Installing Rear Hub On A '40

[Tim Ayers]

Originally Posted by 51woodie

Tim. In my experience, when a shaft is machined properly, the finish is nice and shiny. Some machinists can get a finish that is mirror like. When two parts are lapped together with compound, the finished surface is "ground" away at the high spots by the lapping compound, leaving a dull surface, similar to the surface of 1000 grit wet paper. An example is lapping valves to their respective seats. The seat of a new valve is nice and shiny, but when the valve is lapped, you will see a dull mark the width of the mating surface on the valve seat. Unless you have some real special compound, I think you have it cleaned with what you have done. IMHO.

Well, it was exactly 51 Woodie described. The axle was perfect shiny metal and, after real close inspection, the lapping compound took the sheen off. I got it really clean and should be good to go.

[JSeery]

How taper fits work is not related to Ford. Taper connections are used in a lot of applications and require surfaces that match well and are clean.

[Floyd]

All well said. Now lets throw some numbers on here. Dig out your slide rule and lets go.
As to the key , 1/4 X1/4 X approx. 2.8 ", using typical AISI 1018 material, we get a possible max torque transmission of 540 ft-lbs. per axle side,
A good flathead can make about 185 ft-lbs torque max. Using a low 4:11 rear end, the most that could be seen at the axles would be 760 ft-lbs total. The differential action has to be taken in consideration. This assumes everything is perfect. Real world reduces these numbers by 20 to 50%
Now to the tapered axle and hub. At 250 ft-lbs torque on the axle nut, clean and lapped smooth finish on axle and hub we can get possibly 690 ft-lbs of torque carrying capacity per axle side.
Lets say the differential divides the torque equally, no LSD here, we have a max torque at each axle shaft of 760 / 2 = 380 ft-lbs of torque twisting the axle shaft.
The keyway can handle 540 ft-lbs of torque under ideal conditions which means a perfect fit in the slot of the axle and no fretting, shock loading or fatique etc. If we add in a little safety factor ,1.8, this comes out to a design torque value of 300 ft lbs. which less than and too close to the engine value of 380ft lbs. The key will not do the job alone. See picture above.
As to the self locking taper fit of the axle and hub, which can withstand 690 ft lbs of torque. The engine value of 380 ft-lbs is well within the design limits if everything is perfect. However if a hub is removed for brake work and not reinstalled correctly, say only 150 ft-lbs are applied to the nut, then we can only transmit about 400 ft lbs of torque which is too close to the 380 ft lbs we can see from the engine. Real world conditions can change these numbers drastically and in the wrong direction.
A perfectly assembled taper fit hub and axle will work by itself . A Poorly maintained and set up taper assembly can be prone to failure.
A keyway alone system will barely work in ideal conditions. Poor conditions and it will always fail.
The combination of the taper fit and the keyway and correct axle nut torque is what keeps us safe out on the highway. Its all done with numbers, the Ford guys had good slide rules, we need to use good tools.
My opinion and calcs.

[JSeery]

When abusing Ford banjo rear ends the axle should brake or twist before a properly installed hub would turn on the axle.

[philipswanson]

Thanks for all the great information! Good discussion.

[V8COOPMAN]

Sponsored Links (Register now to hide all advertisements)

Originally Posted by Floyd

The keyway can handle 540 ft-lbs of torque under ideal conditions which means a perfect fit in the slot of the axle and no fretting, shock loading or fatique etc. The key will not do the job alone. See picture above.

A keyway alone system will barely work in ideal conditions. Poor conditions and it will always fail.
The taper fit and correct axle nut torque is what keeps us safe out on the highway.

I know that Floyd spent some meaningful time calculating all those forces. Without re-running his math, I have no reason to doubt that he is at least "in the ballpark", if not spot-on.


I have no desire, nor the physical stamina any more to take-on an almost dumb "Mr. Wizard-like" project, but it would be damned interesting if SOMEONE here on the forum could be convinced to play along and partake in an experiment to remove the KEY from the keyways on the tapered axles on a pre-1949 Ford vehicle. Said 'SOMEONE' would of course replace the drums/hubs, and torque the big nut to the roughly 205-ish ft. lbs. factory figure. Then, a reasonable test drive would follow, not driving like a demented hot rodder, but a normal drive of a mile or two. Return home, remove both big nuts and determine whether or not the hubs had remained tight and in line with their original clocking. I'd be willing to bet a "fifth", purchased tomorrow the 3rd, that a drive on the 4th would prove that the hubs would not slip....more or less convincing folks, for once and for all, that the KEY in the keyway is NOT what drives the hub/wheels/tires.


I believe that there is ONLY ONE person on this forum that MIGHT POSSIBLY be game-enough to undertake this nutty experiment. With that, I wonder if our friend Mart would be up for this challenge with his ol' white pick-up? Surely SOMEONE here must be curious! Mart? Anyone? DD

[Mart]

I'm laughing here, but you did say not like a demented hot rodder...

[Tim Ayers]

I second your Mart nomination. I would think that it would. It is amazing how tight they are on when I'm just placing them on hand tight doing the brakes.

[philipswanson]

Hey if friction between the hub and spindle is what we want, why not use some red Loctite on the spindle for added insurance?

[V8COOPMAN]

Originally Posted by philipswanson

Hey if friction between the hub and spindle is what we want, why not use some red Loctite on the spindle for added insurance?

C'mon, Philip! There's a huge difference between 'friction' and gluing. I believe that the last 90 PLUS years that Fords have been operating with tapered axles and hubs this way should prove that if ya precisely follow Ford's original, printed installation procedures, you won't need "added insurance". Those hubs are difficult-enough to remove as it is (when properly installed). Try to imagine attempting to remove one after some bozo has glued it on. But what the heck do I know?


Can we assume you finally had a successful mating of the shaft into the U-joint? I hadn't heard-back whether or not you finally won that battle. DD

[Floyd]

I will "sixth" the motion that we get the fifth for the fourth and do it on the third.. Make mine Russell's Reserve as I know Jimmy.
Keyways are not efficient and have the load concentrated on only one side. However they are good for limited power transmission like pulleys on straight shafts like your table saw, PTO, band saw, etc. If you want to transmit a lot of power and still have a sliding type assembly fit, you use multiple "keyways" which are then called "splines". Think transmission input and output shafts.
If you want to perfectly align something to a shaft, like a hub for a wheel, you cannot do it with just a sliding fit. The taper joint provides this true alignment IF made correctly. Also the efficiency of the geometry of the correct taper makes it self locking and provides terrific forces that can be used for transmitting power.
Fun fact- If properly made and assembled and tightened, the force to remove the hub will be 13 tons.
Drink up, be safe!

[philipswanson]

Originally Posted by V8COOPMAN

Can we assume you finally had a successful mating of the shaft into the U-joint? I hadn't heard-back whether or not you finally won that battle. DD

Yes but I had to remove one end of the spring first. The wagon spring has extra leaves and it hits the cross member when going in assembled. After that she slipped right in on the second try. I will be putting the Columbia controls in next. Thanks, Phil

[Kube]

I can tell you I had run a lathe with a 6', yeah SIX FOOT swing. The tooling was held in with nothing but taper collets. There were times I'd be taking 3/8" off at a pass.
A properly mated taper is more than enough to handle anything even the strongest flathead can dish out.

[Floyd]

I was waiting for your reply. As a tool and die maker, you have depended on taper fits for every hour of every day that you worked. That is a lot of experience. The good news is that in the tool room you got to enjoy correctly done taper assemblies.

[Kube]

Originally Posted by Floyd

I was waiting for your reply. As a tool and die maker, you have depended on taper fits for every hour of every day that you worked. That is a lot of experience. The good news is that in the tool room you got to enjoy correctly done taper assemblies.

I really enjoyed my career. Best of both worlds for me. I was able to use my hands and my brain. It was nearly always interesting.
Plus, I was actually paid to make things perfect!

That huge lathe I'd mentioned previously was a tad intimidating. I was quite amazed at the chips I was able to remove without the machine giving as much of a grunt.

[40 Deluxe]

Originally Posted by Floyd

All well said. Now lets throw some numbers on here. Dig out your slide rule and lets go.
As to the key , 1/4 X1/4 X approx. 2.8 ", using typical AISI 1018 material, we get a possible max torque transmission of 540 ft-lbs. per axle side,
A good flathead can make about 185 ft-lbs torque max. Using a low 4:11 rear end, the most that could be seen at the axles would be 760 ft-lbs total. The differential action has to be taken in consideration. This assumes everything is perfect. Real world reduces these numbers by 20 to 50%
Now to the tapered axle and hub. At 250 ft-lbs torque on the axle nut, clean and lapped smooth finish on axle and hub we can get possibly 690 ft-lbs of torque carrying capacity per axle side.
Lets say the differential divides the torque equally, no LSD here, we have a max torque at each axle shaft of 760 / 2 = 380 ft-lbs of torque twisting the axle shaft.
The keyway can handle 540 ft-lbs of torque under ideal conditions which means a perfect fit in the slot of the axle and no fretting, shock loading or fatique etc. If we add in a little safety factor ,1.8, this comes out to a design torque value of 300 ft lbs. which less than and too close to the engine value of 380ft lbs. The key will not do the job alone. See picture above.
As to the self locking taper fit of the axle and hub, which can withstand 690 ft lbs of torque. The engine value of 380 ft-lbs is well within the design limits if everything is perfect. However if a hub is removed for brake work and not reinstalled correctly, say only 150 ft-lbs are applied to the nut, then we can only transmit about 400 ft lbs of torque which is too close to the 380 ft lbs we can see from the engine. Real world conditions can change these numbers drastically and in the wrong direction.
A perfectly assembled taper fit hub and axle will work by itself . A Poorly maintained and set up taper assembly can be prone to failure.
A keyway alone system will barely work in ideal conditions. Poor conditions and it will always fail.
The combination of the taper fit and the keyway and correct axle nut torque is what keeps us safe out on the highway. Its all done with numbers, the Ford guys had good slide rules, we need to use good tools.
My opinion and calcs.

Hmm. I don't see the trans. low gear ratio figured in. I used 2.82 as the first gear ratio and came up with a total of 2144 ft. lbs. of torque to divide between the the two axles (minus drivetrain losses-say 20%) so 1700 divided by 2 equals about 850 ft. lbs. max. per axle in first gear.

So from the numbers presented here, it looks like the taper fit alone is not enough! (At least not when the clutch is popped at max RPM in first gear!)

[Aarongriffey]

But Coop, some guys think the rear axle nut is supposed to be tightened just to the point of touching the hub. They think it is like the front wheel bearing nut.
I’ve had guys jump all over the place when the say me tighten that nut.
When that happens I ask them about the nut that holds the steering wheel on.
To prove that the friction will drive the hub just fine a friend of mine bought a ‘26 model T touring and it drove just fine with no axle keys.

[cas3]

i have been waiting for someone to bring up machinist tapers into play, thanks kube for that. jacobs tapers on my r8 mill collets to hold on the drill chuck stay in place without spinning, and without a 200lb torque nut holding it together...just the precise taper fit, never falls apart. well...maybe when too lazy to change collets and try to mill with a drill chuck...but we dont have to go into that, and not mentioning any names here either

[cas3]

to add to Aarons post, every car i have ever owned with tapered axles, many brands, the hubs were loose when i bought it

[cadillac512]

Originally Posted by 40 Deluxe

Hmm. I don't see the trans. low gear ratio figured in. I used 2.82 as the first gear ratio and came up with a total of 2144 ft. lbs. of torque to divide between the the two axles (minus drivetrain losses-say 20%) so 1700 divided by 2 equals about 850 ft. lbs. max. per axle in first gear.

So from the numbers presented here, it looks like the taper fit alone is not enough! (At least not when the clutch is popped at max RPM in first gear!)

Correct! I've built rear axle assemblies for drag cars and that's the way to choose what axle strength is required for a particular engine/trans/torque level/gear ratio. That first gear ratio is a very important factor.


Terry