Explain correct behavior of NWSL driveline couplings

[mmagliaro]

This is something that has puzzled me for some time.
I am talking about the common "dogbone" type of NWSL coupling where you have two cups, one on the motor
shaft, and one on an independent drive shaft holding the worm, with the "dogbone" between them.

Like this (I appropriated this from somebody's on-line article... but I trust they don't mind)


Here's what I don't get.
1. Is the dogbone suppose to fit snugly in those cups or is it supposed to be quite loose?   I ask because all the ones
I've ever seen fit pretty tight... so tight in fact that you cannot push the ball all the way into the cup.

2. While I can see how this forms a quasi-universal joint to compensate for misaligned shafts, I have always found that even the slightest
misalignment makes the drivetrain run quite poor, and in fact, usually piece of rubber tubing ends up doing a better job!

When the coupling is very short, like the one in the photo, the problem is far far worse.

3. What about axial play?  Are these supposed to let the motor and worm shaft freely slide back and forth independently of each
other?  I think they really should.  That's one thing I don't like about rubber-tube couplings and I don't like it about these either.
When the worm shaft gets pushed in its bearings, it pushes right through to the motor armature and vise versa.

Ideally, I would want it to be the kind of coupling used in the Kato Mikado, where the motor and worm shaft are each free to independently float back and forth.

Is the installer (me) supposed to trim  these things a little so they can move axially and have some loose play in the cups?

Thanks.

and forth

[Chris333]

It's been a while, but I think I bored out the centers so the ball just fit.  I have seen factory joints where the ball snaps into the couplings, but once snapped into place it was slightly free.

[Centercab]

I prefer a small piece of rubber tubing when the joint is so close to each other.  In an application like this, it would need some kind of strap to hold the gearbox in place.  Otherwise, the joint would separate.  When I use the cup and doggone, I like to have just a bare minimum of play so nothing binds up.

[u18b]

Old Mehano stuff--- very tight.  "Snaps" in place.


Kumata stuff- dogbone is 100% loose.

Since the KMT cups almost always crack, I have replaced many.
The last on I replaced, I dug and old Mehano coupling out of my parts box.
Ran wonderfully quiet (with a few other tweaks).

[mmagliaro]

Well, ... okay, I'll let the other shoe drop.
I got distracted and started experimenting with ways to regear a Kato Mikado without changing the motor or
going to an expensive motor swap with a gearhead.

I am having some success.  The details will follow when I am ready and have tested it.

But meanwhile, on my experimental engine, I don't have that nice sliding star coupling thing that Kato uses in there,
so I have to substitute it with something.

The worm is held in place in it own bearings with a snap-on retainer above them, so there is no problem with the worm
trying to jump out or walk away with the dogbone in there.   I tried the rubber tubing idea, and it was "okay", but the worm
and motor play that game of loading each other against their bearings.  You can slide the tube this way and that and get things
set up so that both the worm and the motor just have a hair of play either way.   But even so, I wanted them to be able to
slide axially, independently, the way the original Kato design was.

I have now gotten the dogbone to work.   Chris' suggestion about drilling out the cups was key.  The dogbone balls need to
slip in those cups with no binding, and they need to be able to slide back and forth in the cups with no binding.   So drill,
and then a few swipes with a flat jewelers file through the slots in the cups and it works really well.

It's not easy to drill the engineering plastic these things are made of.  It just wants to deflect and bend.  You just have to find a
good size bit an spin it in the cup many times to gradually open up clearance in there.

[Chris333]

I used a bit that looks like a 1/8" end mill in my Dremel tool to drill the cups if I remember correctly.

[LV LOU]

 One thing about universal type shafts like that.Because of the speeding up & slowing down of the cups because of the pin sliding and slightly changing the ratio,it's usually a little smoother having the pins at 90 degrees from each other..Also,the shafts should be inline,and parallel for best operation..

[victor miranda]

Hi Max,
the ball should fit 'not tight'
the 2 or 3 thousands of a normal slip fit is what you want.

the tight fit is a mistake of one sort or another
Kato has knobed ball u-joints that pop together, once they are linked the ball is loose.

your problem is that you do not have a lot of room to get your joint done.
I thought kato used the mike short joint in the c55 also.
it does not allow to much miss alignment in any case.

you can make a pretty good short joint by using the NWSL knobed balls on the two
shafts to be jointed and cutting a piece of brass tubing to go over them
you can get the whole joint done in about the length of the dogbone.
end play is always a concern in terms of clearance, however the joint I describe takes up less than 6 mm total.
and you will have about a half mm of end-play available.

I made one recently for my RR pacific project.
I needed a mule to test a tender pick-up project...

victor

who needs sleep?
I went to take a photo.... and can't get into photobucket....

the ball part of the joint is about 2mm long  so about the shortest you can get the joint is 4.5mm
build the joint carefully.

be careful with the NWSL joints.  they require a flatspot on the shaft.

[mmagliaro]

A flat spot?  Why?  The hole in the cup is round and it presses onto the 1.5mm motor shaft (or worm shaft) just fine.
Is the problem that it will eventually stress-crack unless there is a flatted area to relieve some pressure?
(I could spin the motor or worm and take a gnat's eyelash of metal off so the cup pushes on more easily, and then
depend on LocTite to keep it in place).

Or will the joint start slipping over time?

[Chris333]

Victor is talking about the "dogbone". NWSL makes whole ones or you can make your own with 1.5mm shafting. If you make your own dogbone you file a flat spot for the ball to join. Look at the NWSL sprue they aren't through holes.

[victor miranda]

yeah, what chris said.

the ball part of 'make your own shaft/dogbone.'
It has a flat place.  I have popped a few before I noticed it.

victor

[mmagliaro]

yeah, what chris said.

the ball part of 'make your own shaft/dogbone.'
It has a flat place.  I have popped a few before I noticed it.

victor

Oh  yes yes... The little ball with ears on it.  You can also use it to make a 1-ended joint, where you stick that little thing
on the shaft and use only 1 cup on the other side.   Yeah, it has a little flat side in it, that's right.

[victor miranda]

Hi max,
think outside the balls.

I have more room than I think you have.
in the case of the pacific anyway. 
the problem is that I do not like how little spare there is if I try to install a NSWL dogbone.

so you get the shafts close, put a ball on each end and put a brass double-ended cup over the  pair.
(the other ball is in the loco at the top of the photo.)

the photo I took last night...

[mmagliaro]

Excellent idea, Victor.
And I like that you have the slots in the tube cut at 90 degree angles from each other (as Lou suggests, this probably
makes for better balance and smoother running, because the forces aren't all on one plane spinning on that axis).

I did manage to fit the full dogbone with 2 cups in there (see my new Mikado regearing thread).   I just had to cut the two shafts
(motor and worm) shorter, trim a little off the end of each cup, and finally, I actually sliced a little off the back
of each ball of the dogbone.  It runs beatifully smooth.  All the noise in my regeared Mikado is coming from the
quasi-butchered gear mesh I have in there at the moment.  Unfortunately, the engine runs so well, I am hesitant to take it apart again!

[peteski]

One thing that bothers me is how thin the metal coupling tube is (and the thin sharp edges which the plastic nipples will rub against. Will the edge of the tube eventually cut  into the nipples?

BTW, when I first looked at the photo I thought that Victor was using some sort of strange blue grease. But I think it is some sort of reflection.