Go Back  Bike Forums > Bike Forums > Bicycle Mechanics
Reload this Page >

What exactly will happen if your hubs have a little play?

Notices
Bicycle Mechanics Broken bottom bracket? Tacoed wheel? If you're having problems with your bicycle, or just need help fixing a flat, drop in here for the latest on bicycle mechanics & bicycle maintenance.

What exactly will happen if your hubs have a little play?

Old 12-05-17, 02:14 PM
  #26  
Racing Dan
Senior Member
 
Racing Dan's Avatar
 
Join Date: Aug 2015
Posts: 2,224
Mentioned: 9 Post(s)
Tagged: 0 Thread(s)
Quoted: 1332 Post(s)
Liked 316 Times in 214 Posts
Originally Posted by Andrew R Stewart
I think what Racing Dan is talking about is when there's excessive (and there's that qualifier once again. Our passion is full of judgments and opinions) bearing preload the rolling surfaces go through a work hardening process as they are indented and rebound under the pressure of the balls. I think of it like a too heavy truck for the road's surface. The layers of road will flex and then in time crack.


Not every situation has had scientific research and a published paper. But the decades of hands on findings do speak just the same. Andy
I mean preload in the sense the bearing is loaded, to some degree, even if no external load is applied (no one is riding the bike). Bearing preload is the opposite of bearing clearance.
Racing Dan is offline  
Old 12-05-17, 03:52 PM
  #27  
San Rensho 
Senior Member
 
Join Date: Oct 2005
Posts: 5,820
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 383 Post(s)
Liked 133 Times in 91 Posts
With a properly adjusted bearing, which has absolutely no play in it, then all of the ball bearings always make contact with the Cup and Cone, evenly Distributing all of the forces and weight over the entire assembly. Back in the day when I used to race, many many years ago, people were obsessive about resistance, and ran their beautiful campy hubs with play in them. Within a couple of months, either the cup or the cone was ruined in one section because all of the forces were concentrated on one or two bearings, instead of being evenly distributed over all of them.

So do not adjust hubs so that they have play in them.
San Rensho is offline  
Old 12-05-17, 04:03 PM
  #28  
Andrew R Stewart 
Senior Member
 
Andrew R Stewart's Avatar
 
Join Date: Feb 2012
Location: Rochester, NY
Posts: 18,003

Bikes: Stewart S&S coupled sport tourer, Stewart Sunday light, Stewart Commuting, Stewart Touring, Co Motion Tandem, Stewart 3-Spd, Stewart Track, Fuji Finest, Mongoose Tomac ATB, GT Bravado ATB, JCP Folder, Stewart 650B ATB

Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 4172 Post(s)
Liked 3,792 Times in 2,271 Posts
Originally Posted by 3alarmer
...ball bearings and how they work have been thoroughly researched. Cone and race surfaces are case hardened, at the very least, in a quality bearing assembly. I honestly do not see how what you have described can occur.

Certainly excessive bearing load (to the point where the wheel no longer spins freely under the weight of the valve stem) ought to be avoided. But like @davidad, Mr Brandt, and a lot of other people I have preloaded both headset and wheel bearings on bicycles for many years now, with no noticeable ill effects.

But I'm a test subject group of one, which is why I asked for a source.

Bingo! Because this well adjusted hub is not the usual for the vast majority of bikes sold, at the time of sale. Just about every bike I have pulled out of a box, and had cup and cone hub bearings, had bearings too tight to achieve this standard. The time to adjust the hubs as well as their grade can be isn't time that makes the shop any money, so most shop assemblies I see have really tight bearings. There's so much leverage at the tire over the bearing, that when the bearings are new the preload is hard to notice. Unless the wheel is removed from the bike this preload isn't likely to be discovered until the bearings start to wear badly.


The preload thing is a shade of gray. A tiny amount is good, much more gets bad fast. (and I draw a difference between radial and angular contact WRT radial's sometimes need for end play). Andy
Andrew R Stewart is offline  
Old 12-05-17, 04:14 PM
  #29  
Racing Dan
Senior Member
 
Racing Dan's Avatar
 
Join Date: Aug 2015
Posts: 2,224
Mentioned: 9 Post(s)
Tagged: 0 Thread(s)
Quoted: 1332 Post(s)
Liked 316 Times in 214 Posts
Originally Posted by Andrew R Stewart
Bingo! Because this well adjusted hub is not the usual for the vast majority of bikes sold, at the time of sale. Just about every bike I have pulled out of a box, and had cup and cone hub bearings, had bearings too tight to achieve this standard. The time to adjust the hubs as well as their grade can be isn't time that makes the shop any money, so most shop assemblies I see have really tight bearings. There's so much leverage at the tire over the bearing, that when the bearings are new the preload is hard to notice. Unless the wheel is removed from the bike this preload isn't likely to be discovered until the bearings start to wear badly.


The preload thing is a shade of gray. A tiny amount is good, much more gets bad fast. (and I draw a difference between radial and angular contact WRT radial's sometimes need for end play). Andy
The leverage issue is real. If you adjust a cup and cone wheel to have a tiny bit of play at the rim, when mounted, it will often feel somewhat tight if you rotate the axle, between thumb and index, holding the rim.
Racing Dan is offline  
Old 12-05-17, 07:42 PM
  #30  
3alarmer 
Friendship is Magic
 
3alarmer's Avatar
 
Join Date: Nov 2010
Location: Sacramento, CA
Posts: 22,932

Bikes: old ones

Mentioned: 304 Post(s)
Tagged: 0 Thread(s)
Quoted: 26249 Post(s)
Liked 10,231 Times in 7,100 Posts
Originally Posted by Racing Dan
A wheel that spins freely under the weight of the valve is not preloaded.
...that's an interesting observation. Again, what's your source ?
3alarmer is online now  
Old 12-05-17, 08:50 PM
  #31  
3alarmer 
Friendship is Magic
 
3alarmer's Avatar
 
Join Date: Nov 2010
Location: Sacramento, CA
Posts: 22,932

Bikes: old ones

Mentioned: 304 Post(s)
Tagged: 0 Thread(s)
Quoted: 26249 Post(s)
Liked 10,231 Times in 7,100 Posts
One of the final steps in the bearing manufacturing process is the assembly of the individual bearing components: the outer ring, inner ring, balls and retainer (or ball separator). When the bearings are assembled, it is necessary to have a controlled amount of internal clearance, or looseness between the rings and balls. This is referred to as radial play in most bearing catalogs.

In certain applications, this internal clearance must be removed for a pair of bearings to operate properly. The application of an axial load across a pair of bearings – for the purpose of removing free internal clearances – is called preload.

Benefits of preloading ball bearings include:

Rotational accuracy and precise shaft positioning
Elimination or reduction of ball skidding
Control and reduction of axial and radial deflection under applied load
Noise reduction
Load sharing between bearings
Axial adjustment requires great care, accuracy, proper tooling and cleanliness during the assembly process to avoid excessive preloading and ensure correct conditions. First, two bearings are mounted with the desired axial offset. The shaft and/or housing are threaded. Finally, the internal clearance in the bearings is removed by installing precision ground washers and then tightening the nut or threaded collar.
How much preload should be applied?

In general, provided the design requirements are met, the least amount of preload is desired
...all from Preloading Ball Bearings: Full AST Guide - Bearings : Blog : AST Bearings
3alarmer is online now  
Old 12-05-17, 09:57 PM
  #32  
SquidPuppet
Calamari Marionette Ph.D
 
SquidPuppet's Avatar
 
Join Date: Dec 2013
Location: Coeur d' Alene
Posts: 7,862

Bikes: 3 Chinese Gas Pipe Nerdcycles and 2 Chicago Electroforged Boat Anchors

Mentioned: 75 Post(s)
Tagged: 0 Thread(s)
Quoted: 2358 Post(s)
Liked 33 Times in 26 Posts
Originally Posted by Racing Dan
A wheel that spins freely under the weight of the valve is not preloaded.



My wheel bearings have zero play, NONE, and are therefore preloaded. They are not tight, which would be excessive preload, resulting in an interference fit.

Once spun, my wheels will eventually "pendulum" for a few cycles before they finally come to rest.


Excessive preload can cause increased heat, fatigue, and torque. Insufficient preload can allow resonant vibration causing fretting of the raceways.

Bearing Preload - National Precision Bearing
SquidPuppet is offline  
Old 12-06-17, 04:32 AM
  #33  
Bike Gremlin
Mostly harmless ™
 
Bike Gremlin's Avatar
 
Join Date: Nov 2010
Location: Novi Sad
Posts: 4,424

Bikes: Heavy, with friction shifters

Mentioned: 21 Post(s)
Tagged: 0 Thread(s)
Quoted: 1105 Post(s)
Liked 213 Times in 127 Posts
Originally Posted by Reynolds
+1. Better to have no play, but a little doesn't cause any harm IME.
This speaks more than words:



Just add that play won't result in (at least not measurably) lower drag when a bearing carries rider's weight (compared to an optimally preloaded bearing, not overtightened one), and will result in wheel's play.
Bike Gremlin is offline  
Old 12-06-17, 11:24 AM
  #34  
RubeRad
Keepin it Wheel
 
RubeRad's Avatar
 
Join Date: Aug 2011
Location: San Diego
Posts: 10,238

Bikes: Surly CrossCheck, Krampus

Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 26 Post(s)
Liked 3,399 Times in 2,510 Posts
That graph is super helpful! The takeaway being, better to err a little on the side of loose than tight
RubeRad is offline  
Old 12-06-17, 12:25 PM
  #35  
pdlamb
Senior Member
 
Join Date: Dec 2010
Location: northern Deep South
Posts: 8,847

Bikes: Fuji Touring, Novara Randonee

Mentioned: 36 Post(s)
Tagged: 0 Thread(s)
Quoted: 2577 Post(s)
Liked 1,900 Times in 1,192 Posts
Er, I think you read the graph wrong. Peak bearing life happens with just a little more than zero preload.
pdlamb is offline  
Old 12-06-17, 12:39 PM
  #36  
Metis
Junior Member
 
Join Date: Mar 2011
Posts: 114
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 21 Post(s)
Likes: 0
Liked 0 Times in 0 Posts
Not sure if this is helps but last winter I rode with play in the hubs for several months without any noticeable difference. Took it to a bike shop and they couldn't figure out why the wheel shimmied from side to side. Rode it a while longer and took it into another shop which immediately realized the cartridge bearings were shot. Replaced them and the wheel spun like new. Pretty sure the first mechanic thought I had a cup and cone bearing and tried tightening up the hub...
Metis is offline  
Old 12-06-17, 12:44 PM
  #37  
Bike Gremlin
Mostly harmless ™
 
Bike Gremlin's Avatar
 
Join Date: Nov 2010
Location: Novi Sad
Posts: 4,424

Bikes: Heavy, with friction shifters

Mentioned: 21 Post(s)
Tagged: 0 Thread(s)
Quoted: 1105 Post(s)
Liked 213 Times in 127 Posts
Originally Posted by RubeRad
That graph is super helpful! The takeaway being, better to err a little on the side of loose than tight
I added a sentence below the graph, just in case, but obviously didn't explain well enough. The no-play zone is what you should aim for - i.e. preload, no play. Just make sure not to overdo with it. Tightening until the play has just disappeared is usually good enough.
Bike Gremlin is offline  
Old 12-06-17, 12:45 PM
  #38  
SquidPuppet
Calamari Marionette Ph.D
 
SquidPuppet's Avatar
 
Join Date: Dec 2013
Location: Coeur d' Alene
Posts: 7,862

Bikes: 3 Chinese Gas Pipe Nerdcycles and 2 Chicago Electroforged Boat Anchors

Mentioned: 75 Post(s)
Tagged: 0 Thread(s)
Quoted: 2358 Post(s)
Liked 33 Times in 26 Posts
Originally Posted by pdlamb
Er, I think you read the graph wrong. Peak bearing life happens with just a little more than zero preload.
I think what Rube is saying is that once you begin to move away from "perfect", the consequences are greater and occur more rapidly if you move towards increased preload vs towards increased clearances.

I think. And that's if I am reading the graph correctly as well.
SquidPuppet is offline  
Old 12-06-17, 12:56 PM
  #39  
RubeRad
Keepin it Wheel
 
RubeRad's Avatar
 
Join Date: Aug 2011
Location: San Diego
Posts: 10,238

Bikes: Surly CrossCheck, Krampus

Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 26 Post(s)
Liked 3,399 Times in 2,510 Posts
Yes, the Squid understands. 'perfect' is not possible, so if you keep getting what you percieve is a little too far on either side, one type of error is preferable.
RubeRad is offline  
Old 12-06-17, 04:29 PM
  #40  
WizardOfBoz
Generally bewildered
 
WizardOfBoz's Avatar
 
Join Date: Aug 2015
Location: Eastern PA, USA
Posts: 3,036

Bikes: 2014 Trek Domane 6.9, 1999 LeMond Zurich, 1978 Schwinn Superior

Mentioned: 20 Post(s)
Tagged: 0 Thread(s)
Quoted: 1152 Post(s)
Liked 340 Times in 250 Posts
My setup uses Polk speakers (fronts, which I have bi-amped, central, subwoof, rears) and a Yamaha RX-V1000. The RX-V1000 came with a microphone that you plug in, and then you run a calibration routine, and the unit sets the equalization constants. Sounds great.

I think if you have a ballsy enough amp, and you work hard to get the EQ set right, the 901s (any version) will sound fine. If your amp is a good one, but not so powerful (e.g. 20 or 30 W per channel), you might be better suited to using another speaker. Other folks here, more expert than I, have suggested some names.
WizardOfBoz is offline  
Old 12-06-17, 05:28 PM
  #41  
3alarmer 
Friendship is Magic
 
3alarmer's Avatar
 
Join Date: Nov 2010
Location: Sacramento, CA
Posts: 22,932

Bikes: old ones

Mentioned: 304 Post(s)
Tagged: 0 Thread(s)
Quoted: 26249 Post(s)
Liked 10,231 Times in 7,100 Posts
.
...42.
3alarmer is online now  
Old 12-06-17, 05:38 PM
  #42  
RubeRad
Keepin it Wheel
 
RubeRad's Avatar
 
Join Date: Aug 2011
Location: San Diego
Posts: 10,238

Bikes: Surly CrossCheck, Krampus

Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 26 Post(s)
Liked 3,399 Times in 2,510 Posts
In response to WizardOfBoz, I would just say that my preferred bread to bake is the no-knead refrigerator dough recipe from Artisan Bread in 5 Minutes a Day.
RubeRad is offline  
Old 12-07-17, 04:49 AM
  #43  
Bike Gremlin
Mostly harmless ™
 
Bike Gremlin's Avatar
 
Join Date: Nov 2010
Location: Novi Sad
Posts: 4,424

Bikes: Heavy, with friction shifters

Mentioned: 21 Post(s)
Tagged: 0 Thread(s)
Quoted: 1105 Post(s)
Liked 213 Times in 127 Posts
Originally Posted by RubeRad
Yes, the Squid understands. 'perfect' is not possible, so if you keep getting what you percieve is a little too far on either side, one type of error is preferable.
If you tighten until all the play is gone (for QR axles it is tested with the QR locked - loose QR should allow a bit of play) - then you are usually getting somewhere between the zero preload (and zero play) and optimal preload. Loosening further, so that there is play is not a good choice in my knowledge and experience. Bearnings get pitted and the wheel has play - (bad for braking, little less for steering as well).
Bike Gremlin is offline  
Old 12-07-17, 08:52 AM
  #44  
kcblair
Old Legs
 
kcblair's Avatar
 
Join Date: Nov 2016
Location: Mass.
Posts: 1,212

Bikes: '80 Strayvaigin, '84 Ciocc Aelle-Shimano 105, '90 Concorde Astore /Campy Triple ,85 Bridgestone 500/Suntour, 2005 Jamis Quest, 2017 Raleigh Merit 1, Raleigh Carbon Clubman

Mentioned: 10 Post(s)
Tagged: 0 Thread(s)
Quoted: 302 Post(s)
Liked 33 Times in 22 Posts
I was taught a loooooong time ago, to adjust hub bearing to a point, there is very slight play when wiggling the axle side to side. When the wheel is installed and QR closed, there should be no play and the wheel should rotate under the weight of the valve. I have 4 sets of wheels from the 80's, using this practice, and absolutely no wear to the cones and races.

A few years back, I bought a new replacement wheel for a bike at the LBS, and the adjustment was so tight, you could feel the indexing of the bearings.

I will continue with this practice.

Oh, found the reference. The Best Of Repair Stand, by Don Cuerdon.

"Check the adjustment by alternately turning the axle between the thumb and forefinger sideways. The axle should turn freely and feel slightly loose (to allow for compression when the wheel is fasten to the frame).

When the wheel is installed, there should be no side play at the rim, and the weight of the valve stem (positioned at 3 or 9 o'clock) should be enough to turn the wheel . Keep fusing until you get it right."

KB
kcblair is offline  
Old 12-07-17, 09:06 AM
  #45  
wphamilton
Senior Member
 
wphamilton's Avatar
 
Join Date: Apr 2011
Location: Alpharetta, GA
Posts: 15,280

Bikes: Nashbar Road

Mentioned: 71 Post(s)
Tagged: 0 Thread(s)
Quoted: 2934 Post(s)
Liked 341 Times in 228 Posts
Originally Posted by Slaninar
This speaks more than words:

I think it speaks the same thing as "I'd rather have a little play than too tight with no play", but try to get in between those two.
wphamilton is offline  
Old 12-07-17, 09:09 AM
  #46  
Bike Gremlin
Mostly harmless ™
 
Bike Gremlin's Avatar
 
Join Date: Nov 2010
Location: Novi Sad
Posts: 4,424

Bikes: Heavy, with friction shifters

Mentioned: 21 Post(s)
Tagged: 0 Thread(s)
Quoted: 1105 Post(s)
Liked 213 Times in 127 Posts
Originally Posted by wphamilton
I think it speaks the same thing as "I'd rather have a little play than too tight with no play", but try to get in between those two.
It is far from that. From the graph it is visible that no play part give shorter lifetime. As well as wheel that... well - has play.

It is just important not to overdo the preload - but that is not hard - just stop tightening as soon as all the play is gone. There is a margin - as you can tell from the graph.

One thing that isn't shown in the graph: for cup and cone bearings, leaving play will result in pitting on the cones (at least), which is more bother to source and more expensive to replace (than just balls). Making it tight just enough to eliminate play allows for hubs to last for decades - just replace balls from time to time.
Bike Gremlin is offline  
Old 12-07-17, 12:23 PM
  #47  
wphamilton
Senior Member
 
wphamilton's Avatar
 
Join Date: Apr 2011
Location: Alpharetta, GA
Posts: 15,280

Bikes: Nashbar Road

Mentioned: 71 Post(s)
Tagged: 0 Thread(s)
Quoted: 2934 Post(s)
Liked 341 Times in 228 Posts
Originally Posted by Slaninar
It is far from that. From the graph it is visible that no play part give shorter lifetime. As well as wheel that... well - has play.
Seriously? Too tight has a very steep line for rated life, and just too the right of the maximum (a little play) has much less slope. Best is anywhere in the immediate region beyond "too tight'. It is precisely what I said.
wphamilton is offline  
Old 12-07-17, 12:27 PM
  #48  
Racing Dan
Senior Member
 
Racing Dan's Avatar
 
Join Date: Aug 2015
Posts: 2,224
Mentioned: 9 Post(s)
Tagged: 0 Thread(s)
Quoted: 1332 Post(s)
Liked 316 Times in 214 Posts
Originally Posted by SquidPuppet
My wheel bearings have zero play, NONE, and are therefore preloaded. They are not tight, which would be excessive preload, resulting in an interference fit.

Once spun, my wheels will eventually "pendulum" for a few cycles before they finally come to rest.


Bearing Preload - National Precision Bearing

That makes no sense. "Inteference" is almost the definition of preload. That is, Preload is the opposite of clearance. Also the post I replied to claimed a wheel should be both free spinning under the weight of the valve(!) but be preloaded as well ... This sounds suspiciously like having your cake and eating it too.

Trouble with "internet quote war" is that there is an inherent bias to searching certain terms like "bearing preload". Try searching for "bearing clearance" instead. Then you get this (SKF):

Bearings must have the appropriate operating clearance to operate satisfactorily (→ Importance of selecting correct clearance/preload).
In most cases, bearings require a certain degree of clearance (→ Selecting internal clearance). However, in some cases, they may require preload (i.e. negative clearance → Selecting preload).


As a general rule:

Ball bearings should have an operating clearance that is virtually zero.

Cylindrical, needle, spherical and CARB toroidal roller bearings typically require at least a small operational clearance.

Tapered roller and angular contact ball bearings should have a small operational clearance, except in applications where a high degree of stiffness or positional control is required, in which case they can be mounted with a degree of preload.


Selecting internal clearance or preload

or you can read the SKF catalog that has a lot if interesting info:


"Clearance versus preload

For most applications, bearings operate with
some residual clearance. Normally, a positive
operating clearance near zero is the optimum
(† diagram 2).
A somewhat greater clearance may be more
suitable for:
• high-speed applications to reduce frictional
heat
• form errors on the shaft or housing seat
such as ovality
The initial internal clearance prior to mounting
and permissible reduction after mounting, depend
on the type and size of the bearing. The
reduction in clearance due to an interference
fit may require greater initial internal clearance
than Normal to avoid preloading the
bearing († fig. 15, page 167).
Preload (negative operating clearance) has
advantages, but can also be risky. If a high degree
of stiffness is required, light preload can
be suitable († Bearing preload, page 214).
A light preload may also be required when
there is a very light or no external load on the
bearing in operation.
However, there is a risk that too much
preload causes the bearing to overheat, which
further increases preload, friction and heat.
This cycle can continue until the bearing
seizes.
It could be argued that preload is acceptable,
provided the bearing operates in a zone that
does not exceed light preload († diagram 2,
zone between 0 and –1). In this case, however,
there is an increase in friction and frictional
heat.
Although all bearing types can run with some
preload, SKF recommends a positive operating
clearance. This is particularly important for
roller bearings such as cylindrical roller, needle
roller, spherical roller and CARB toroidal roller
bearings."


https://www.google.dk/url?sa=t&rct=j...tHGX9qtuKpVRpv

Page ~212.


What Im getting at is that yeah, preload is "a thing", but none of the applications that require actual preload has much resemblance to bicycle wheels and usually preload is employed for a variety of reasons other than bearing longevity. Anybody can make up their own mind as to how much, or little, or none, they want to preload their wheel bearings, but claiming bearings should always be preloaded is patently false even if you operate with novel definitions of "preload".

Last edited by Racing Dan; 12-07-17 at 12:31 PM.
Racing Dan is offline  
Old 12-07-17, 01:16 PM
  #49  
3alarmer 
Friendship is Magic
 
3alarmer's Avatar
 
Join Date: Nov 2010
Location: Sacramento, CA
Posts: 22,932

Bikes: old ones

Mentioned: 304 Post(s)
Tagged: 0 Thread(s)
Quoted: 26249 Post(s)
Liked 10,231 Times in 7,100 Posts
Originally Posted by Racing Dan
Also the post I replied to claimed a wheel should be both free spinning under the weight of the valve(!) but be preloaded as well ... This sounds suspiciously like having your cake and eating it too.
...it's magical when you get it right.

Originally Posted by Racing Dan
Trouble with "internet quote war" is that there is an inherent bias to searching certain terms like "bearing preload". Try searching for "bearing clearance" instead. Then you get this (SKF):

Bearings must have the appropriate operating clearance to operate satisfactorily (→ Importance of selecting correct clearance/preload).
In most cases, bearings require a certain degree of clearance (→ Selecting internal clearance). However, in some cases, they may require preload (i.e. negative clearance → Selecting preload).


As a general rule:

Ball bearings should have an operating clearance that is virtually zero.

Cylindrical, needle, spherical and CARB toroidal roller bearings typically require at least a small operational clearance.

Tapered roller and angular contact ball bearings should have a small operational clearance, except in applications where a high degree of stiffness or positional control is required, in which case they can be mounted with a degree of preload.


Selecting internal clearance or preload

or you can read the SKF catalog that has a lot if interesting info:


"Clearance versus preload

For most applications, bearings operate with
some residual clearance. Normally, a positive
operating clearance near zero is the optimum
(† diagram 2).
A somewhat greater clearance may be more
suitable for:
• high-speed applications to reduce frictional
heat
• form errors on the shaft or housing seat
such as ovality
The initial internal clearance prior to mounting
and permissible reduction after mounting, depend
on the type and size of the bearing. The
reduction in clearance due to an interference
fit may require greater initial internal clearance
than Normal to avoid preloading the
bearing († fig. 15, page 167).
Preload (negative operating clearance) has
advantages, but can also be risky. If a high degree
of stiffness is required, light preload can
be suitable († Bearing preload, page 214)
.
A light preload may also be required when
there is a very light or no external load on the
bearing in operation.
However, there is a risk that too much
preload causes the bearing to overheat, which
further increases preload, friction and heat.
This cycle can continue until the bearing
seizes.
It could be argued that preload is acceptable,
provided the bearing operates in a zone that
does not exceed light preload († diagram 2,
zone between 0 and –1).
In this case, however,
there is an increase in friction and frictional
heat.
Although all bearing types can run with some
preload, SKF recommends a positive operating
clearance. This is particularly important for
roller bearings such as cylindrical roller, needle
roller, spherical roller and CARB toroidal roller
bearings."


https://www.google.dk/url?sa=t&rct=j...tHGX9qtuKpVRpv

Page ~212.

Originally Posted by Racing Dan
What Im getting at is that yeah, preload is "a thing", but none of the applications that require actual preload has much resemblance to bicycle wheels and usually preload is employed for a variety of reasons other than bearing longevity. Anybody can make up their own mind as to how much, or little, or none, they want to preload their wheel bearings, but claiming bearings should always be preloaded is patently false even if you operate with novel definitions of "preload".
...now you're just making stuff up. If anything, a bicycle wheel most resembles the application in your reference that I have bolded in your quote from it. As to definitions, that's exactly why I've been asking for references in this thread since it started.

Nothing anyone writes here is going to have much effect on how things get done IRL, but it would be nice if we could occasionally have an informed discussion in here on topics that have some real world research and answers without the butthurt that inevitably follows in defense of positions that might be based on something else.


But why would we want to change longstanding tradition at this point, eh ?
3alarmer is online now  
Old 12-07-17, 01:16 PM
  #50  
SquidPuppet
Calamari Marionette Ph.D
 
SquidPuppet's Avatar
 
Join Date: Dec 2013
Location: Coeur d' Alene
Posts: 7,862

Bikes: 3 Chinese Gas Pipe Nerdcycles and 2 Chicago Electroforged Boat Anchors

Mentioned: 75 Post(s)
Tagged: 0 Thread(s)
Quoted: 2358 Post(s)
Liked 33 Times in 26 Posts
Originally Posted by Racing Dan
That makes no sense. "Inteference" is almost the definition of preload.
Interference is when there is binding. Snug to the point that the adjustment is interfering with efficient operation. That is the definition of too much, or incorrect, or poorly adjusted preload.

Correct preload is exactly what it says. Before load. Peload does not mean loaded or tight. It simply means that clearances have been adjusted out.

Correct Preload = zero play. Eliminating clearances is the literal definition of being preloaded.

Correct Preload does not = tight/binding/interfering.

I agree with SKF here, and it sounds like a bicycle application to me. We don't want wobbly wheels or headsets. And we certainly want slopless crank operation.

If a high degree
of stiffness is required, light preload can
be suitable
I agree here as well, and this sounds a lot like a bicycle application too.

A light preload may also be required when
there is a very light or no external load on the
bearing in operation.
What Im getting at is that yeah, preload is "a thing", but none of the applications that require actual preload......
Actual preload is a matter of degrees, ranging from light (and it's many benefits) to excessive and it's destructive consequences.

When we buy and install a brand new cartridge bottom bracket it has zero play. That is because it has been set with a light preload at the factory. Later in it's service life the bottom bracket will exhibit some play or slop. It is worn and that has caused a loss of it's preload. AKA it's toast.

Last edited by SquidPuppet; 12-07-17 at 01:22 PM.
SquidPuppet is offline  

Thread Tools
Search this Thread

Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service -

Copyright © 2024 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.