BMW Luxury Touring Community banner

1 - 14 of 14 Posts

·
Premium Member
Joined
·
177 Posts
Discussion Starter #1
I've been reviewing final drive threads and trying to decide whether to do a preemptive rebuild. I'm at about 43k, with no service records from PO(s). So I don't know if it has ever had an issue.

Last oil change had a fine coat of metal particles on the magnetic plug, but nothing too concerning (no obvious fragments).

Watched CharlieVTs video again (nice work, thanks!), but wondered whether anyone has tried to use crushed solder or lead as a method of measuring the gap prior to shimming. Searched the site, but nothing turned up.

It appears to be used by some shadetree mechanics in checking tolerances for differential rebuilds (which fits my description, I once overhauled a truck engine after pulling it with a
come-along chained to the ceiling joist in the garage and rolling it around on the floor to get to the top and bottom ends).

Method would be to assemble the drive with soft lead shot or solid core solder in the gap and torque to spec., disassemble and measure the thickness of the crushed lead. It would be similar to using Plastigage, but instead of measuring the width, would just measure the crushed thickness.

Might not be any easier, but would seem to be an alternate method for those of us with more fishing shot than dial indicators.
 

·
Registered
Joined
·
4,404 Posts
jrh2020 said:
I've been reviewing final drive threads and trying to decide whether to do a preemptive rebuild. I'm at about 43k, with no service records from PO(s). So I don't know if it has ever had an issue.

Last oil change had a fine coat of metal particles on the magnetic plug, but nothing too concerning (no obvious fragments).

Watched CharlieVTs video again (nice work, thanks!), but wondered whether anyone has tried to use crushed solder or lead as a method of measuring the gap prior to shimming. Searched the site, but nothing turned up.

It appears to be used by some shadetree mechanics in checking tolerances for differential rebuilds (which fits my description, I once overhauled a truck engine after pulling it with a
come-along chained to the ceiling joist in the garage and rolling it around on the floor to get to the top and bottom ends).

Method would be to assemble the drive with soft lead shot or solid core solder in the gap and torque to spec., disassemble and measure the thickness of the crushed lead. It would be similar to using Plastigage, but instead of measuring the width, would just measure the crushed thickness.

Might not be any easier, but would seem to be an alternate method for those of us with more fishing shot than dial indicators.

What you are thinking of doing is using some type of "PlastiGauge" or other compressible material which will retain it's dimensions for measurement after you have disassembled the parts.

Here's your problem as I see it: The crown wheel bearing is a press fit into the cover. In order to remove the bearing from the cover, you'll either need to press or hammer it out which will be bad for a new bearing. Or you can heat the cover and bearing (bearing will be on the crowngear hub) to about 250 degrees F. Heating allows removal of the bearing from the cover using finger pressure. BUT heating to 250 degrees is gonna destroy the accuracy your "plastigauge" measuring material. (Some covers can be removed with less heat, there seems to be some variation here.)

IF you come up with a "plastigauge" material that will maintain dimensional stability with the cover heated enough for easy removal you may have a method.
 

·
Premium Member
Joined
·
1,178 Posts
My concerns with this method would be in no particular order.

1) can you get the lead shot placed evenly around the bearing so it does not side load or tip and give you a false reading?

2) How much force does it take to compress the lead shot? Example: 500 lbs of force = .001 movement or preload on the bearings...

I have only seen plastigage used between two solid metal surfaces like on rod bearings etc. I have never seen plastigage used where one surface can tilt or move like in the wheel bearing assembly.

not sure how valid these concerns are but would at least have to consider them.

Roy
 

·
Registered
Joined
·
12,165 Posts
bigbear said:
My concerns with this method would be in no particular order.

------------------------------------------
2) How much force does it take to compress the lead shot? Example: 500 lbs of force = .001 movement or preload on the bearings...
-------------------------------------------
Roy
That was my concern, if wire solder is used the pressure to flatten it may be more than the proper preload force would be, but even if less you do not know how much. So, there is no way to relate the flattened material to the real clearance.
 

·
Registered
Joined
·
2,176 Posts
Darned good idea though, nice thinking. Only way to know is for 'someone' to try it out and compare it to conventional methods. Would be real interesting, don't you think - CURTIS?
 

·
Registered
Joined
·
4,404 Posts
fpmlt said:
Darned good idea though, nice thinking. Only way to know is for 'someone' to try it out and compare it to conventional methods. Would be real interesting, don't you think - CURTIS?
Nah, I'm not gonna do it. ;)

It has got me thinking again about other "low tech" ways of doing the measurement though.
 

·
Premium Member
Joined
·
1,178 Posts
Having never had one of these final drives apart yet. Do you need to keep the housing at temperature of 250 degrees to take the measurements or just get it broke loose the first time? If temp can be lower during measurement phase the plastigage may be a viable alternative. Not sure at what temp the plastigage turns into plastigoo though.

What is the force required to compress plastigage? Could this still be the concern that Dave S. and I expressed earlier?

Agreed good idea and may just require a little refinement. Did Charlie VT get everthing dialed in the first time? I bet he has modified and improved his procedure from his first try to todays method.

Roy
 

·
Registered
Joined
·
4,404 Posts
bigbear said:
Having never had one of these final drives apart yet. Do you need to keep the housing at temperature of 250 degrees to take the measurements or just get it broke loose the first time? If temp can be lower during measurement phase the plastigage may be a viable alternative. Not sure at what temp the plastigage turns into plastigoo though.

What is the force required to compress plastigage? Could this still be the concern that Dave S. and I expressed earlier?

Agreed good idea and may just require a little refinement. Did Charlie VT get everthing dialed in the first time? I bet he has modified and improved his procedure from his first try to todays method.

Roy
Yes, there is sort of an art to measuring using the dial indicator method. The drives don't all "behave" the same way due to differences in tolerances. Also, drives that have failed often have changes in the bearing seat of the FD cover due to the race having spun.

When doing Dman's dial indicator technique, I heat the cover to 250degrees F and assemble the drive. I put a film of grease on the crownwheel bearing seat in the cover to help things move. I repeat measurements as the cover cools. At first the measurements well be a little on the big side, then as the cover cools the measurements settle down and become repeatable. Then as the cover cools further, the bearing outer race will no longer slide in the cover and measurements are not longer meaningful. Not all drives are the same. Some are easy to get measurments from even with the cover relatively cool, others the bearings doesn't want to move even with the cover close to 250 degrees. This is a function of the variation between covers and bearing in the interference fit between them.

Way back someone suggested a feeler gauge approach. I and others dismissed it as unreliable. I have experimented with it but cannot get the results using feeler gauges to match up with the other two methonds I use.

Here's what the feeler gauges I experimented with look like. Someone might try fussing with this method, but I can't get it to work reliably. Feeler gauges just weren't meant to work around corners and in a blind, shallow gap.
 

Attachments

·
Premium Member
Joined
·
177 Posts
Discussion Starter #9
CharlieVT said:
Here's what the feeler gauges I experimented with look like. Someone might try fussing with this method, but I can't get it to work reliably. Feeler gauges just weren't meant to work around corners and in a blind, shallow gap.
Agreed, I have trouble with feeler gauges when I have plenty of elbow room. Have to have some sense of the drag when moving the gauge to be accurate, and I don't do it enough to stay in practice.

Good points on the preload the crushing process could make. You wouldn't want to crush too much material at the same time. Perhaps just a couple of strands of solder (I'm thinking 1/16" or so) placed 180° apart would give results, but not impart too much preload.

Too hot in the garage to worry it now, perhaps I'll compare to the dial indicator method this winter and report back.

Thanks for the feedback.
 

·
Registered
Joined
·
637 Posts
It is hard to beat a dial indicator for quick and accurate measurements of small displacements. The dial test indicators are cheap (< $40) and regular 1 inch dial indicators are especially cheap (< $20) as everyone thinks they can't read metric motions. But WE know better..... :D )

It is the supporting fixture and technique that is important and not so easily achieved.
 

·
Premium Member
Joined
·
7,677 Posts
Yes, there is sort of an art to measuring using the dial indicator method. The drives don't all "behave" the same way due to differences in tolerances. Also, drives that have failed often have changes in the bearing seat of the FD cover due to the race having spun.

When doing Dman's dial indicator technique, I heat the cover to 250degrees F and assemble the drive. I put a film of grease on the crownwheel bearing seat in the cover to help things move. I repeat measurements as the cover cools. At first the measurements well be a little on the big side, then as the cover cools the measurements settle down and become repeatable. Then as the cover cools further, the bearing outer race will no longer slide in the cover and measurements are not longer meaningful. Not all drives are the same. Some are easy to get measurments from even with the cover relatively cool, others the bearings doesn't want to move even with the cover close to 250 degrees. This is a function of the variation between covers and bearing in the interference fit between them.

Way back someone suggested a feeler gauge approach. I and others dismissed it as unreliable. I have experimented with it but cannot get the results using feeler gauges to match up with the other two methonds I use.

Here's what the feeler gauges I experimented with look like. Someone might try fussing with this method, but I can't get it to work reliably. Feeler gauges just weren't meant to work around corners and in a blind, shallow gap.
On the drives you can measure "cold", how much does the reading vary between room temp and 250?
 

·
Registered
Joined
·
17 Posts
CharlieVT; snipped......... Way back someone suggested a feeler gauge approach. I and others dismissed it as unreliable. I have experimented with it but cannot get the results using feeler gauges to match up with the other two methonds I use. Here's what the feeler gauges I experimented with look like. Someone might try fussing with this method said:
When I did my final drive a year ago or so , after i checked with a dial indicator , out of curiosity i checked with some wire feeler gauges ( that came bent at a 90 degree angle) that i had. came within .0005 of the dial indicator, which i chalked up to feel. Haven't done enough to compare or have any worthwhile data. Like I said, just was curious !!
 

·
Registered
Joined
·
63 Posts
The problem with a dial indicator-derived reading is the friction of the OD of the bearing outer race in its housing may affect clamp up motions. Heating the housing increases the housing to bearing clearance about .001 inch per inch (diameter) for 160 degF temp rise (that's a good approximation to remember for any aluminum vs steel expansion). But if there's not to much bearing OD vs housing ID friction it probably is the best scheme we have. The same expansion ratio will hold true for the short axial length of the force/clamp-up path.

It also works for metric providing the temperature delta is given in Fahrenheit. Or furlongs per furlong for 160 degF etc., whatever.............
 

·
Registered
Joined
·
4,426 Posts
The problem with a dial indicator-derived reading is the friction of the OD of the bearing outer race in its housing may affect clamp up motions. Heating the housing increases the housing to bearing clearance about .001 inch per inch (diameter) for 160 degF temp rise (that's a good approximation to remember for any aluminum vs steel expansion). But if there's not to much bearing OD vs housing ID friction it probably is the best scheme we have. The same expansion ratio will hold true for the short axial length of the force/clamp-up path.

It also works for metric providing the temperature delta is given in Fahrenheit. Or furlongs per furlong for 160 degF etc., whatever.............
The OD of the bearing and the ID of the cover are not the issue you need to worry about. It is the thickness of the cover change from cold to hot and if that maintains the same expansion rate of .001 per inch per 160 degF, then the cover depth being measured being only 3/4 of an inch would be off by no more than .001 compared to when cold so that is an acceptable margin of error seeing that the over shimming is typically in the .004 to .008 inch range. If you shoot for the center of the preload range, you would likely still be " in spec" even with that error in measurements. After that, it is all about how well did you do the measurements as to whether you stay " in spec" or not.
 
1 - 14 of 14 Posts
Top