Optimal kinematic viscosity for mimimal wear?

Status
Not open for further replies.
Originally Posted By: BuickGN
Originally Posted By: CATERHAM
Originally Posted By: Junior
Originally Posted By: BuickGN


I guess what I'm looking for is an admission that higher power outputs regardless of temperature require higher viscosities to keep things separated.


In general for a given bearing, higher loading or lower RPM will require a higher viscosity. But the viscosity selected has to take into account the actual operational conditions and physical size of the bearing. For example, and F1 engine has a high specific power output per liter but I have read they use low viscosity oils. Of course they spin 19,000 rpm and are not a fair comparison when speaking in terms of passenger car engines.

The point is, viscosity is selected based on the application not just the power output alone.





You don't know what they use for a fact so don't post opinions as such.

I get condemned for posting my "race engine" results in the GN with my stock clearances and production parts using a thick oil to survive yet it's ok for you guys to mention Top Fuel and F1. And consider that valide. Here's a question, how much torque do F1 cars make?? You may find you answer in there.


BuickGN, my mentioning something I read is not stating it as fact. This is a discussion board, not a technical paper where I am required to credit my source. Your arguments are based on your experiences and the data you collected. You studied the data and developed your own conclusions. Like it or not, that is your opinion. And, in my post I did mention the F1 is not a fair comparison to typical automotive.

The funny thing is, I agree with you that if you take a stock engine and increase the power it produces, you probably need to increase the viscosity.

Where I don't agree with you is where you try to use power and torque to mean the same thing. Power is a calculated value based on (in terms of an internal combustion engine) two measurable values, speed and torque. You can have two engines that make the exact same power but operate at very different speeds and have very different torque curves.

In answering your torque question, lets take the example of doubling the power of an engine. There are two basic ways to accomplish this. 1)keep the RPM constant and double the torque. 2) keep the torque constant and double the speed. In both cases the overall power output has been doubled.

In example 1, Yes you are most likely going to have to increase the viscosity because that torque relates to higher loading on the bearings. This is consistent with what I originally stated about increased load requires increased viscosity.

In example 2, the torque load did not increase so we do not necessarily need to increase the viscosity because the increased bearing speed is going to increase the size of the hydrodynamic wedge in the bearing. Increased speed does present it's own challenges which I admit I am ignoring to keep this example simple.

Like I said before, viscosity is selected for the application. A blanket statement of needing a higher viscosity for higher power output is not correct without knowing the operating conditions and engine design. And to repeat myself, there are formulas and tables to calculate the required viscosity for a journal bearing based on the speed, load, bearing design and desired hydrodynamic wedge thickness. It's not a guessing game. It is a compromise to select a lubricant the will have viscosity characteristics that will work across the expected operational profile of the bearing.
 
Originally Posted By: Junior
Originally Posted By: BuickGN
Originally Posted By: CATERHAM
Originally Posted By: Junior
Originally Posted By: BuickGN


I guess what I'm looking for is an admission that higher power outputs regardless of temperature require higher viscosities to keep things separated.


In general for a given bearing, higher loading or lower RPM will require a higher viscosity. But the viscosity selected has to take into account the actual operational conditions and physical size of the bearing. For example, and F1 engine has a high specific power output per liter but I have read they use low viscosity oils. Of course they spin 19,000 rpm and are not a fair comparison when speaking in terms of passenger car engines.

The point is, viscosity is selected based on the application not just the power output alone.





You don't know what they use for a fact so don't post opinions as such.

I get condemned for posting my "race engine" results in the GN with my stock clearances and production parts using a thick oil to survive yet it's ok for you guys to mention Top Fuel and F1. And consider that valide. Here's a question, how much torque do F1 cars make?? You may find you answer in there.


BuickGN, my mentioning something I read is not stating it as fact. This is a discussion board, not a technical paper where I am required to credit my source. Your arguments are based on your experiences and the data you collected. You studied the data and developed your own conclusions. Like it or not, that is your opinion. And, in my post I did mention the F1 is not a fair comparison to typical automotive.

The funny thing is, I agree with you that if you take a stock engine and increase the power it produces, you probably need to increase the viscosity.

Where I don't agree with you is where you try to use power and torque to mean the same thing. Power is a calculated value based on (in terms of an internal combustion engine) two measurable values, speed and torque. You can have two engines that make the exact same power but operate at very different speeds and have very different torque curves.

In answering your torque question, lets take the example of doubling the power of an engine. There are two basic ways to accomplish this. 1)keep the RPM constant and double the torque. 2) keep the torque constant and double the speed. In both cases the overall power output has been doubled.

In example 1, Yes you are most likely going to have to increase the viscosity because that torque relates to higher loading on the bearings. This is consistent with what I originally stated about increased load requires increased viscosity.

In example 2, the torque load did not increase so we do not necessarily need to increase the viscosity because the increased bearing speed is going to increase the size of the hydrodynamic wedge in the bearing. Increased speed does present it's own challenges which I admit I am ignoring to keep this example simple.

Like I said before, viscosity is selected for the application. A blanket statement of needing a higher viscosity for higher power output is not correct without knowing the operating conditions and engine design. And to repeat myself, there are formulas and tables to calculate the required viscosity for a journal bearing based on the speed, load, bearing design and desired hydrodynamic wedge thickness. It's not a guessing game. It is a compromise to select a lubricant the will have viscosity characteristics that will work across the expected operational profile of the bearing.


I actually agree with you completely. This is what I've been trying to say. If you increase the frequency of combustion events but don't add torque, the viscosity requirements won't change or won't change much. Double the torque and they do.

This was the point of asking the question how much torque F1 engines make. Going from a bad memory I've seen some in the 200-300lb range.
 
Originally Posted By: SuperBusa


Good question ... and yes, that's a factor. Also people need to realize that dragsters and F1s are designed to last a race or two, then they get a teardown and rebuild, or replaced all together. When you are looking for a few more HP with a disposable engine it's a different "application" of oil.



2009 rules mandate that F1 teams get 8 engines total that must last season. 17 races this year. Gearboxes must last 4 races.
 
Originally Posted By: BuickGN


I actually agree with you completely. This is what I've been trying to say. If you increase the frequency of combustion events but don't add torque, the viscosity requirements won't change or won't change much. Double the torque and they do.

This was the point of asking the question how much torque F1 engines make. Going from a bad memory I've seen some in the 200-300lb range.


300 lb-ft at 18000rpm is about 1000Hp.

Yeah, I thought we agreed but just had a terminology problem.

My point was that what the mfg recommends is adequate across the typical consumers driving profile and intended use of the vehicle. Deviate from that through engine modifications and/or duty cycle and one needs to rethink the viscosity question.
 
Originally Posted By: BuickGN
Originally Posted By: FZ1
Well,that's part of the fun. You just correct it. Lol. What kind of times does your little Buick run?
Originally Posted By: BuickGN
Originally Posted By: FZ1
You're breakin' that car too much! You need to get some tires with less rear traction so you can spin them more easily and,therefore, take the strain off you engine and drive train. Then it won't break so often.


Street tires spin at 70mph. It's as much a matter of safety as it is performance. Breakage is ok, it's part of the game. When you consider the car hasn't broken in 3 years at this level using mostly stock parts, I can't really be upset.


10.60@126 with a best mph of 129mph. I've taken 300lbs out of it and added another 75hp or so so it should run good next time out.

Spinning is fun sometimes but things happen very fast at higher speeds and while I've driven it sideways a lot and never spun out, my luck will run out one day. I used to enter the freeway around my favorite onramp at 45mph and punch it as the corner straightened and let it drift over 3 lanes into the fast lane.


Well finally something on which we agree; drifting is a load of fun!
 
Originally Posted By: Junior
Originally Posted By: BuickGN


I actually agree with you completely. This is what I've been trying to say. If you increase the frequency of combustion events but don't add torque, the viscosity requirements won't change or won't change much. Double the torque and they do.

This was the point of asking the question how much torque F1 engines make. Going from a bad memory I've seen some in the 200-300lb range.


300 lb-ft at 18000rpm is about 1000Hp.

Yeah, I thought we agreed but just had a terminology problem.

My point was that what the mfg recommends is adequate across the typical consumers driving profile and intended use of the vehicle. Deviate from that through engine modifications and/or duty cycle and one needs to rethink the viscosity question.


Peak HP and TQ don't don't occur at the same RPM though. Just because they rev to 18K doesn't mean they make peak HP at 18K, and even if they do, they aren't making peak TQ there, as it always occurs at a lower RPM point.

To extrapolate a bit:

My 302 made peak torque at a hair over 3,000RPM. Peak HP didn't occur until 5,800RPM. That was with my old cam.

So, if the F1 car makes peak TQ of 300lb-ft at 12,000RPM, but peak HP at 17,000RPM, then we must assume at that 17,000RPM, the TQ produced is going to be less.

It would be interesting to see a dyno plot for one of those engines.
 
Originally Posted By: OVERK1LL
Originally Posted By: Junior
Originally Posted By: BuickGN


I actually agree with you completely. This is what I've been trying to say. If you increase the frequency of combustion events but don't add torque, the viscosity requirements won't change or won't change much. Double the torque and they do.

This was the point of asking the question how much torque F1 engines make. Going from a bad memory I've seen some in the 200-300lb range.


300 lb-ft at 18000rpm is about 1000Hp.

Yeah, I thought we agreed but just had a terminology problem.

My point was that what the mfg recommends is adequate across the typical consumers driving profile and intended use of the vehicle. Deviate from that through engine modifications and/or duty cycle and one needs to rethink the viscosity question.


Peak HP and TQ don't don't occur at the same RPM though. Just because they rev to 18K doesn't mean they make peak HP at 18K, and even if they do, they aren't making peak TQ there, as it always occurs at a lower RPM point.

To extrapolate a bit:

My 302 made peak torque at a hair over 3,000RPM. Peak HP didn't occur until 5,800RPM. That was with my old cam.

So, if the F1 car makes peak TQ of 300lb-ft at 12,000RPM, but peak HP at 17,000RPM, then we must assume at that 17,000RPM, the TQ produced is going to be less.

It would be interesting to see a dyno plot for one of those engines.


I know they don't. I just selected a rpm point and calculated the Hp. I'm familiar with torque rise, power bulge and governor slope as applied to diesel engines. I don't directly care about Hp, give me a torque curve.

As far as the F1 torque curve, I would imagine it would be relatively flat between 12,000rpm and 17,000rpm for linear power delivery.
 
Originally Posted By: Junior

I know they don't. I just selected a rpm point and calculated the Hp.


Ahhh, OK, since you used the "top" number that BuickGN quoted, I wasn't sure, which is why I posted what I did.

Quote:
I'm familiar with torque rise, power bulge and governor slope as applied to diesel engines. I don't directly care about Hp, give me a torque curve.


Don't forget "area under the curve" as well
wink.gif


Quote:
As far as the F1 torque curve, I would imagine it would be relatively flat between 12,000rpm and 17,000rpm for linear power delivery.


For sure, I imagine it would be quite flat. I would love to see a dyno pull from one though, I think it would look quite neat.
 
Originally Posted By: Ben99GT
Regarding F1 engines, here's an interesting read: http://www.epi-eng.com/piston_engine_technology/comparison_of_cup_to_f1.htm


Great read!

Quote:
At the end of the 2006 season, Formula One engines typically used a 20,000 RPM redline (sometimes even throughout the course of a Grand Prix), and produced a peak power of about 755 BHP at above 19,000 RPM, with a peak torque of about 214 lb-ft (290 nm) at 17,000 RPM.
 
Originally Posted By: virginoil


So after 18 pages, will the BITOG genius's please tell me what is the universal grade for street use, temperate -20 to 40 deg C.
for a 2 litre to 4 litre engine, regardless of vehicle brand or type.


synthetic 10w-30 in my book. i'm no genius either. has a decent HTHS, and low VII.
BTW please elaborate where in WA you run into -20 deg C temps? must be kalgoorlie ey.
 
Last edited:
Crinkles

I'm about 20km south of Perth. I threw in -20 deg C, just to get an oil that covers most operating conditions.

After 24 pages of debate, I want some recommendations / conclusions.

I like reading the BITOG "Optimal kinematic viscosity for minimal wear?”, war and peace edition version that is being released page by page
 
Last edited:
Originally Posted By: OVERK1LL
So, if the F1 car makes peak TQ of 300lb-ft at 12,000RPM, but peak HP at 17,000RPM, then we must assume at that 17,000RPM, the TQ produced is going to be less.

It would be interesting to see a dyno plot for one of those engines.


Agree with what you said, but I'd bet that a hi-pro engine like an F1 would have a (relatively) pretty flat torque curve due to near perfect volumetric efficiency. If a torque curve is nearly flat, then the HP curve will be nearly linear. They obviously make peak HP at or very near redline, as they spend a lot of time there and shift right at redline when going through the gears.
 
Originally Posted By: Junior
I don't directly care about Hp, give me a torque curve.


As you know, HP is derived by measuring T at speed (RPM). But it's really HP that counts ... of course it takes the right gearing to made full use of the power. If you could use a motor that made 50 ft-lbs of torque at 100,000 RPM it would make 952 HP ... but the gearing would be a challenge to make it useful for a wheel driven car, etc.

Originally Posted By: Junior
As far as the F1 torque curve, I would imagine it would be relatively flat between 12,000rpm and 17,000rpm for linear power delivery.


Yep ... same thing I said, but I didn't see your post before making the same comment.
thumbsup2.gif
 
Originally Posted By: virginoil

After 24 pages of debate, I want some recommendations / conclusions.


Consult your Owner's Manual.
whistle.gif
 
Originally Posted By: CATERHAM
Originally Posted By: ekpolk
Originally Posted By: johnbottilaw


If you drive a BMW M series car, then 10w-40 or 20w-50 is a dangerously "thin" oil.

I disagree with this. In the owner's manual for M engines it says 40 and 50 weights are acceptable and not a problem if 60 weight unavailable. I have UOA's from my 2008 M3 using 30 weights that show far less wear than shorter change intervals with the 10w-60.


That fine, I'm no expert on BMWs. I used this example only to highlight the problem is using hazy, subjective terms such as "thick" or "thin". Turning back to the example, what's "thick" to my 0w-20 Camry might be very "thin" for the M-series BMW. I should have made what I was saying clearer.
cheers3.gif



It's all about temperature. That's why it's gets very misleading to talk about any SAE grade of oil without that temperature context. If you're talking kinematic viscosity you've largely eliminated the temperature component from the the discussion.
I love to tell my friends I'm using a 0W-20 oil at the track for the shock effect. The reality is I'm running a 9 cSt oil. And on the street typically a 17 cSt oil.


my point exactly, you nailed it!
 
I wanted to touch back here for a bit.

Quote:
Wear at 50°C was extremely small due to significant reduction in the severity of contacts due to the increased film thickness resulting from increased oil viscosity. The wear rate under this condition was near zero. However, the wear rate increased when the oil temperature was raised to 100°C again.

Also, the friction torque decreased with increased speed at 100°C because the valvetrain operates in the mixed or EHD lubrication regime. However, at 50°C the friction torque variation with speed became flatter compared to that observed at 100°C. This is due to fewer mechanical interactions between asperities on two sliding surfaces due to increased oil viscosity.


This is exactly the opposite on the Schneider Cylinder/Ring wear paper. In that study, wear was highest when the engine was cold and evaporated as the engine warmed (visc obviously being lower). This is 180° out of that wear relationship.
 
Lol. I think that we also agree that thinner is better,as long as it's not too thin,right?
 
Status
Not open for further replies.
Back
Top