Why not to re-use an old shaft coupling..

[Maine Sail]

The video below show's the difference a little bit of rust makes.

The fit of a straight coupling to a shaft should always be a light interference fit. This means it will not just slide on.

0.001"is the MAXIMUM allowable variation between the OD of the shaft and the ID of a split coupling which clamps the shaft to get a bite. To put that in perspective a sheet of standard 20 Lb. paper is supposed to be .0038" thick. This means a sheet of copy paper is 380% larger or thicker than the maximum allowable variation between shaft and split coupling. Wow! Straight couplings do not clamp the shaft and the need to be an interference or light press fit with no "slop" or clearance..

If you remove a coupling, and there was rust, between it and the shaft, it is a very good idea to replace the coupling and have the new one fitted and faced to the shaft by a competent machine or prop shafting shop.

Even the shaft for my C-36 ordered from Catalina's in-house shop was a light tap or light interference. I used a phenolic (plastic) hammer to tap this one onto the shaft. This is one of the hammers I used for sounding hulls so the audio sounds is worse than it really is. It would not just slide on by hand nor come off by hand. I used my gear puller to get it back off. If you need to use hammering blows it's too tight but if it just slides on with no resistance at all it is probably to loose. The fit should ideally be a light tap fit or what my prop shop manager refers to as an "interference fit", which is a probably a technically incorrect definition based on the ability to "tap" it on, but one that gets the point across to most customers. As I understand it an "interference fit" is one that will usually require a press hence "light press" or "light interference" fit.

The last thing you want to do is ruin an expensive shaft due to slop, or worse, put the boat in reverse and have your shaft come out of the coupling.

The new coupling in this video was hand fitted and machine faced by a shafting shop and is the proper fit. Total cost was under $85.00, including the coupling. Some fit tad tighter but this one is about as loose as you'd want it to tap onto the shaft.

Unfortunately with an old shaft it sometimes needs to be dressed and it's tough to get a proper fit. Some shops will still send them out the door this way but in reality they should not. couplings are generally shipped under-sized and then made to fit the shaft properly.

The old coupling in this video used to fit just as snug as the new one does but the layer of rust can clearly be a lot.

6.5.5.1 Transmission Coupling End - The coupling end of the propeller shaft and/or the coupling shall be sized to a diameter that permits a maximum clearance of .001 in.

That is for SPLIT couplings not straight. straight couplings fit much tighter.

I recorded this quite a while ago but never got around to uploading it until a discussion recently came up where this would be pertinent.

[ttyouyube]http://www.youtube.com/watch?v=onmkD4PnRlk[/ttyouyube]

 

[jibes138]

This post is giving me fits

I build parts with clearance fits of .0002 to .0005 on diameters that have to be round within .000050 and straight within .0001. Both mating parts are lapped to achieve these fits and tolerances. I think the message here should be that you want a slight interference fit so that the interference fit provides the primary retention of the shaft and not just the set screws which should only be a secondary retention for safety reasons. With engine vibrations and slight mis - alignment you do not want relative motion of the shaft and coupling that would happen with a clearance fit. Now for a 1.000 shaft a .001 interference is a pretty tight interference and will go together very hard. I recommend the interference fit be matched to .0002 to .0005. The hoop stress created should be adequate for retention and yet it will go together a little easier. Keep in mind these are not precision parts and the measuring equipment available in an average machine shop in a boat yard is probably not capable of really controlling this, but it is what I would strive for. To simplify the assembly the shaft should be frozen in liquid nitrogen and the coupling heated, put a hose clamp on the shaft to control how far down the shaft you want the coupling to slide before freezing the shaft. Using thick gloves take the two parts and put them together, should be a slip fit under these conditions. Once the parts reach equilibrium you will have a nice assembly with little to no chance of coming apart in service, no relative motion between the two parts, and secondary retention via the set screws. The assembly should then be faced square as recommended by Mainsail. I have also done assemblies where zinc chromate primer is put on the shaft before freezing so galvanic corrosion inhibiter is inside the joint after assembly.
Just a few tricks of the trade. Keep in mind for interchangeable parts for a .0001 to .0011 diametral fit you need to control each part diameter to +/- .00025 since .0005 of the tolerance will be consumed by each part. Holding parts to these tolerances on simple engine lathes and Bridgeports is questionable other than a highly skilled machinist and even then the roundness and taper come into play inherent in the machine ways and spindle bearings. If you match fit them the whole tolerance can be applied to just one of the parts so you have a little wiggle room.

 

[Stu Jackson]

I understand, but

jibes, you wrote: "To simplify the assembly the shaft should be frozen in liquid nitrogen and the coupling heated, put a hose clamp on the shaft to control how far down the shaft you want the coupling to slide before freezing the shaft..."

Since the shaft has to be slid into the boat before engaging the coupling, how do you do that?

 

[tommays]

The shaft has to be drilled so the SCREWS are the main holding force and then the heads get saftey wire

 

[jibes138]

More on fits

The reason you want to freeze the shaft in liquid nitrogen is stainless steel will typically have a coefficient of expansion of .0000096 inches per inch per degree F. So liquid nitrogen at -321 degrees F provides about 400 degrees F temperature differential. So 400 (assuming 80 degrees ambient air) times.0000096 times 1 inch diameter = .00384 shrinkage of the shaft. Now if you have a .001 interference fit this will mean the shaft is .00284 smaller in diameter than the bore it is going into. This should slide together quickly and easily before the temperature warms up the shaft even without heating the coupling. If you heat the coupling to say 180 degrees you will get another .001 clearance to play with.
Note every material has a unique coefficient of expansion so look up the actual for whatever materials your parts are made from, this is provided as an example only of an assembly process widely in use in industry.

 

[tommays]

We use bearing heaters all the time so dont you ruin the shaft pressing them on

BUT

The heat freeze thing does not work out in most engine bays due to the space

 

[jibes138]

To answer Stu

Good question, you could tilt the boat up on end so the shaft can sit in a vat of liquid N2.
Maybe build a dam around the shaft with something and pour it on while collecting it in a steel tray?
Maybe a better approach would be to use dry ice in the boat. You could sit a chunk on the shaft, still provides a delta T of 189 degrees to 80 degrees ambient (dry ice is minus 109 degrees F) so 189 x .0000096 x 1 = .0018 shrinkage of the shaft.

 

[paulj]

Very interesting subject....my have to get out my 1961 Machinests Hand book.

Slip fit vs press fit, I'm not sure what the original design called for at this joint and if through experience with this joint a new method should be used.

I was wondering what the Rockwell hardness of the shaft and the Rockwell hardness of set screws were and the type of end on the set screws.
Seams there would be some gauling or displacement of metal on the shaft
to secure the flange with set screws.

Maybe I would drill dimples in the shaft after alignment for the set screws just to ease my mind on what is really holding the flange on the shaft.

Grinding the shaft and lapping the bore for tighter tolerances very expensive and I'm not sure.... I may have to break this joint in the near future to make some repairs.


paulj :troll:

 

[tommays]

The shaft has to have "dimples" to be done right it would pull right OUT in reverse without them

 

[Don12364]

If the interference really is 1 thou then heating the mild steel coupling by 83°C (150°F) should do the trick. So putting it into a pot of boiling water should be sufficient and no need for liquid nitrogen, though cardice would be simpler for the shaft (-78°C = -108°F).
In practical terms a gas blowlamp (used above deck) to heat the coupling to 200°C (so that when you spit on it the spit boils) is the way I would go. This would give a 2 thou increase in the coupling bore.

 

[Maine Sail]

Maybe I would drill dimples in the shaft after alignment for the set screws just to ease my mind on what is really holding the flange on the shaft.

No worries if that is the original Catalina shaft you will already have dimples. It is a SAE J 756 standard for marine shafting.

Grinding the shaft and lapping the bore for tighter tolerances very expensive and I'm not sure.... I may have to break this joint in the near future to make some repairs.


paulj :troll:

If your shaft is anything like the one on my 310 and the one I ordered from Catalina for my C-36 it was a light tap or press fit from the factory. I pulled my C-310 shaft after only a few moths to install a PSS and the fit was quite tight. I still had it checked for fit and face and the coupling was within spec to re-use it. This has been the only coupling I have come across that was reusable and within spec when checked because it had not yet had enough time to form much of a layer of rust.

From my train of thought fitting and facing a new coupling, or even just paying to have it checked, is cheaper than a whole new shaft if it happens to get ruined from a sloppy fit..

Holding parts to these tolerances on simple engine lathes and Bridgeports is questionable other than a highly skilled machinist and even then the roundness and taper come into play inherent in the machine ways and spindle bearings. If you match fit them the whole tolerance can be applied to just one of the parts so you have a little wiggle room.

Most shafting shops I have used or worked with hand fit and hand finish the coupling to the shaft, though I have not seen how Catalina did it. It takes time but seems to ensure you get a proper fit. I agree, and so does my friend that runs the shop, that the required fit can not be done solely by the machines most prop shaft shops use.

They usually dress the shaft lightly and remove any burrs near the dimple holes & key way then fit the coupling to the shaft by hand. The coupling in the photo took about 15-20 minutes to hand fit, I was there watching it. It was then checked for face run-out to be within or under the max allowed of .002".

For the coupling to the shaft it's ream, check, ream, check, ream, check, ream, check. Kind of monotonous but it is a fairly precise fit you are hoping to achieve. The shaft clean up only took about 10 minutes including checking the final tolerance for diameter. The shaft was then checked for straightness and it took about 10-15 minutes to set up and check at varying points along the shaft.

New coupling was about $45.00 and total bill was well worth it and under $100.00 but I think closer to $85.00..

Checking the shaft run out:

 

[jibes138]

need a little more

If the interference really is 1 thou then heating the mild steel coupling by 83°C (150°F) should do the trick. So putting it into a pot of boiling water should be sufficient and no need for liquid nitrogen, though cardice would be simpler for the shaft (-78°C = -108°F).
In practical terms a gas blowlamp (used above deck) to heat the coupling to 200°C (so that when you spit on it the spit boils) is the way I would go. This would give a 2 thou increase in the coupling bore.

The problem with either heat or cold is handling the parts. Going too hot also has the hazard of starting something on fire. Just going to say 180 degrees F would only give a line on line fit. You need a few thousands of clearance so they slam together quickly as a clearance fit then immediately seize up as the temperatures normalize. In the non boating shaft world a connection like this is typically done with a spline, for example the output shaft from a car engine to a transmission. A spline is an expensive but highly precise connection made just for this purpose. For cars the process is highly automated to keep the cost down. The male spline is usually done on a hobbing machine and for very precise splines it can be ground on a Reishauer type machine or even form ground. The female spline is either broached or shaped with a gear shaper. i have also used internal form grinders to grind very precise internal splines. A spline connection for this joint would be ideal as it would typically have some grease in it and the rust problem would be solved.
A splined joint would still have the issue of retention of the shaft. A thrust bearing would be ideal but I have see a cross pin used on a splined shaft for retention. I once owned a Triumph Spitfire car (big mistake) and the rear axles were independent suspension. These axles were had a splined joint at the differential housing and a cross pin held it in place. One the pin was installed it was deformed on the opposite side with a press to keep it from coming out. This would be a much better solution for a boat shaft as it would go together quite easily and the relationship of the coupling face to the splined bore could be controlled as a detail part so no need to machine it square as a shaft coupling assemlby. Broaching the coupling spline would only take about 20 seconds. Hobbing the shaft would be pretty cheap in volumes but the variable lengths of all the different boat shafts might pose a problem. One solution would be to have a splined male piece that could attach to the shaft by brazing it on, inertia welding, EB welding it, or a shrink fit using an induction heater might be adequate, the ring gear on a flywheel is usually put on this way and I can't remember hearing of one slipping loose from the torque of a starter.
So maybe we have a better mousetrap??