Full float rearends

[Silver69Camaro]

Originally Posted by chicane

Don't need... and did not need to re-read your post. I read your comment's correctly the first time.

My reply was to educate the rest of the forum members that don't know any better and/or that do not understand shaft specifics. Even comments in generalization can mis-lead or mis-represent the point trying to be conveyed.

Gotcha. It appeared the contents of your message were directed towards me.

[chicane]

Oh... no no no... its all good Matt.

You are right... it does look as I was specifically drawing the point towards you individually. That wasnt the intention... and I realize that my general sarcasm isnt known to all amoungst the masses.

[ccracin]

See my comments or questions within the quote, it's easier. Thanks for taking the time as well.

Originally Posted by chicane

Instead of falling suspect to any speculation I have called upon our resident bearing specialist... CarlC to address this subject. Hopefully he will reply shortly.

I agree, I am not an expert either.

CC- Yes... that term being used is rather generic. But... Ill take a stab at a simple explaination as I under stand it.

So, with the SF rear set-up do the bearings have a pre-load in this case? The pre-load being a torque value applied to the fasteners of the bearing retainer after all clearance is taken up?

Bearings with a tapered geometry, can be difficult to directly measure radial clearances. For this reason, radial clearances are commonly converted to axial clearances. (Radial clearance equals axial clearance times the cotangent of the bearing contact angle.)

Agreed, this is exactly why the application specific testing was done to develop the clearance matrix in our application.

The amount of clearance between bearing rollers and raceways measured in this axial direction is called endplay. The optimal initial endplay depends on a number of factors, including shaft and housing materials, bearing fits, and operating temperatures.

With this, endplay adjustments in tapered roller bearings are made during assembly or installation to compensate for the expected thermal expansion during operation and system deflections. Axial preload in tapered roller bearings is produced by displacing one bearing ring axially in relation to the other by an amount corresponding to the desired preload force. Generally, each bearing is individually and manually adjusted using shims or spacer sleeves, or by tightening torques from means likend to the bearing retainer.



In addition to manual methods, there are also several automated techniques. In the case of double-row tapered roller bearings, a preset assembly can be supplied, or the bearings can be adjusted manually at assembly by machining spacers.

Agreed, pre-machined spacers work well.

Careful monitoring and observation of applications can sometimes reveal early warning signs of clearance and endplay-related problems. Vibration monitoring equipment, for example, can detect the excessive axial and radial shaft movements caused by too much endplay. Temperature sensors can detect-increases in heat associated with impending bearing failure. This may also be accomplished by reading the surface condition that may provide a picture of what the failure mode is... radially spalling, micro spalling, impact deformations etc etc.

Just in case your interested, SKF has a nice canned system for predicting bearing failure on the market. It's trade name is the "Copperhead" System

CarlC... take it away.

Thanks again, I agree with everything stated. (Not that you were looking for agreement. :shrug) I was very curious mostly about the pre-load vs. initial clearance idea. We never had good luck with pre-load. Then again all applications are different. I guess we should quit boring the rest of the board.

[ccracin]

Originally Posted by CarlC

Tom's on it.

Preload can be + or -. Often in the bearing industry, but mostly related to machine tool spindle bearings, clearance will be referred to as negative preload.

Due to the nature of the beast tapers are set using an axial endplay (bench endplay) value. On horizontal applications it can be a real bear to get the BEP set correctly since gravity is not your friend. On larger assemblies it is recommended to set the BEP in a vertical fixture. Trying to measure radially usually does not have good results due to trying to keep the system aligned, off-center housing mass loads, no access to both sides of the shaft, multi-piece rings, etc. It can be a very time consuming and arduous process.

Smaller TRB's like those used as discussed are great for the do-it-yourself'er since the preload is built in. There are very few self-contained TRB assemblies like these anywhere in the industrial field. I like them, and have proved very reliable for me.

Thanks for your reply as well CarlC. This is exactly why I raised my first question about the delta cup to cone being the only verification question. We prefered doing our bench checks vertically as you stated, but also as you stated it is not always possible. Thus, the need for the custom fixtures and testing. Things most backyard mechanics don't have the luxury of. Thanks for the info. Gotta love this stuff.

[CarlC]

Originally Posted by ccracin

We never had good luck with pre-load.

From a practical standpoint with field-matched parts, you never will. Some basic thermal expansion calcs + a bit more help get you close for an initial BEP. I'm working on one now for a Komatsu haul truck (the mega-sized variety) where the OE spec for BEP is absolutely stupid, about 4X more than what it should be, and what has proven NOT to work in the field. But proving it to the OE of the traction motor is another story. But, the OE gets $Millons$ every year for rebuilds.

Bearing life is maximized when a slight preload (micron level) is applied. The hard part is getting it set right, and then having equal shaft and housing thermal expansion so that the preload stays under control. The preload reduces the chances of rolling element slippage, where instead of true rolling motion the rolling element skids across the raceway. This is, usually, not much of a problem for us. But imagine the ACC/DEC on a 15K RPM 40-Taper machine tool spindle. 15K - 0 RPM in a second is tough to control from an inertia standpoint and requires multiple levels of preload, temperature, lubrication, and warmup control.

Preload, both negative and positive, requires clamping of the inner and outer rings. For traditional front axles using TRB's the cones are clamped via the adjustment nut, and the outer rings/cups the hub. For the rear axle bearings discussed, it requires the inner rings be pressed on together and clamped in place, often using another sweated on ring for retainment. The outer ring(s) are held in by a retainer plate. For the rear TRB's discussed here they are a clearance (+) design.

[chicane]

Originally Posted by ccracin

Just in case your interested, SKF has a nice canned system for predicting bearing failure on the market. It's trade name is the "Copperhead" System.

Always interested.

Yep, Copperhead. I am familiar with SKF's device's. A few years back I was involved in a study that utilized that device, as well as a bunch of other COTS. It actually turned into a full blown study on the interpretation of high speed vibration data. We also had a full compliment of Andre' vibration equipment... displacement, seismic, velocity transducers and an array of different accelerometers. Neat stuff.

Originally Posted by ccracin

So, with the SF rear set-up do the bearings have a pre-load in this case? The pre-load being a torque value applied to the fasteners of the bearing retainer after all clearance is taken up?

If the retainer is designed correctly... yes. Or if the bearing design is that like a set 20... yes. But, it also relies on the design of the retainer. If you take into account of the tolerance stacking and growth factors I am sure that it could hold 0.001-0.0005" with relative ease. The race diamention's are another factor... obviously. End gap aligment, axial loading and diamentional tolerance are also big players here too.

Bearing theory and extensive application experience show that a "line-to-line" setting (with no endplay or preload) or a slight preload at the operating temperature provides maximum performance and bearing life. The appropriate mounted bearing preload at normal operating temperatures depends on the bearing load. Under radial loading, a tapered bearing is subjected to both radial force and force in the axial direction.

Then again... were talking about street cars. Temperature fluctuations and an uncontrolled environemnt make it difficult to maintain this level of scrutiny.

[tyoneal]

[QUOTE=chicane;140528]Yes.

Chicane:

This is not mean as a challenge as much, as I am interested in learning more about this topic.

Could you elaborate further about whom you speak about, and what cars do they drive. Also, what are they doing to put that kind of stress on their cars?

You may be referring to yourself. Since I don't know who you are, I am interested in your background in an attempt to understand the boundaries where a full floater would be necessary, vs. being a fun extra to have on a car.

I look forward to your answer.

Regards,

Ty

[chicane]

PM sent.

[tyoneal]

Originally Posted by chicane

PM sent.

PM Sent.

[ccracin]

C'mon guys No Fair! LOL Anyway thanks Chicane and CarlC foor participating in this discussion. You guys have confirmed alot of what I was thinking and have given me more to think about. I hope some other people have benefited as well. Most have to spend thousands on school to get this type of info. Good Stuff.

I would also be interested in the type of car or cars you have Chicane. I did a search and didn't yield anything.

Later,