Mech Engineers: question ? - BMW Luxury Touring Community
 
LinkBack Thread Tools Display Modes
post #1 of 17 Old Nov 3rd, 2008, 7:13 pm Thread Starter
Senior Member
 
CharlieVT's Avatar
 
Join Date: Jun 2002
Location: The Deep South of Vermont
Posts: 4,395
Mech Engineers: question ?

This is gonna be fun.

When my FD rebuild video first became available on the site, I posted asking for suggestions and comments regarding the technique.
I just got a long and thoughtful msg from Les in AL.
Here's the jist of Les' observation:
A new crown wheel bearing has some axial play (slop, tolerance, whatever you want to call it), that can be large enough to be a significant precentage of the movement measured using the technique described in the FD rebuild video.

The effect of this axial play in the bearing is that the shim space is over-estimated by the amount of this axial play when using the technique described in my video. This can result in calculating a preload that is less that would be calculated by using the BMW technique described in the Service Manual.

Is this significant? Les thinks it can be and he has a well thought out description of why; (it does go on for about 1000 words). I am not an engineer and don't know the specs on these bearings but I understand his logic.

If you are one of our resident mech engineers, I'd like to forward Les' observations to you (I don't think he'll mind) for your review and comments.

The goal is to improve and refine the technique for those interested in doing a DIY final drive rebuild.

Thanks in advance.
CharlieVT is offline  
Sponsored Links
Advertisement
 
post #2 of 17 Old Nov 3rd, 2008, 8:54 pm
Wrencher Extraordinaire
 
jzeiler's Avatar
 
Join Date: May 2004
Location: Hazel Green, AL, USA
Posts: 13,497
Re: Mech Engineers: question ?

Curtis,


As I think about it your method is sound and takes into account the axial play. Just take the measurement while applying the lift. If you pull it to maximum travel then let it relax you will get the play error, but keeping pressure on the lift will show a proper measurement.

John
2009 K1300GT Red Rocket
2009 R1200GS (Gone)
2005 K1200LT Ocean Blue Blue Wizard 110 K and counting...
2006 Bushtec Turbo+2 Spell
2004 330 Ci Convertable
K4AN

Have ridden a Motorcycle in all 48
But lack DE, MA, RI and CT with the 2005 LT

To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.
jzeiler is offline  
post #3 of 17 Old Nov 3rd, 2008, 10:55 pm
Lifetime Supporter
 
PadG's Avatar
 
Join Date: Aug 2008
Location: Solon, OH, USA
Posts: 4,836
Re: Mech Engineers: question ?

Send it along Charlie! I will be more than happy to comment!

Pad. Gajajiva
Solon, OH, USA

2015 R1200RT (San Marino Blue Met.)
2014 R1200RT (Quartz Metallic Blue - Returned to BMW)
2007 R1200RT (Sold!)


Once Upon a Time........
1963 Norton Dominator 650 SS
1960 Triumph Bonneville (T120)
1960 Triumph Thunderbird (6T)
1952 Triumph Thunderbird (6T)
1932 Triumph 500
1952 BSA Goldstar
PadG is offline  
 
post #4 of 17 Old Nov 3rd, 2008, 11:05 pm
Lifetime Supporter
 
PadG's Avatar
 
Join Date: Aug 2008
Location: Solon, OH, USA
Posts: 4,836
Re: Mech Engineers: question ?

Quote:
Originally Posted by jzeiler
Curtis,


As I think about it your method is sound and takes into account the axial play. Just take the measurement while applying the lift. If you pull it to maximum travel then let it relax you will get the play error, but keeping pressure on the lift will show a proper measurement.
Well, if there are any axial play at all, what you say is not logical. When you lift the drive axle up (via the wheel mount flange), the play is taken up on the "up" side (let's call it that). Now, when you let the axle go to its resting position, the play is in the opposite side as the weight of the axle pull the inner race toward the "down" side. Therefore, any measurements that you take will include this so-called slop.

I have two questions that needs to be looked into. The first is if there is any play at all? If yes, then how much? I have a strong feeling that if the standard bearings do have axle play, then it would have been accounted for in the preload specifications. However, I feel rather strongly that there are zero play when the ball bearing is installed. However, I will reserve my comments until I have looked at all the data and dig into the facts!

Pad. Gajajiva
Solon, OH, USA

2015 R1200RT (San Marino Blue Met.)
2014 R1200RT (Quartz Metallic Blue - Returned to BMW)
2007 R1200RT (Sold!)


Once Upon a Time........
1963 Norton Dominator 650 SS
1960 Triumph Bonneville (T120)
1960 Triumph Thunderbird (6T)
1952 Triumph Thunderbird (6T)
1932 Triumph 500
1952 BSA Goldstar
PadG is offline  
post #5 of 17 Old Nov 4th, 2008, 5:23 am Thread Starter
Senior Member
 
CharlieVT's Avatar
 
Join Date: Jun 2002
Location: The Deep South of Vermont
Posts: 4,395
Re: Mech Engineers: question ?

Here is a segment from the pm I got from Les. (He told me he doesn't mind my sharing his comments, so they are.) He wrote me a long and very thoughtful msg and I think he has a point. Not being an engineer, I didn't consider that the axial play of the crownwheel bearing might be as great as he has measured.

"Most of the following you obviously know already, but I just wanted to make sure we're at least starting on the same page. To set up the final drive preload, we need to determine the space or gap between the Carrier Bearing (CB) outer race (OR) and the Cover. Once we know this dimension, we add the factory specified preload amount and this becomes the required shim pack thickness. My concern is that the method shown in the video does not take into account the axial clearance of the CB. I think this has been discussed before, but has not, to my knowledge, been completely addressed. I'm afraid the procedure shown in the video guarantees that the shim pack calculation will always be shy by an amount equal to the axial clearance of the CB. How so ? Here’s what I see happening. When we move the CB OR from the bottomed out (pushed down) position to the topped out (pulled up) position and visa versa, we are doing this though the CB inner race (IR) via the Hub to which it is solidly connected. The bottomed out or lowest position of the Hub/IR is established by the tapered roller bearing (TRB) at the opposite end of the Hub being seated in it’s race which itself is pressed into the housing. The topped out or up position is established when the CB OR is fully seated up against the Cover. Per the video, this movement is measured multiple times with a dial indicator placed against the Hub. The Hub is carefully moved down/up and indicator readings are taken until the measurements consistently repeat. The repeatable measurement is then used as a base line to which the specified preload amount is added to establish the shim pack thickness. Here is where I see a problem. Only when the cover is first installed is everything truly bottomed out. That is, the Hub/IR is bottomed because the TRB is pushed down against it’s outer race in the housing. The CB OR is bottomed out by having been pushed down as far as it can go through installation of the cover which is in direct contact with it. Basically: Cover pushes CB OR pushes CB IR pushes Hub pushes TRB pushes Housing. (pretty sure we're on the same page to this point) This "all bottomed" position is where we are zeroing the dial indicator and most importantly it is where the CB is in its' zero preload position. The distance from cover to the CB OR at this position is the critical or baseline dimension upon which the amount of preload must be established. This point is crucial to understanding what I see as wrong with the process in question. When we raise the Hub, the CB IR immediately moves up with it as does the CB OR (immediately this time only). This “assembly” continues to move up /outward until the CB OR tops out against the Cover. Let’s call this the initial measurement. So far so good, but the next step is where the measurement error creeps in. When we push the Hub back down for subsequent measurements, the CB IR moves immediately with the Hub, but the CB OR race will only begin to move down once the axial clearance of the CB is taken up in the downward direction (opposite of that when we first pushed the cover on). The CB OR downward movement will “lag” CB IR downward movement by an amount equal to the axial clearance of the CB. The Hub/IR will still bottom as before and the indicator will still return to zero, but the CB OR will not have not moved back to the exact position it was when we first installed the cover. Likewise, when we again pull/lift up the Hub/IR for another measurement, the CB OR will not begin to move up until all of the axial clearance is taken up in the upward direction. Remember, we set the initial bottomed (zero preload) position of the CB OR by pushing directly on it with the cover, but now we are moving the CB OR up/down indirectly via the CB IR. Consequently, this movement is now influenced by the CB axial clearance. The indicator will consistently track up/down movement of the hub and subsequent readings will repeat, but after the first up/down cycle, the CB OR movement will always be less than that of the Hub/IR movement. Consequently, the initial upward measurement is the only one that can possibly represent the correct baseline dimension between the Cover and the CB OR for our preload calculation. So how big a deal is this? As it turns out for a C3 class bearing as used here it can be a pretty big deal. A C3 class bearing of this size can have up to .0101” of axial clearance. What this means is that the inner and outer races can be moved relative to each other by up to .0101” along the axial centerline of the bearing and still be in spec. As an example, I have a brand new 17-ball carrier bearing which has an axial clearance of 0.008” (measured using a holding fixture on the inner race and multiple dial indicators on the outer race). This is a perfectly good bearing, yet it's axial clearance is 4x the total preload tolerance! If we used this bearing in the above process and are at the point of pushing the hub back down after the initial upward measurement, the Hub/IR would move downward .008" before the CB OR would even began to budge. When the Hub/IR bottoms, the indicator will properly return to zero, but the CB OR would now be .008” higher up in the cover than it was at the initial installation. In this case, the CB OR will always move .008" less than the Hub/IR. Unless we can push down directly on the CB OR, after the first up/down cycle it will always be .008" higher in the cover than it's initially bottomed position. Again, the initial measurement is the only one that represents the true distance between the cover and the CB OR when the CB is in it's zero preload position. All subsequent measurements will be shy by the amount of the CB's axial clearance. This is because we are moving the CB OR via the CB IR which are "connected" to each other by, in this case, .008" of "slop". We are measuring off of the Hub/IR, but what we really want is the true CB OR movement. On the video the initial measurement was mentioned as being around .015”, with subsequent measurements at .0135, .0130,.0130. Those measured values indicate that the bearing being installed had an axial clearance of about .002. This means the calculated shim pack for that drive was .002” too thin. While there’s certainly no danger of too much preload using this method, this amount of error could easily lead to selecting a shim pack which results in insufficient or possibly no preload. Either case is not good. The axial clearance must be taken into account if proper preload to ever be accurately established. If we know the axial clearance of the CB prior to installation, then the method shown in the video would probably work quit well. Although the method shown in the shop manual may be a bit difficult to tool (and repeat), it by nature avoids any measurement errors caused by the CB’s axial clearance."

Here's a comment from another mech engineer (Jim) regarding Les' observations on axial play: "I can definitely say that he is correct about axial clearance in this C3 fit (ie loose) 61917 deep groove ball bearing. The SKF specs state this bearing has a RADIAL clearance of 30-58 microns. The specs dont state the AXIAL clearance, but this can be 8-10 TIMES the radial clearance. A good description here: http://www.gmn.de/front_content.php...idart=37&lang=2

So - lets assume worst case of 10X. 30 microns times 10 = 300 microns = 0.0118". 58 times 10 = 580 microns = 0.0228"

Thus Les's measurements for axial clearance are in the ballpark for this bearing."
CharlieVT is offline  
post #6 of 17 Old Nov 4th, 2008, 5:38 am
Lifetime Supporter
 
DanMartin's Avatar
 
Join Date: Jun 2002
Location: Fulshear, TX, USA
Posts: 2,082
Re: Mech Engineers: question ?

That is good. My eyes started glazing over halfway through. Read it through several times though. Well written with good points.
Thanks for sharing.


dan
DanMartin is online now  
post #7 of 17 Old Nov 4th, 2008, 8:24 am
Lifetime Supporter
 
cfell's Avatar
 
Join Date: Apr 2004
Location: Round Rock, TX, USA
Posts: 7,794
Re: Mech Engineers: question ?

Yeah.. what Les said.. .. really.

I'm thinking this is really a good approach for initial setting of the "preload".

From Les' clinically solid observations I agree on the "distance" measurements. It doesn't verify the final "preload rotational force", the real reason for the "measured preload".

Has this been "considered" previously?

If the answer is "yes", please quit reading.





What this all assumes is correct preload tension... which is actually provided by the "cover" against the preload shims, transferring through the assembly to the FD housing.

When rebuilding "automobile differentials" I've done it using the "shim measurement" method .... with marginal results. What really worked was checking the 'preload rotational pressure". This means, set preload, then check how much it took to start the assembly to turn.... usually in "inch" pound torque.

Now, add to this "expansion" due to heat and you will either "tighten" or "loosen" the required preload.

All this really "assumes" proper case strength and expansion/contraction based upon temperature.

Axial and torsional loading will cause additional "flex" in the cover which will change pressure on the bearing set preload.

Based upon average running temps I've seen, 100 degree swings are possible from "cold" of a Texas morning to "operating temp".

We also need to wonder about the forces on the TRB which could "push" on the cover. If the cover is of insufficient strength, or damaged by bearing failure stresses, it is conceivable this would have varying degrees of change added or subtracted from the 'preload'.

Taking into account this information, I can only surmise (another word for ass-u-me).... the BMW Engineers took all this into account... right?

Now, what are we to do?

Buy a brand new FD each time? No guarantee the FD case strength is sufficient to maintain the "preload".

Now, lemme ask this...

What is the rotational preload measured at the rear wheel of YOUR bikes?

Anyone wanna go check?
Anyone wanna check when cold/operating temp?

Yes, you'll have additional "preload" rotational load of the driveshaft and universal joints in the driveshaft... but there SHOULD be an average with some "tolerance" amount....

Loose or tight would indicate higher potential for failure, I "assume" =)

Next time someone "rebuilds" their "FD", it would be interesting to test the "Preload" at first assembly of the FD, then check when installed on the bike... then you'll have a "baseline" for your machine.... then check for variances occassionally.

I'll wager they vary quite a bit, even on new bikes.

Ok, I'll stop there because the pain meds are beginning to engage...

...............
J.M.J...
Dcn Channing

To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.
cfell is offline  
post #8 of 17 Old Nov 4th, 2008, 8:27 am
Senior Member
 
bigbear's Avatar
 
Join Date: Dec 2005
Location: Winnebago, IL, USA
Posts: 1,178
Re: Mech Engineers: question ?

Charlie,
first off I am NOT a degreed mechanical engineer.

My thoughts on this procedure are that as long as you are measuring the "Hub" movement tha calculations are correct for amount of shims required.

Reasoning:
Since the Crown Bearing Outer Race (CB OR) is NOT the stopping point, the Tapered roller bearing is, when the hub assembly is pushed in. This negates or in esence eliminates the CB Axial free play from the equation.


IF the CB OR was hanging up or stopping the hub before the Tapered roller bearing seats then the CB Axial free play "would" be an issue.

Roy

Roy Gregersen

Ride Slow, Ride Fast, Always Ride Safe
85 K100RT sold
02 LTC DOA 9/21/14
12 R1200RT
bigbear is offline  
post #9 of 17 Old Nov 4th, 2008, 3:02 pm
LGW
Junior Member
 
Join Date: Aug 2002
Location: Huntsville, AL, USA
Posts: 27
Re: Mech Engineers: question ?

Quote:
Originally Posted by bigbear
Charlie,
first off I am NOT a degreed mechanical engineer.

My thoughts on this procedure are that as long as you are measuring the "Hub" movement tha calculations are correct for amount of shims required.

Reasoning:
Since the Crown Bearing Outer Race (CB OR) is NOT the stopping point, the Tapered roller bearing is, when the hub assembly is pushed in. This negates or in esence eliminates the CB Axial free play from the equation.


IF the CB OR was hanging up or stopping the hub before the Tapered roller bearing seats then the CB Axial free play "would" be an issue.

Roy
Roy,

Yes, the Hub/CB IR stops movement in the downward direction when the Tapered bearing seats. However, it is the CB OR that stops movement in the upward direction when it seats against the cover. This is where the error takes place because the CB races "mis-track" each other by the amount of the CB axial play. If we could cycle through our measurement process by lifting up on the Hub/CB IR and then pushing down on the CB OR to reseat everything there would not be an issue and all readings, including the first one, would repeat. The big difference between the first measurement and the subsequent measurements in the original process is telling us the initial conditions (positions) have changed. The CB OR is no longer in it's initial position after the first lift up/push down cycle. And that difference is the axial clearance of the bearing.

Regards,
Les Willey

Last edited by LGW; Nov 4th, 2008 at 3:38 pm.
LGW is offline  
post #10 of 17 Old Nov 4th, 2008, 4:30 pm Thread Starter
Senior Member
 
CharlieVT's Avatar
 
Join Date: Jun 2002
Location: The Deep South of Vermont
Posts: 4,395
Re: Mech Engineers: question ?

Quote:
Originally Posted by bigbear
Charlie,
first off I am NOT a degreed mechanical engineer.

My thoughts on this procedure are that as long as you are measuring the "Hub" movement tha calculations are correct for amount of shims required.

Reasoning:
Since the Crown Bearing Outer Race (CB OR) is NOT the stopping point, the Tapered roller bearing is, when the hub assembly is pushed in. This negates or in esence eliminates the CB Axial free play from the equation.


IF the CB OR was hanging up or stopping the hub before the Tapered roller bearing seats then the CB Axial free play "would" be an issue.

Roy
Did I say this was going to be fun?
After a good night's rest and a cup of coffee, I'm going to analyze this again.

At first I thought Les was right, but now I think Ray has it.

Having gone round and round visualizing Les' concept, and Roy's comment that "....Crown Bearing Outer Race (CB OR) is NOT the stopping point, the Tapered roller bearing is, when the hub assembly is pushed in...." makes sense to me.

In the technique shown in the video, when the hub is in the "down" position the taper roller bearing is seated. Where the outer race of the crown bearing is with respect to the inner race doesn't matter; this is the "zero" position with the hub firmly pushed or tapped down into the housing. With the cover heated, the outer race will slide down in its seat in the cover and never meet a stopping point. The hub assembly as a whole comes to a stop when the taper bearing is seated.

When the hub is pulled up with leverage (tire irons were used in the video) the crownwheel assembly pushes on the crownbearing inner race in an upward direction, the inner race presses the balls against the outer race, which then has a limit of travel when it hits the bottom of the seat in the final drive cover. When the hub is held in the "up" position the crown bearing is compressed in the same direction that it will be when shims are installed. That is to say, the axial play in the crown bearing is not interfering with the measurement, as the bearing is being loaded with leverage applied by the tire irons in the same direction of force that will be applied by the shims.

Since we are measuring the extent of travel of the hub from its most down position with the taper bearing seated in its race, to the hub's most up position with the crown bearing firmly seated in the cover, and the axial play taken out of the crown bearing with pressures in the same direction that will be applied by the shims, we are measuring the shim space accurately. When the specified preload is added to that space we have the needed shim thickness.

If we were taking our measurement in the down position from the outer race of the crown bearing, Les' observation would be absolutely correct. But since we are measuring the hub, the amount of axial play doesn't matter. The bearing's dimension is only measured when it is loaded in the same direction that the shims will load it.

I think the initial larger measurement shown in the video which changed with subsequent measurements I believe was a result of the dial indicator not being properly zeroed or having been jarred slightly. If I had been more careful in setting up the dial indictor for the first measurement, that difference would not have been seen.

My next project is to make a BMW "special tool" out of a spare final drive cover I have and compare the BMW service manual technique for calculating preload with the technique shown in the video.

Anyone have a final drive they want rebuilt? It will be rebuilt having been measured for preload every possible way I can.

Thanks to Les and everyone for their comments, theories, and insights. My goal is to make the FD rebuild technique as simple and reliable as possible so that anyone considering a DIY job has the best info this collective group can put forth.

I just got my copy of November BMWON and have started the FD article. I did read the FD article that was in MCN (THANKS GRIF!) and it looks very much like the same article, but I will read it again with interest.

PS I had a nice ride on the LT this afternoon. It is getting close to the end of riding season here in the Deep South of Vermont, some of the roads had been salted because of recent snow. I never think about my FD when riding the LT, I only think about final drives when reading and posting here about them.
CharlieVT is offline  
post #11 of 17 Old Nov 4th, 2008, 5:22 pm
Lifetime Supporter
 
PadG's Avatar
 
Join Date: Aug 2008
Location: Solon, OH, USA
Posts: 4,836
Re: Mech Engineers: question ?

Charlie,

Wow.....that's hard reading, with everything in a single paragraph!!

The major issue that I have with Les' analysis is the fact that he had based the axial movement of the inner race to outer race of an "unloaded" bearing, that is one straight out of the box and unassembled! You can't do that. If you do, then you will see lots of movements! The missing factor that was not taken into account is the preload on that bearing when it is installed! No, I am NOT talking about the axial preload on the taper roller bearing with those shims that you have been working with.

It needs to be understood that every "rolling type" bearings needs to have pre-load applied, and the amount is usually specified by the manufacturers. That ball bearing is pre-loaded by having interference fits between the inner race and the drive shaft in addition to the interference fit between the outer race and the housing!

What the radial pre-load does is to eliminate the gaps that exist between the balls and the races. This is necessary to make the bearing works properly. If you guys are interested, I can get into more details regarding radial pre-loads of these bearings and how it effect the performance.

Incidentally, do NOT think that a grade C3 bearing is less precise than grade C2, say! The reason why, I, as a designer, would pick grade C2 or CN or C3 or even C4 is NOT for level of precision, but for the amount of interference fits that I want to have when I assemble the bearing! How much interference I want will be dependent on many factors, such as shaft and housing materials, expected thermal conditions, and so on. The key thing though is that when the bearing, regardless of grade, is assembled, it will have the amount of radial pre-load that is specified by the manufacturer!

Pad. Gajajiva
Solon, OH, USA

2015 R1200RT (San Marino Blue Met.)
2014 R1200RT (Quartz Metallic Blue - Returned to BMW)
2007 R1200RT (Sold!)


Once Upon a Time........
1963 Norton Dominator 650 SS
1960 Triumph Bonneville (T120)
1960 Triumph Thunderbird (6T)
1952 Triumph Thunderbird (6T)
1932 Triumph 500
1952 BSA Goldstar
PadG is offline  
post #12 of 17 Old Nov 4th, 2008, 5:54 pm
Senior Member
 
JATownsend's Avatar
 
Join Date: Oct 2003
Location: Bloomfield Hills, MI, USA
Posts: 1,134
Re: Mech Engineers: question ?

Quote:
Originally Posted by CharlieVT
...Anyone have a final drive they want rebuilt? It will be rebuilt having been measured for preload every possible way I can.
Charlie,

I recently (post CCR '08) acquired a very nice '05 FD. I have it squirreled away as a BU...ready to ship! The lack of LT FDs is this country is nutz.

Its a low mileage unit in nice shape. I've not opened it. If you're interested, let's talk.

J. Averill Townsend
Bloomfield Hills, MI


IBA# 24374

2002 K1200LTC - Silver
1978 R100/7 - Very, very Black
2004 Bushtec Quantum - Silver, of course...(SOLD)

To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.
JATownsend is offline  
post #13 of 17 Old Nov 4th, 2008, 5:57 pm
LGW
Junior Member
 
Join Date: Aug 2002
Location: Huntsville, AL, USA
Posts: 27
Re: Mech Engineers: question ?

Curtis / Ray,

After having made several scaled drawings at various stages of the process to help me visualize what and Ray you are saying, I can see where you guys are coming from and I have to admit that it makes sense. I hope this hasn't caused any undue concern (or eyestrain) for anyone ... I'll go back to sleep now

Take care,
Les


Quote:
Originally Posted by CharlieVT
Did I say this was going to be fun?
After a good night's rest and a cup of coffee, I'm going to analyze this again.

At first I thought Les was right, but now I think Ray has it.

Having gone round and round visualizing Les' concept, and Roy's comment that "....Crown Bearing Outer Race (CB OR) is NOT the stopping point, the Tapered roller bearing is, when the hub assembly is pushed in...." makes sense to me.

In the technique shown in the video, when the hub is in the "down" position the taper roller bearing is seated. Where the outer race of the crown bearing is with respect to the inner race doesn't matter; this is the "zero" position with the hub firmly pushed or tapped down into the housing. With the cover heated, the outer race will slide down in its seat in the cover and never meet a stopping point. The hub assembly as a whole comes to a stop when the taper bearing is seated.

When the hub is pulled up with leverage (tire irons were used in the video) the crownwheel assembly pushes on the crownbearing inner race in an upward direction, the inner race presses the balls against the outer race, which then has a limit of travel when it hits the bottom of the seat in the final drive cover. When the hub is held in the "up" position the crown bearing is compressed in the same direction that it will be when shims are installed. That is to say, the axial play in the crown bearing is not interfering with the measurement, as the bearing is being loaded with leverage applied by the tire irons in the same direction of force that will be applied by the shims.

Since we are measuring the extent of travel of the hub from its most down position with the taper bearing seated in its race, to the hub's most up position with the crown bearing firmly seated in the cover, and the axial play taken out of the crown bearing with pressures in the same direction that will be applied by the shims, we are measuring the shim space accurately. When the specified preload is added to that space we have the needed shim thickness.

If we were taking our measurement in the down position from the outer race of the crown bearing, Les' observation would be absolutely correct. But since we are measuring the hub, the amount of axial play doesn't matter. The bearing's dimension is only measured when it is loaded in the same direction that the shims will load it.

I think the initial larger measurement shown in the video which changed with subsequent measurements I believe was a result of the dial indicator not being properly zeroed or having been jarred slightly. If I had been more careful in setting up the dial indictor for the first measurement, that difference would not have been seen.

My next project is to make a BMW "special tool" out of a spare final drive cover I have and compare the BMW service manual technique for calculating preload with the technique shown in the video.

Anyone have a final drive they want rebuilt? It will be rebuilt having been measured for preload every possible way I can.

Thanks to Les and everyone for their comments, theories, and insights. My goal is to make the FD rebuild technique as simple and reliable as possible so that anyone considering a DIY job has the best info this collective group can put forth.

I just got my copy of November BMWON and have started the FD article. I did read the FD article that was in MCN (THANKS GRIF!) and it looks very much like the same article, but I will read it again with interest.

PS I had a nice ride on the LT this afternoon. It is getting close to the end of riding season here in the Deep South of Vermont, some of the roads had been salted because of recent snow. I never think about my FD when riding the LT, I only think about final drives when reading and posting here about them.
LGW is offline  
post #14 of 17 Old Nov 4th, 2008, 6:14 pm Thread Starter
Senior Member
 
CharlieVT's Avatar
 
Join Date: Jun 2002
Location: The Deep South of Vermont
Posts: 4,395
Re: Mech Engineers: question ?

Quote:
Originally Posted by LGW
Curtis / Ray,

After having made several scaled drawings at various stages of the process to help me visualize what and Ray you are saying, I can see where you guys are coming from and I have to admit that it makes sense. I hope this hasn't caused any undue concern (or eyestrain) for anyone ... I'll go back to sleep now

Take care,
Les
Les,
Your original PM to me was thoughtful and I much appreciated getting it. I have harbored doubts about the FD rebuild technique, wondering what limitations, errors there might be. Your inquiry caused a questioning of premises and deeper analysis of what we were doing. I brought my experice as a shade tree mechanic but everything I think I understand about the final drive I got from this group. I am relieved to think that I haven't led anyone far astray with the video, but knowledge is always a work in progress and I consider only that you have made a contribution to that end.
Thanks, and warm regards from the deep south of Vermont
CharlieVT is offline  
post #15 of 17 Old Nov 4th, 2008, 6:37 pm
LGW
Junior Member
 
Join Date: Aug 2002
Location: Huntsville, AL, USA
Posts: 27
Re: Mech Engineers: question ?

Thanks Curtis. You're absolutely right. Knowledge truly is a work in progress and right or wrong there's always opportunity for learning. AMHIK Hopefully in the future I can give back to this great forum something truly helpful.

Ya'll take care
Les

Quote:
Originally Posted by CharlieVT
Les,
Your original PM to me was thoughtful and I much appreciated getting it. I have harbored doubts about the FD rebuild technique, wondering what limitations, errors there might be. Your inquiry caused a questioning of premises and deeper analysis of what we were doing. I brought my experice as a shade tree mechanic but everything I think I understand about the final drive I got from this group. I am relieved to think that I haven't led anyone far astray with the video, but knowledge is always a work in progress and I consider only that you have made a contribution to that end.
Thanks, and warm regards from the deep south of Vermont
LGW is offline  
post #16 of 17 Old Nov 4th, 2008, 7:07 pm
LAF
Lifetime Supporter
 
LAF's Avatar
 
Join Date: Jul 2007
Location: Harrisburg , PA, USA
Posts: 3,041
Re: Mech Engineers: question ?

Just an amazing exchange of thoughts!

If BMW could have you guys for week we would be good to go!

I thought I read that many felt after installing these bearings in FD rebuilds, that installing them at the factory was a thought on FD failures?

And I also thought I have read that some have found the hub may be out of spec?

So if a bearing does needs an amount of preload on the cages, and races, that would support that theory very well, right?

I mean I have used heat and cold to put cam bearings on cams in Harley's before many times and it goes like butter.

How is BMW installing them?

And if a hub is out, the bearing is not getting the correct preload weather it is loose or tight, right?

How would a layman measure the tolerances of a bearing race and retaining ring, going over a hub? Would it be measured by pressure of what it took to spin the bearing? Would the race contort enough to measure? The ball bearings themselves?

I have nothing but questions trying to comprehend all this.

Lee
17.5 R1200 GSW Black Low
15 R1200 RTLC San Marino Blue Metallic (Sold)
10 Liquid Silver FJR1300 (Sold)
O7 Biarritz Blue Metallic LT (Totaled 2010)
ATGATT I am breathing proof.
BMW MOA #135959
LAF is offline  
post #17 of 17 Old Nov 4th, 2008, 8:15 pm
Senior Member
 
bigbear's Avatar
 
Join Date: Dec 2005
Location: Winnebago, IL, USA
Posts: 1,178
Re: Mech Engineers: question ?

You would have to measure the hub where the Crown bearing inner race presses on. This measurement should be taken with a micrometer that would read accurately to 0.0001" or 1/ 10,000th of an inch. Then you would have to find what the bearing manufacturer considers the correct diameter should be.

Same thing for the bearing ID.

Not exactly sure how much interference is normal for this size bearing.

Roy

Roy Gregersen

Ride Slow, Ride Fast, Always Ride Safe
85 K100RT sold
02 LTC DOA 9/21/14
12 R1200RT
bigbear is offline  
Reply

Quick Reply
Message:
Options

Register Now



In order to be able to post messages on the BMW Luxury Touring Community forums, you must first register.
Please enter your desired user name, your email address and other required details in the form below.

User Name:
Password
Please enter a password for your user account. Note that passwords are case-sensitive.

Password:


Confirm Password:
Email Address
Please enter a valid email address for yourself.

Email Address:
OR

Log-in











Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page
Display Modes
Linear Mode Linear Mode



Similar Threads
Thread Thread Starter Forum Replies Last Post
Engineers Lynn_Keen Humor 12 Mar 2nd, 2008 8:15 pm
Why German Engineers are not on my Chirstmas List tlash Northern California 6 Oct 6th, 2007 8:49 pm
Understanding Engineers mconlogue Chit Chat 6 Jul 22nd, 2006 7:11 am
Understanding Engineers mconlogue Humor 0 Mar 13th, 2006 1:45 pm

Posting Rules  
You may post new threads
You may post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On

 
For the best viewing experience please update your browser to Google Chrome