How much Tolerance needed for bearing inserted into aluminium alloy ?

[firasf1dream]

For example, if the bearing has been manufactured with certain amount of internal radial clearance, and it is installed into the housing with sufficient interference at the outer race OD to eliminate the internal radial clearance, it will cause failure of the bearing.

what do you mean by sufficient interference to eliminate the internal radial clearance ?! isn't it sufficient ?

[riff_raff]

Say your cylindrical roller bearing is manufactured with a standard (C3) radial internal clearance of around .002". And you install the bearing into a housing bore that has much greater radial stiffness than the thin section bearing outer race, with a diametral interference of .005". This would likely cause the outer bearing race to collapse enough to produce seizure of the bearing.

If your 75mm OD roller bearing is manufactured to a typical RBEC class 5 precision, the OD tolerance will be around +0"/-.0004". And due to CTE mismatch between the steel bearing race and aluminum housing bore, you would design the parts to have a RT interference fit sufficient to prevent the outer bearing race from becoming free to rotate within the aluminum bore at peak operating temps.

[firasf1dream]

Say your cylindrical roller bearing is manufactured with a standard (C3) radial internal clearance of around .002". And you install the bearing into a housing bore that has much greater radial stiffness than the thin section bearing outer race, with a diametral interference of .005". This would likely cause the outer bearing race to collapse enough to produce seizure of the bearing.

If your 75mm OD roller bearing is manufactured to a typical RBEC class 5 precision, the OD tolerance will be around +0"/-.0004". And due to CTE mismatch between the steel bearing race and aluminum housing bore, you would design the parts to have a RT interference fit sufficient to prevent the outer bearing race from becoming free to rotate within the aluminum bore at peak operating temps.

ok i have one more question, would loctite or any strong adhesive do the job for keeping the other diameter of the bearing from rotating or moving if the maximum force that will be on is around 90 to 100N ?

[flynfrog]

its not ideal but I have done it. You want the bearing locker type I think its usually green

[flynfrog]

keep in mind you still need to have a pretty close tolerance for it to work. Its a much better idea to build it right the first time.

[firasf1dream]

i have worked with the red one before, for a nut of a bicycle, but you say for bearing it's different the green

about the tolerance yes sure it will be close, the problem why i am doing this is because i am not making the parts on a CNC machine no, i am making the molds with a laser machine for plexi that have a 0.1 diameter, and half are handmade molds that will be casted into aluminium alloy

[dans79]

ok i have one more question, would loctite or any strong adhesive do the job for keeping the other diameter of the bearing from rotating or moving if the maximum force that will be on is around 90 to 100N ?

Even with loctite, you still need to be within a few thousands of the actual diameter, as loctite is not designed to be a gap filler.

If it's important I'd recommend these steps.

1. determine the load the bearing will see and it's operating temperature.
2. Talk with the bearing manufacture and they should be able to tell you what size the bore should be given the info in #1.
3. drill the hole undersized 1mm or 2mm
4. bore the hole undersized 0.3mm or so (boring the hole accurately locates it)
5. ream the hole to size. (reaming accurately sizes the hole)

[dans79]

the problem why i am doing this is because i am not making the parts on a CNC machine

You don't need a cnc machine, you can use a manual lathe or a manual mill.

[firasf1dream]

the problem why i am doing this is because i am not making the parts on a CNC machine

You don't need a cnc machine, you can use a manual lathe or a manual mill.

but wouldn't it be expensive still ?

[firasf1dream]

ok i have one more question, would loctite or any strong adhesive do the job for keeping the other diameter of the bearing from rotating or moving if the maximum force that will be on is around 90 to 100N ?

Even with loctite, you still need to be within a few thousands of the actual diameter, as loctite is not designed to be a gap filler.

If it's important I'd recommend these steps.

1. determine the load the bearing will see and it's operating temperature.
2. Talk with the bearing manufacture and they should be able to tell you what size the bore should be given the info in #1.
3. drill the hole undersized 1mm or 2mm
4. bore the hole undersized 0.3mm or so (boring the hole accurately locates it)
5. ream the hole to size. (reaming accurately sizes the hole)

thank you dans79, yes reaming it will be good but should that be done by hand or with a rotor machine ?

[dans79]

but wouldn't it be expensive still ?

are you talking about paying someone to do it, or buying the machines to do it yourself?

thank you dans79, yes reaming it will be good but should that be done by hand or with a rotor machine ?

I'd recommend a chuck-able machine reamer.

[firasf1dream]

but wouldn't it be expensive still ?

are you talking about paying someone to do it, or buying the machines to do it yourself?

thank you dans79, yes reaming it will be good but should that be done by hand or with a rotor machine ?

I'd recommend a chuck-able machine reamer.

i meant to do it by a small drill with a bit for grinding,
i went today to the workshop of one of my professors, so we discussed the process, he said that an upright is better to be machined than caster but it would be more expensive so then we discussed about casting it and then do the bore with a lathe machine but before that i must make another FAE simulation to see if Aluminium would work or bronze and not break if yes then i will proceed because unfortunately the casting workshop does only aluminium, bronze and some other material

[riff_raff]

The way a anaerobic compound such as Loctite works is by increasing the coefficient of friction at the contact interface. The cured anaerobic compound layer itself has very little shear strength. If there is any significant clearance at assembly between parts with a CTE mismatch, the Loctite bond will separate from thermal cycling. With a suitable fit between the bearing OD and aluminum housing bore, the Loctite adhesive bond should work OK for preventing the bearing race from spinning in the housing bore.

[mep]

I must admit I am not impressed by the comments here so maybe I can add my view on the issue:

1. Tolerance and fit
First of all go and check what bearing manufacturers suggest as fitting and tolerance. Even though that would be the first thing to do and it was the first suggestion made here you still have not done that!
Why you keep discussing here without any own research?
Just for the sake of it I quickly did it myself:
Selected a bearing; figure the outer tolerance of the bearing; check what tolerance the manufacturer suggests; calculate interferences and play; -> be able to choose a tolerance.
This took just a few minutes on the internet. First I wondered why nobody just told you a possible tolerance but I guess they just don’t want to do the homework for you – so I will not do either.

2. Casting vs. machining
You want to cast an aluminium upright in the scale of 1:4 but you are unable to do any machining?
An upright in this scale is about the size of a watch, most of the surfaces do need machining. To cast such a part just does not give you any benefit. The casting process alone opens up so many questions. Are you sure the casting will be of good quality, with no cavities included, good surface finish, correct material properties, no internal stress, no deformation?
And who manufactures the moulding tools? Doing the tools will be a bigger job than milling a little upright. I want to see the place who offers you an aluminium cast for less than the machining costs and I want to see similar parts they have done before.

3. Gluing of bearings
It is not necessary to glue the bearings in. The key is that the design is a complete assembly which makes it impossible that any part can fall apart even if the fitting is too lose. Again, look up how a bearing assembly is done correctly. After you have done step one (looking up the fitting) you will understand that with the given tolerance of the bearing ring it is likely that the outer ring has a little play. Nevertheless, don’t come up with the idea that you can get away with a roughly machined oversized hole by filling the gap with glue. That’s no proper engineering. Also if everything is done correctly you should not need to prevent the bearing ring from spinning in the housing otherwise you could have designed a plain bearing from beginning on.

4. FEA!
Why you do simulations when you are not able to calculate a fitting and tolerance yourself?
All you need can be calculated by hand. Seriously, it pisses me of when people talk about simulations but don’t have the slightest glue. Learn your basics first!
You can also calculate the increase of size with temperature Greg Locock mentioned just by hand.
Also if heat shrinking the bearing in will work.
BTW. : Something with your professor must be wrong as well.

[firasf1dream]

I must admit I am not impressed by the comments here so maybe I can add my view on the issue:

1. Tolerance and fit
First of all go and check what bearing manufacturers suggest as fitting and tolerance. Even though that would be the first thing to do and it was the first suggestion made here you still have not done that!
Why you keep discussing here without any own research?
Just for the sake of it I quickly did it myself:
Selected a bearing; figure the outer tolerance of the bearing; check what tolerance the manufacturer suggests; calculate interferences and play; -> be able to choose a tolerance.
This took just a few minutes on the internet. First I wondered why nobody just told you a possible tolerance but I guess they just don’t want to do the homework for you – so I will not do either.

2. Casting vs. machining
You want to cast an aluminium upright in the scale of 1:4 but you are unable to do any machining?
An upright in this scale is about the size of a watch, most of the surfaces do need machining. To cast such a part just does not give you any benefit. The casting process alone opens up so many questions. Are you sure the casting will be of good quality, with no cavities included, good surface finish, correct material properties, no internal stress, no deformation?
And who manufactures the moulding tools? Doing the tools will be a bigger job than milling a little upright. I want to see the place who offers you an aluminium cast for less than the machining costs and I want to see similar parts they have done before.

3. Gluing of bearings
It is not necessary to glue the bearings in. The key is that the design is a complete assembly which makes it impossible that any part can fall apart even if the fitting is too lose. Again, look up how a bearing assembly is done correctly. After you have done step one (looking up the fitting) you will understand that with the given tolerance of the bearing ring it is likely that the outer ring has a little play. Nevertheless, don’t come up with the idea that you can get away with a roughly machined oversized hole by filling the gap with glue. That’s no proper engineering. Also if everything is done correctly you should not need to prevent the bearing ring from spinning in the housing otherwise you could have designed a plain bearing from beginning on.

4. FEA!
Why you do simulations when you are not able to calculate a fitting and tolerance yourself?
All you need can be calculated by hand. Seriously, it pisses me of when people talk about simulations but don’t have the slightest glue. Learn your basics first!
You can also calculate the increase of size with temperature Greg Locock mentioned just by hand.
Also if heat shrinking the bearing in will work.
BTW. : Something with your professor must be wrong as well.

thank you mep for your reply

i am going to start by telling you that i live in a country where we do not have automotive engineering first, then you can't find anyone here that has a proper automotive experience in term of engineering so about my professor that will give you the answer, nothing is wrong with him, it's not his field yet he was able to give me guidance, but i am not all the time with him, i work alone at home + in my university we do not have labs or else, so i will be the first to hit in a project like mine + the manufacturing process on my own in the history of it !

about the bearing, i looked over the internet but couldn't find any manual about it, it's a bearing i got from the yankee-legend website, when i got it back last year i had no idea whatsoever about what i was doing,
about calculation now i am reading in the shigley book about bearing to be able to make calculations and understand more about this subject

about the mold for casting i had to do it myself a plexi core + hand made cardboard arms here are pictures of the uprights but these pics were taken before i put the paste filler to have a better surface
http://i96.photobucket.com/albums/l196/ ... foobof.jpg
http://s96.photobucket.com/user/mac_987 ... 9.jpg.html
http://i96.photobucket.com/albums/l196/ ... idnqzy.jpg

i decided to make the molds myself because using a CNC machine or else or machining from 0 would cost big fortune, so i asked one of the professors who has a workshop for machining and he told me you will be able to make casting but after that you have to make machining of the bore for the bearing and for the surface on which the caliper will be assembled