Here's a simple mod that's been in my brain for awhile so finally decided to go ahead and make up a modified rear axle bearing lock nut plate. That's the "washer" with four holes that fits on a full float rear axle spindle between the rear axle outer hub bearing and the lock nut/collar.
(Scroll down to the photos to avoid all the words)
Background:
Was in the middle of a rear axle service and after using a torque wrench to install and settle the new bearings (following the FSM procedure using 43 ft lbs x 2) I found that at one position where the lock nut would line up with the notch on the spindle (so that the set screws would line up with holes in the plate) the bearing preload was at the very low end of the scale (6 lbs rotational torque), and it felt too easy to rotate the hub by hand.
And when I moved the locking collar tighter to the next available position the bearing preload was now at the very high end of the range (12-13lbs) if not a bit past. To get to that position I had to set my torque wrench to 55-60 ft lbs and once I got there the hub was very hard to turn, way too much preload.
So I decided to have four extra holes drilled into the lock plate in-between the four holes already there for a total of eight holes.
This mod may cause a bit of debate, ie: seems like most people set the rear wheel bearing preload by feel because there's too many variables when setting preload if using a fish scale, or maybe they just slam it with a sledge hammer like a well-known highly experienced Land Cruiser mechanic in Australia (he strikes the mounted tire, not the hub, to help settle the bearings).
This topic (not using a fish scale) for setting bearing preload is fairly well settled for the Front axle, but is it for the rear??
The way I look at it, it doesn't hurt to use some method to tell you where you are in the range (~5.7-12.8 lbf), and it is in the FSM, so I went with an ANALOG (mechanical) fish scale ie: a simple spring dial-type scale.
Reason being I've found that the digital "fish" scales, all use some sort of load cell to determine weight which (all) seem to have a certain amount of delay in the readout so as you're pulling to check rotational starting torque of the hub you get numbers on the digital display like 3.9, 5.2, 7.5, 9.3,--- then the hub rotates, but wait, did it rotate at 7.5 or 9.3 or ??
Point is IME, when using a simple mechanical spring dial gauge it's easier to determine when the rear hub starts to rotate compared to a digital instrument. There are better dial type measuring torque wrenches that are very accurate but they cost a few hundred dollars and you'd need another tool to attach it to the hub.
FWIW I've found that the numbers I was getting on the rear hubs using a spring gauge were reproducible about 7-8 times out of 10 ie: occassionally I would get a lower number and sometimes I'd get a higher number, but rotating the hub again (and tossing out those "outlier" numbers) that the preload numbers (starting torque using the dial gauge) were very similar ie: if you took 10-15 measurements and tossed the wild ones, most came out the same.
FWIW I also checked rotational torque with zero preload on a set of new (grease packed) bearings along with the new design Terrain Tamer HD Casette type hub seals, and got just 1 lb starting torque.
----------------------------------------------------------------------------------
I realize that the vast majority of people go by feel when setting preload on their wheel bearings and that's all good. The point of this thread however is that IMHO there just weren't enough choices (hole positions) in the original lock nut plate design to get the bearing preload I wanted (whether set by feel or by using a gauge, or both, which is the method I'm using).
I suppose that one way to look at it is that this design with only four holes in the lock nut plate has worked well for something like fifty years so why bother?? Can't argue with that other than my recent experience as mentioned above: the first set screw position after following the FSM procedure was at the very low side of the range and felt too loose and the next available position was at the very top of the range and felt way too tight (repeating myself).
Either way, as always, the photos tell the story:
----------------------------------------------------------------------
First is a photo of the current style OEM plate for the rear axle full-floating bearing lock nut plate. They still have four holes but now come with what appears to be a thin ?Molybdenum coating and the hole chamfers are less pronounced compared to an original (25 year old) plate (not shown, yet).
The next group of photos below show my modified lock plate with four extra holes drilled in-between the original four holes.
Note that the ID of the plate is not a perfect circle, more like two D's back to back which changes as you get near the bottom. This means the measurements to locate the holes (the "bolt circle") must be taken from the outer edge of the plate (which is a circle). FWIW these were modified for me by a part-time (old-timer) machinist to my rough specs. Will post the few numbers I have but this was done on pencil and paper, so I don't have any fancy Computer graphics to post up.
Others who are machinists and/or handy with computer programs I'm sure will be able to put this into some sort of Machine/software program for anyone to reproduce. (a Royalty fee of $75 will be collected via paypal for each plate modified. See my attorney's contact information at the end of this tome).
Four new holes at 12 and 6, 9 and 3 O'clock positions. Note the hole at 6 O'clock looks like it's too close to the inner edge but it is in the "bolt circle" (on the same circular line as the other holes). It just ends up in that location due to the design of the plate. Most forces on a set screw at that location IMO would be radial ie: left or right which is the purpose of the screw, so the lock nut doesn't back off.
Once the holes are all in the same bolt circle (so they line up with the threaded holes in the locking nut) the other key specification is that each set of two opposing holes are in the same line/axis radially (directly across from each other on the circle).
---------------------------------------------------------------------------------
The next photo shows the diameter of the unthreaded tip of the set screw that goes into the hole in the plate. The holes in the plate itself are approximately 4.7mm or 0.185 inches. We noted there was slight variability in the hole diameters in each original plate and between original plates. There also was some variablity in the degree of chamfering around the holes in the original plates however more recent production plates have very little chamfer.
Main point if anyone wants to do this themselves is that the holes in the plate are larger than the set screw tip by somewhere around 0.4mm or 0.0156 inches ie: a sloppy fit to allow some leeway for installing the set screws.
Note also that the set screws have a tapered tip, and the lock plate is a loose fit on the spindle, so when you tighten up the lock nut the plate that it's squeezing down on has a tendency to shift IME. So with the tapered tip, if it can catch part of a hole that isn't centered, as the screw is tightened down the plate should shift into position for the full diameter of the post to pass into that hole, if that's clear as Mud.
Same measurement in inch decimals:
Just tossed this last photo in to show having more holes is not a new idea, this is a lock nut from some other vehicle (not a Land Cruiser)
(Scroll down to the photos to avoid all the words)
Background:
Was in the middle of a rear axle service and after using a torque wrench to install and settle the new bearings (following the FSM procedure using 43 ft lbs x 2) I found that at one position where the lock nut would line up with the notch on the spindle (so that the set screws would line up with holes in the plate) the bearing preload was at the very low end of the scale (6 lbs rotational torque), and it felt too easy to rotate the hub by hand.
And when I moved the locking collar tighter to the next available position the bearing preload was now at the very high end of the range (12-13lbs) if not a bit past. To get to that position I had to set my torque wrench to 55-60 ft lbs and once I got there the hub was very hard to turn, way too much preload.
So I decided to have four extra holes drilled into the lock plate in-between the four holes already there for a total of eight holes.
This mod may cause a bit of debate, ie: seems like most people set the rear wheel bearing preload by feel because there's too many variables when setting preload if using a fish scale, or maybe they just slam it with a sledge hammer like a well-known highly experienced Land Cruiser mechanic in Australia (he strikes the mounted tire, not the hub, to help settle the bearings).
This topic (not using a fish scale) for setting bearing preload is fairly well settled for the Front axle, but is it for the rear??
The way I look at it, it doesn't hurt to use some method to tell you where you are in the range (~5.7-12.8 lbf), and it is in the FSM, so I went with an ANALOG (mechanical) fish scale ie: a simple spring dial-type scale.
Reason being I've found that the digital "fish" scales, all use some sort of load cell to determine weight which (all) seem to have a certain amount of delay in the readout so as you're pulling to check rotational starting torque of the hub you get numbers on the digital display like 3.9, 5.2, 7.5, 9.3,--- then the hub rotates, but wait, did it rotate at 7.5 or 9.3 or ??
Point is IME, when using a simple mechanical spring dial gauge it's easier to determine when the rear hub starts to rotate compared to a digital instrument. There are better dial type measuring torque wrenches that are very accurate but they cost a few hundred dollars and you'd need another tool to attach it to the hub.
FWIW I've found that the numbers I was getting on the rear hubs using a spring gauge were reproducible about 7-8 times out of 10 ie: occassionally I would get a lower number and sometimes I'd get a higher number, but rotating the hub again (and tossing out those "outlier" numbers) that the preload numbers (starting torque using the dial gauge) were very similar ie: if you took 10-15 measurements and tossed the wild ones, most came out the same.
FWIW I also checked rotational torque with zero preload on a set of new (grease packed) bearings along with the new design Terrain Tamer HD Casette type hub seals, and got just 1 lb starting torque.
----------------------------------------------------------------------------------
I realize that the vast majority of people go by feel when setting preload on their wheel bearings and that's all good. The point of this thread however is that IMHO there just weren't enough choices (hole positions) in the original lock nut plate design to get the bearing preload I wanted (whether set by feel or by using a gauge, or both, which is the method I'm using).
I suppose that one way to look at it is that this design with only four holes in the lock nut plate has worked well for something like fifty years so why bother?? Can't argue with that other than my recent experience as mentioned above: the first set screw position after following the FSM procedure was at the very low side of the range and felt too loose and the next available position was at the very top of the range and felt way too tight (repeating myself).
Either way, as always, the photos tell the story:
----------------------------------------------------------------------
First is a photo of the current style OEM plate for the rear axle full-floating bearing lock nut plate. They still have four holes but now come with what appears to be a thin ?Molybdenum coating and the hole chamfers are less pronounced compared to an original (25 year old) plate (not shown, yet).
The next group of photos below show my modified lock plate with four extra holes drilled in-between the original four holes.
Note that the ID of the plate is not a perfect circle, more like two D's back to back which changes as you get near the bottom. This means the measurements to locate the holes (the "bolt circle") must be taken from the outer edge of the plate (which is a circle). FWIW these were modified for me by a part-time (old-timer) machinist to my rough specs. Will post the few numbers I have but this was done on pencil and paper, so I don't have any fancy Computer graphics to post up.
Others who are machinists and/or handy with computer programs I'm sure will be able to put this into some sort of Machine/software program for anyone to reproduce. (a Royalty fee of $75 will be collected via paypal for each plate modified. See my attorney's contact information at the end of this tome).
Four new holes at 12 and 6, 9 and 3 O'clock positions. Note the hole at 6 O'clock looks like it's too close to the inner edge but it is in the "bolt circle" (on the same circular line as the other holes). It just ends up in that location due to the design of the plate. Most forces on a set screw at that location IMO would be radial ie: left or right which is the purpose of the screw, so the lock nut doesn't back off.
Once the holes are all in the same bolt circle (so they line up with the threaded holes in the locking nut) the other key specification is that each set of two opposing holes are in the same line/axis radially (directly across from each other on the circle).
---------------------------------------------------------------------------------
The next photo shows the diameter of the unthreaded tip of the set screw that goes into the hole in the plate. The holes in the plate itself are approximately 4.7mm or 0.185 inches. We noted there was slight variability in the hole diameters in each original plate and between original plates. There also was some variablity in the degree of chamfering around the holes in the original plates however more recent production plates have very little chamfer.
Main point if anyone wants to do this themselves is that the holes in the plate are larger than the set screw tip by somewhere around 0.4mm or 0.0156 inches ie: a sloppy fit to allow some leeway for installing the set screws.
Note also that the set screws have a tapered tip, and the lock plate is a loose fit on the spindle, so when you tighten up the lock nut the plate that it's squeezing down on has a tendency to shift IME. So with the tapered tip, if it can catch part of a hole that isn't centered, as the screw is tightened down the plate should shift into position for the full diameter of the post to pass into that hole, if that's clear as Mud.
Same measurement in inch decimals:
Just tossed this last photo in to show having more holes is not a new idea, this is a lock nut from some other vehicle (not a Land Cruiser)
Last edited: