Ceramic Bearings

[Terry Keeley]

Would attaching the bearing using 601 Loctite do the same thing as a tight fit?

Was very meticulous about trying that & it didn't hold.

 

[Jim Allen]

Some people have attempted to use ceramic cements to hold the inner race of a full ceramic bearing to a crankshaft. This also does not work! The inner race will shatter every time because of the very different expansion rates involved.

JA

 

[Terry Keeley]

I have a couple CMB 67's I run in my SAW boat, the crank in one has that "firm press fit" it the rear bearing, the other is loose.

I can run regular C3 fit WIB's in the "firm press fit" motor no problem, the loose fit will burn them up in a couple runs (high 20K range, signs of slipping). I can run looser fit ceramic balled bearings from Greg Settles in that motor and the bearing survives but the loose fit causes the rod to rub in the case and possibly the piston skirt to hit ABDC, so for me and that motor it's not the ideal situation.

I should mic the good crank to know what a "firm press fit" is, it's probably only a couple tenths under nominal. I'm sure Jim &/or Jack would know.

 

[dwilfong]

Terry why do you think the loose fitting crank will not destroy the loose bearings?

Dose it look like the race is spinning if you inspect the inner race?

Why dose the inner race spin in the first place?

Friction building heat and making the bearing grow making more heat and growing more feeding on its self?

Chain reaction to destruction?

 

[Jim Allen]

I have a couple CMB 67's I run in my SAW boat, the crank in one has that "firm press fit" it the rear bearing, the other is loose.

I can run regular C3 fit WIB's in the "firm press fit" motor no problem, the loose fit will burn them up in a couple runs (high 20K range, signs of slipping). I can run looser fit ceramic balled bearings from Greg Settles in that motor and the bearing survives but the loose fit causes the rod to rub in the case and possibly the piston skirt to hit ABDC, so for me and that motor it's not the ideal situation.

I should mic the good crank to know what a "firm press fit" is, it's probably only a couple tenths under nominal. I'm sure Jim &/or Jack would know.

Terry,

Every radial ball bearing manufactured as per AFBMA tolerances comes with the inner race's ID .0002" smaller than the nominal size. For example, a bearing with a listed ID of 15 mm (.5906"), would not allow a precision ground shaft whose OD is .5096" to pass through. A shaft whose OD is .5904" will pass through. Hardened precision ground shafts for checking inner race ID's are available from William T. Hutchinson. Ring gages are available from various tool suppliers for checking outer race OD's. The outer race's OD is always .0002" larger than the listed size.

An interference fit of .0003" on the inner race can not be assembled with a "firm press fit" by hand, providing that the crankshafts OD is correct. Most manufactured engines are using crankshafts that are .0001" to .0002" smaller in size than the nominal size listed. This is done to facilitate quick & easy assembly. In all Nelson race engines, the crankshaft's OD's in the areas where the bearing's ID's are sitting is .0001" to .0002" bigger than the nominal size. For example if the crankshafts nominal OD was listed as 15 mm (.5906"), the crankshaft's OD where the bearings are sitting would measure .5907" to .5908". The remainder of the crankshaft would be .0002'" smaller than 5904".

Outer races will have, in some instances, as much as .0008" interference fit. If the housing is made of a rigid material, such as hardened steel, it will be necessary to use a bearing with much more than a C-3 radial clearance. This could also be true for a housing made of aluminum. A bearing that is manufactured to a special radial clearance would be classified as having a S radial clearance. All the bearings used in the Nelson & my race engines are classified as such.

Jim Allen

 

[dwilfong]

Terry a loose crank bearing will not make the piston hit the crank. If the crank dose touch the case it needed some clearance any way.

You are only talking a clearance change of less then .001 total.

 

[Jim Allen]

delete

 

[Terry Keeley]

Terry why do you think the loose fitting crank will not destroy the loose bearings?

Dose it look like the race is spinning if you inspect the inner race?

Why dose the inner race spin in the first place?

Friction building heat and making the bearing grow making more heat and growing more feeding on its self?

Chain reaction to destruction?

I'm quite sure the inner race on the loose crank is spinning, you can see "skid marks" on the crank where the bearing sits. I think heat from this friction causes the inner race to grow and the bearing tightens up. Too little radial clearance causes the rest of the bearing to overheat and fail. I've pulled bearings after a couple runs that the outer race was blue from excess heat.

The looser fit bearings from Greg might have enough clearance to allow this expansion, also the ceramic balls may not grow or transfer heat to the outer race so they can survive the spinning.

Terry a loose crank bearing will not make the piston hit the crank. If the crank dose touch the case it needed some clearance any way.

You are only talking a clearance change of less then .001 total.

You might be right there, I didn't measure it but it seemed like a lot more than 0.001".

 

[Terry Keeley]

Terry,

Every radial ball bearing manufactured as per AFBMA tolerances comes with the inner race's ID .0002" smaller than the nominal size. For example, a bearing with a listed ID of 15 mm (.5906"), would not allow a precision ground shaft whose OD is .5096" to pass through. A shaft whose OD is .5904" will pass through. Hardened precision ground shafts for checking inner race ID's are available from William T. Hutchinson. Ring gages are available from various tool suppliers for checking outer race OD's. The outer race's OD is always .0002" larger than the listed size.

An interference fit of .0003" on the inner race can not be assembled with a "firm press fit" by hand, providing that the crankshafts OD is correct. Most manufactured engines are using crankshafts that are .0001" to .0002" smaller in size than the nominal size listed. This is done to facilitate quick & easy assembly. In all Nelson race engines, the crankshaft's OD's in the areas where the bearing's ID's are sitting is .0001" to .0002" bigger than the nominal size. For example if the crankshafts nominal OD was listed as 15 mm (.5906"), the crankshaft's OD where the bearings are sitting would measure .5907" to .5908". The remainder of the crankshaft would be .0002'" smaller than 5904".

Outer races will have, in some instances, as much as .0008" interference fit. If the housing is made of a rigid material, such as hardened steel, it will be necessary to use a bearing with much more than a C-3 radial clearance. This could also be true for a housing made of aluminum. A bearing that is manufactured to a special radial clearance would be classified as having a S radial clearance. All the bearings used in the Nelson & my race engines are classified as such.

Jim Allen

I was sure you'd know Jim, you have quantified so many things that us mere mortals only understand by "touch and feel".

So Henry likes a 3-4 tenths interference fit on that rear bearing that turns in the 30K range, I take it the nominal 0.0002" fit wasn't enough?

How much more radial clearance do the special "S" bearings he uses have? Does he use ceramic balls like the Metkemeijer brothers?

 

[dwilfong]

How dose the inner race surface look on the loose clearance ceramic bearings?

Dose it show sign of spinning?

I will take the VAC 1.05 apart after this race and see what is what. This was a new never run crank and the bearing just slipped on.

Thy are full complement bearings done by Greg. Thy are the same size as you use in the .67

Will have more than 10 gallons threw the eng by then...........that is about 3 hours of run time.

 

[Terry Keeley]

How dose the inner race surface look on the loose clearance ceramic bearings?

Dose it show sign of spinning?

I will take the VAC 1.05 apart after this race and see what is what. This was a new never run crank and the bearing just slipped on.

Thy are full complement bearings done by Greg. Thy are the same size as you use in the .67

Will have more than 10 gallons threw the eng by then...........that is about 3 hours of run time.

The crank that's under size (loose fit) has obvious sings of slipping no matter what bearing is used, it's just that Greg's ceramic balled bearings with the smaller balls and greater internal clearance survives where the stock WIB's don't.

If the bearing "just slipped on" the crank I'm sure the inner race will be sliding on the crank. Take some close up pix if you can!

 

[dwilfong]

No problem will do it next week.

 

[Jim Allen]

"I was sure you'd know Jim, you have quantified so many things that us mere mortals only understand by "touch and feel".

So Henry likes a 3-4 tenths interference fit on that rear bearing that turns in the 30K range, I take it the nominal 0.0002" fit wasn't enough?

How much more radial clearance do the special "S" bearings he uses have? Does he use ceramic balls like the Metkemeijer brothers?"



Terry,

If you actually have the .0002" interference fit you asked about between the inner race & the crankshaft, a heavy press fit with your hand will put the parts together. Larger amounts will require a manual press. A press is required to assemble & disassemble all Nelson front ends. Please be aware that the surface finish of the parts involved will also have an effect & in this case, both surfaces are ground to at least a 10 to 12 micro inch finish; the same finish that would be found on any quality dowel pin. The bearings that we use have a special radial clearance of .0011" to .0014" before they are mounted. These numbers apply to both all metal (52100 steel) & ceramic hybrid bearings with 52100 steel races. They cannot be purchased in small numbers, but WIB will supply any clearance desired if you purchase the entire run or 1000 pieces.

I designed & built a bearing tester to make sure the bearings are to specification. Rob Metkemeijer's brother is no longer alive. He uses both all metal & ceramic hybrid type bearings in his .45 speed engine. In the air the engine, the engine turns 38,000 RPM. There is a video on U-tube of a test flight. There are no crankshaft bearing problems, however there are connecting rod wear problems. He & I had an in depth discussion about a steel roller rod for his engine when he visited Aero Precision Machine in 2014.

Jim Allen

 

[Terry Keeley]

There you go again Jim, quantifying for the rest of us, thanks!

Not sure what the Nelson uses for a rear bearing but I looked up a C3 fit for the 6001, a popular 45 rear. It's given as 11 - 25 Um or about 0.0004" - 0.001". Your "S" bearings therefore have about 0.0005" more radial clearance (on average), almost twice as much! But isn't that fit just taken up by the tighter press used on the crank and housing? So once installed is would have about the same clearance as a looser fitting C3?

Another Q if you have time. What's your take on ceramic hybrid (ie: 52100 races with ceramic balls) bearings? I see there are different ceramics available nowadays, Boca has some descriptions:

http://www.bocabearings.com/bearing-types/balls

Less friction? Less heat transfer? One type better than another?

Did I say one question?

Found a recent F3D pylon race, not sure who's motors are being used but impressive to say the least!



ps: steel roller bearing rods are one of the best things to come to our engines since the tuned pipe IMHO. Hours of run time with just changing the needles, the rod and crank stays round and wears very, very little...

 

[Jim Allen]

Terry,

Radial internal clearances for single row deep groove ball bearings will be determined by the bore diameter of the bearing in question. For example, a 10 to 18 mm bore bearing, rated at C/3, SHOULD HAVE, .0004" to .0010" radial clearance. A 18 to 24 mm bore bearing, rated at C/3, SHOULD HAVE, .0005" to .0011" radial clearance. The problem with these numbers is, how does the purchaser tell if the bearing is made to the specification given. Do you tell by feel, how freely the bearing spins, how quiet it spins or how long it spins? Google any precision bearing company's radial internal clearance options. You may see things listed such as C/1, C/2, STANDARD, C/3, C/4 & C/5. Some companies will list radial clearances with letters such as L, N, P, K, T & Special.

"Your "S" bearings therefore have about 0.0005" more radial clearance (on average), almost twice as much! But isn't that fit just taken up by the tighter press used on the crank and housing? So once installed it would have about the

same clearance as a looser fitting C/3?" No, not quite. What we are doing is making sure we stay on the looser side of a C/4 radial clearance before mounting, which would be .0007" to .0013". Keep in mind that "NO ONE" is going to make

every part exactly to the nominal number. There are tolerances in all the parts involved, crankshaft OD. housing ID, inner & outer race diameters & ball diameters. We want our assemblies to be on the looser side after mounting & end up with about .004" to .006" axial play.

One reason I prefer to use only Silicon Nitride (Si3N4) balls is because the balls do not shatter when failing like other ceramic materials do. In fact, they tend to fail in a manner similar to that of steel balls. Other advantages include 40% less density than steel which means less centrifugal forces at high speeds; high modulus, 50% higher rigidity than steel; high hardness, Rc 75 to 80 compared to 58 to 64 for steel; high compressive strength, 5 to 7 times that of steel; low coefficient of friction; low thermal expansion, 25% of typical steel bearings; corrosion resistance, chemically inert.

The ceramic hybrids we have used starting in 1985 & are still using contain Silicon Nitride balls with AISI 52100 steel races. We have seen an improvement in performance with very high polishing of both races. We also use only AISI 52100 for all steel bearings. Only grade 3 & 5 balls are used & it is impossible to tell the difference with visual observation.

Jim Allen

 

[Terry Keeley]

Again Jim, thanks for taking time to pass on your incredible knowledge and experience, much appreciated.

I was kinda shocked when I saw the wide tolerance window for C3 fits, it you get one on the tight side and it's fit to the crank and housing is tight you could end up with very little room to spare. Now I see why you shoot for a looser fit.

Looks like I might have to build a bearing tester similar to yours. The boyz in the shop don't get no rest!

 

[Mike Larson]

In over my head on the tech.. wow!.. I'd had experience with ceramics from Greg years ago.. and the difference was there.. my then engine builder KB from Virginia used minimum his rear bearing to a ceramic hybrid.. some of my engines had dual front and rear ceramics... modified on road buggy novas to marine use and they achieved higher top end enough to warrant the change.. I still have a couple that run after 11 yrs, although theyre obsolete... maybe Greg will come on, but, basically his skill ( this is the tolerance knowledge you guys are writing about) was spot on with a rebearing effort from a factory bearing.. been 10 yrs.. and Kevin got out when the world didnt understand exactly the C3 vs 5, vs whatever you guys are posting now... but bottom line the precision toleranced fit Greg achieves makes the difference... my findings over the course and infant compared to whats posted, on quite a few engines had me see:.. an massive over rev would, take a hybrid bearing OUT. Rough water beating the hell outta them hurt... one ball would fracture and it was over... id maybe had a pair... spun the race on the crank too... out... i didnt take many out, but knew when they went.. I ran and always ran 16% oil... back then Byrons yellow 45 to 50% custom order fuel and id great luck there... idk.. that little edge some guys gotta have is there... the expense... well its up to you... my finding over time... its a razor blade... walk on it.. In all reality.. nova factory bearings rate high... and minimumally... a steel or chrome based bearing will exhibit signs youll see and possibly save you a catastrophic failure... ... so... some go... and... good luck... Mike

 

[dwilfong]

Good point Mike..... a ceramic hybrid will not show signs like a steel bearing as to when to change them.

Every thing has a fatigue rate and will fail at some piont.

You still have to change them out but not nearly as much.

What is the service time ???????...............Live and learn.................

 

[sbowden21]

That's funny novarossi have ceramics and they run just fine.

The 46DD has a ceramic big bearing and a high speed small.

2 seasons on an inboard 3 season on an outboard no bearing changes here...

 

[Glenn Quarles]

Yeah the nova bearings are pretty dand good for factory! This was very important to Mr Rossi and he had a very good idea of what clearances he was looking for. During my visit to the factory I watched him measure bearings. He was the only one that did that job! Just like Greg, he had thousands of balls of different sizes to correct the bearing clearances. The factory ones will never be as good as Greg's, but they are good!

GQ