and the last two are job shops.

http://www.jobshop.com/techinfo/papers/grinding2.shtml

http://www.americanmachinist.com/304/Issue/Article/False/9165/

http://www.generalpolygon.com/index.htm

http://www.stoffelpolygon.com/

 

[smarte]

Fox: I have one shaft mounted vertically in bearings. The second shaft will have a motor rotor pressed on to it and be outboard (above). I need a fully rigid and aligned connection between the two shafts to ensure the rotor and stator are aligned. Hence a flexible coupling is inappropriate. Space is tight and we will be custom machining the two shafts.

Mike: I have components of a smaller diameter than the ends of the shaft, hence a need for a 2-piece design.

Unclesyd: Thanks, for the info. I don't think a polygon drive is the best fit for me as my torque is very low but I did find the articles interesting

 

[desertfox]

Hi smarte

Only way I can see for you to do this interference fit and thread is:-

1/ Ensure that the mating shaft ends engage with
each other before the threads engage

2/ This would mean selecting tolerences for the
threads to be as loose as possible to allow
for misalignment whilst still being able to
engage them.

I see problems with this however firstly for engagement of
the threads you after overcome the interference fit you select between the shafts secondly you can't see the threads
your trying to connect.
If your lucky enough to engage the threads, you now have to tighten them by rotating one of the shafts against the interference fit till its fully home and as you don't say what size shafts or interference you intend to use I cannot get a feel for the torque required to make the assembly.

regards desertfox
for there misalignment

 

[smarte]

Desert Fox:

At the moment I have designed the connection so there are about 5 threads engaged on a 1-7/16 UNEF profile, BEFORE the interferences (top and bottom) engage. The bottom interference engages slightly before the top. This way you can thread the shafts together and then use the threads to pull the shafts fully home. I am concerned a bit with screwing an interference together as I have never seen this done before.

In your post you say the interference must engage first, this is not the case unless I am missing something. It is true you cannot see the threads but I don't see this as a problem. I was thinking of a 0.5-1.5 thou interference on the diameter but haven't done the calcs. yet to see the stresses and torques required.

Sean

 

[desertfox]

Hi smarte

Provided the clearence in your threaded connection is greater than your interference fit your right in that you
can engage the threads first,the clearence in the threads must also allow for any eccentricity of thread centres during manufacture. I was thinking that the thread
engagement might control alignment of the two shafts.
I have never seen two shafts with an interference fit screwed together and depending on the fit etc you will have to make provision somewhere on the shafts for clamping and rotating respective shafts.
Over what length of engagement will the interference fit be
and what are the respective shaft sizes?

regards

desertfox

 

[smarte]

Desert Fox:

I just want to make sure we are both talking about apples here so I have drawn a little sketch (is there a better way to post sketches?)

[tt] _______________________
|________ |______
^^^^^^^^^^------ ^^^^^^^^^^^------
|
|
______| ______
--------^^^^^^^^^^ -------^^^^^^^^^^^
| |_______________________[/tt]

With this design 5 threads (^) are engaged before the interferences (_ and -'s) comes into contact. The length of the interferences are about 0.21 inches on 1.455" and 1.390" diameters.

 

[GregLocock]

If you replace the parallel bore with a locking taper you'll have a nicer joint, in my opinion.



To include an image in your post:
(img

http://www.mysite.com/images/happy.gif

)

but replace the parentheses with square brackets


Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[smarte]

Greg Locock:

A locking taper, which is screwed together, has a very high change of picking up and galling when you undo it. (I had a bad experience with this before)

I am also looking at using a locking taper and separate bolt for drawing the shafts together. The idea with the parallel bore and integral thread was suggested to me. I did not have any experience with this type of connection so I posted my question on this board.

 

[GregLocock]

Ah, yes that is a good point, I was gently wondering to myself why we always use a loose bolt or nut to assemble tapers.

_______________________
|___ |______
|_____^^^^^^^^^^___--- ^^^^^^^^^^^------
|


Or some variation on that with a similar series of steps on the female part would be easier to assemble. How important is dissassembly?



Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[MikeHalloran]

Your thread choice suggests that the OD may be as small as 1.50", which in turn suggests that it's time to do the math and check the stresses.

Be aware that assembly reduces the volume of two closed cavities, one around the threads and one at the bottom of the female part's bore. You might consider venting those cavities.

The joint is multiply constrained. The threads by themselves would tend to align the parts as the shoulder and female shaft end are drawn together. Both straight bores also tend to align the parts. If all the parts are perfect, it could work. If any of the salient features are imperfectly located, then assembly will induce stresses within, as the various locating features fight each other. The female part in particular seems a manufacturing nightmare, given the tolerances so far implied. You might allleviate some of the assembly stresses with well radiused necks at either end of the male threads; then you have to balance lateral flexibility of the necked areas against allowable eccentricity tolerances.

Absent a complete picture of all your actual constraints, I'd be inclined to go with a locking taper and a long slender drawbolt down the center.




Mike Halloran
Pembroke Pines, FL, USA

 

[unclesyd]

Getting the concentric holes at the stated precision isn't that big of problem. Ideally one would use gun drills to make the holes and the ream if necessary.
The internal threads will be a problem unless you have enough quantity to warrant a tap with a long pilot, though you could hold pretty close in a lathe.

I would gun drill the small diameter, then the major diameter for the threads, the bore the larger diameter. This is just one scenario so a talk with a gun drilling shop would be of great benefit.

Another possibility is bore the small dia. then ream; the use a step drill with a non drill pilot or counterbore with pilot to make the thread major dia. then bore the large diameter.

Special Tools

http://www.national-tool.com/index.html

Diamond Hones

http://www.accucutdiamond.com/tooling.htm

Gun drills

http://www.drillmasterseldorado.com/index.htm

Gun drills

http://www.tbt-usa.com/index.htm

 

[desertfox]

Hi smarte

Yes I envisaged what you have drawn.
I don't understand your diameters of 1.455" and 1.390" unless you mean there the outer diameters of the respective
shafts.
I meant what were the diameters of the two parts that will form the interference.

regards

desertfox

 

[Tofflemire]

Hi Smarte,

If you are custom machining the shafts why not make one shaft with the ends milled to fit your bearings and motor?

Tofflemire

 

[smarte]

GregLocock:

To get the sketch to come through right, try using the monospaced text formatting {tt}monospaced text{/tt} but with square brackets. My actual part does have reliefs as you have shown to make assembly easier, I just did not show them in my sketch. (If others want to see Greg's sketch copy and paste it into the reply window.) Disassembly is not the most important feature but it will be taken apart a few times in its lifetime.

MikeHalloran:

One of the shafts is hollow, however you are correct there are two cavities at the start and end of the thread. I will vent these, thanks for the tip. The original design I looked at (and am still considering) has a hollow shaft that is closed at one end, with a short drawn down bolt. Building in flexibility is an interesting idea.

Desertfox:

These are the interference diameters; 1.390" is the standard tap drill size for a 1-7/16 UNEF thread. Although it may change slightly depending on the desired thread tolerance and clearance between the first interference face on the male threaded shaft and female thread. 1.455" is a slightly larger diameter than the major diameter of the thread (1.4375").

Tofflemire:

I have components which mount in the middle of the shaft which are have a smaller bore than either end, hence the need to split the shaft into 2 pieces.


Thanks to everyone for all the help, it is greatly appreciated!!