Has there ever been a study...

[Scout1]

...that measures when an oil film breaks down and metal to metal contact occurs in an engine?

I hope this isn't a really stupid question but I've been reading some about bearing design and lubrication and to be honest, the math goes over my head. I understand RPM, viscosity and load along with bearing design are pretty much the variables that determine when an oil film will fail.

So I got to thinking (dangerous, right?

), why not instrument an engine to detect when metal to metal contact happens; low RPM/high load (lugging) or is it more likely at high RPM? How much happens at startup and how about shutdown? You could run different oils through it to compare performance. Since oil is an insulator, could you electrically isolate the crankshaft from the block and then detect by continuity when metal to metal contact occurs? Has something like that ever been done?

I guess I'm wondering how much operating margin we have in todays engines? I'd like to think the whole time my engine is running all the bearings are riding on a nice cushion of oil. But in reality, does metal to metal contact happen once in a blue moon or is it fairly frequent?

 

[Shannow]

Had some discussion here a while back

Ricardo paper

 

[A_Harman]

I participated in an uncontrolled experiment once when the oil film broke down. It took about 3 seconds after the oil pump shaft broke for the engine to seize. Engine speed went from 6000 to 0 in about the same 3 seconds. Welded all the bearings to the crank. That was a bad day.

I suppose it would be possible to measure capacitance between the crank journals and bearings in a running engine and correlate the readings to previously calibrated values measured under controlled conditions. The most critical ones to measure would be rod bearings, and they would be the most challenging because of the irregular motion of the rod. I wouldn't be surprised if it has been done before.

 

[Al]

Way too complicated..speaking as a Mechanical Engineer.

 

[turtlevette]

That's a good idea. Challenges are. How does the monitor make contact with the crank? Then what do you do about it when the film does breakdown. Is it anything different than if you weren't monitoring.

 

[zpinch]

Originally Posted By: A_Harman
I participated in an uncontrolled experiment once when the oil film broke down. It took about 3 seconds after the oil pump shaft broke for the engine to seize. Engine speed went from 6000 to 0 in about the same 3 seconds. Welded all the bearings to the crank. That was a bad day.

I suppose it would be possible to measure capacitance between the crank journals and bearings in a running engine and correlate the readings to previously calibrated values measured under controlled conditions. The most critical ones to measure would be rod bearings, and they would be the most challenging because of the irregular motion of the rod. I wouldn't be surprised if it has been done before.


It would take a [censored] of alot longer than 3 seconds for an engine to seize. Unless the engine was very hot, and 100% load? Then, yes.

 

[Scout1]

Originally Posted By: turtlevette
That's a good idea. Challenges are. How does the monitor make contact with the crank? Then what do you do about it when the film does breakdown. Is it anything different than if you weren't monitoring.


So my thought is the crank could be monitored easily by just putting a brush on the output shaft. Lets say you have a timing chain with the neoprene gear so that works but are there other parts of the crankshaft that normally make electrical contact? I'm thinking about the thrust bearing. Anyway, if you could electrically isolate the crank you should be able to monitor when it contacts the bearings by monitoring electrical continuity periodically, right? I'm assumming when the oil film breaks down, an electrical current could pass between journal and the bearing. And you would need to record the contacts through some sort of monitoring system that could correlate it with time, RPM, load and oil temperature (viscosity).

I would think some university graduate student would get him/herself sponsored to build something like this for a thesis, don't you think?

 

[turtlevette]

Originally Posted By: Scout1
Originally Posted By: turtlevette
That's a good idea. Challenges are. How does the monitor make contact with the crank? Then what do you do about it when the film does breakdown. Is it anything different than if you weren't monitoring.


So my thought is the crank could be monitored easily by just putting a brush on the output shaft. Lets say you have a timing chain with the neoprene gear so that works but are there other parts of the crankshaft that normally make electrical contact? I'm thinking about the thrust bearing. Anyway, if you could electrically isolate the crank you should be able to monitor when it contacts the bearings by monitoring electrical continuity periodically, right? I'm assumming when the oil film breaks down, an electrical current could pass between journal and the bearing. And you would need to record the contacts through some sort of monitoring system that could correlate it with time, RPM, load and oil temperature (viscosity).

I would think some university graduate student would get him/herself sponsored to build something like this for a thesis, don't you think?


It's a great idea. Write up a patent.

 

[Scout1]

Originally Posted By: Shannow
Had some discussion here a while back

Ricardo paper


Thanks! interesting read. Nice trick to be able to measure the film thickness all the way through a revolution! From the looks of this, would you say we probably don't have much metal to metal contact in our bearings?

This was 15w40 oil... wonder what 0w20 would do?

 

[Shannow]

Originally Posted By: Scout1
Thanks! interesting read. Nice trick to be able to measure the film thickness all the way through a revolution! From the looks of this, would you say we probably don't have much metal to metal contact in our bearings?


Honda in some of their papers on 16s are admitting to more operation in mixed/boundary, so I think that it's more prevalent now than before...don't know how close/much but...but they are heading more in that direction.

A thought that's been developing in my head is that the US proclivity for the Automatic transmission helps thinner oils to perform.

Reason being that the hydrodynamic "film" is produced by speed difference in the bearings, viscosity and load.

Drive off from idle in an auto, and speed and load come simultaneously...do it in my Nissan with turbodiesel and manual, and when warm, I can release the clutch (relatively quickly) at idle...load increases, but revs DROP...to about 500...load up, revs down, film thickness deteriorates.

A manual is likely more sensitive than an auto...just my theory ATM, still working on it really.

Originally Posted By: Scout1
This was 15w40 oil... wonder what 0w20 would do?

Try this thread...

 

[jaynissan12]

Only test I ever saw was one done by Royal Purple, I think its accessible on their website, but that was obviously to their gain so take it for what its worth.

 

[A_Harman]

Originally Posted By: zpinch
Originally Posted By: A_Harman
I participated in an uncontrolled experiment once when the oil film broke down. It took about 3 seconds after the oil pump shaft broke for the engine to seize. Engine speed went from 6000 to 0 in about the same 3 seconds. Welded all the bearings to the crank. That was a bad day.



It would take a [censored] of alot longer than 3 seconds for an engine to seize. Unless the engine was very hot, and 100% load? Then, yes.


Yep, it was running full power at the time. Basically a worst case scenario.

 

[Yah-Tah-Hey]

In 1979, a series of operational mistakes caused one of our GE D8 turbine/generator sets to lose oil supply to the shaft mounted oil pump and the unit coast down from 3600 RPM without lubrication. I had a piece of melted babbitt from one of the 16" bearings in my desk for many years.

 

[Ken2]

Metal to metal contact never occurs. If it does, you get towed.

One of the responsibilities of the additive package in the oil is to provide the materials for boundary lubrication. This prevents metal to metal contact until the hydrodynamic oil wedge has time to build.

Heard of a "spun" bearing? The bearing shell spinning is actually the third step in the bearing failure. First the oil film breaks down, usually from excessive force in a modified engine or oil supply failure. Next the bearing metal wipes and the shell seizes (welds) to the journal. Third, the shell spins.

 

[Scout1]

Originally Posted By: Ken2
Metal to metal contact never occurs. If it does, you get towed.



Ummm... then what causes all those scuff marks you often see on bearings when an engine is rebuilt? Is it only start-up and shut-down wear?

 

[Fordtrucktexan]

You could just pull the motor and plasti-gauge the bearings for an exact measure of wear.

I'll see if I can dig up some old bearings and post pictures of them. It's kinda rare to see any noticable wear as long as the engine has had oil in it.

My 01' 1.9 Saturn is currently sitting at my dads driveway after it spun a bearing due to loss of oil pressure (no oil). When I pull the motor I'll post the pics of that damage

 

[deven]

Originally Posted By: Fordtrucktexan
You could just pull the motor and plasti-gauge the bearings for an exact measure of wear.

I'll see if I can dig up some old bearings and post pictures of them. It's kinda rare to see any noticable wear as long as the engine has had oil in it.

My 01' 1.9 Saturn is currently sitting at my dads driveway after it spun a bearing due to loss of oil pressure (no oil). When I pull the motor I'll post the pics of that damage

We need more posters like you with "real" data rather than "empirical" data.