M20 Air/Oil separator

I thought Jay's experience of no change in temp was enough data.

Mike
MU-2


"Matt Whiting" wrote in message
...
Mike Rapoport wrote:

Didn't we go through this already. More sump capacity doesn't provide
"more
lubrication and cooling". You noted this yourself when you installed
the
separator and the oil temp was unchanged.

Discussing it doesn't resolve it. I agree it doesn't provide more
lubrication, but I've yet to see any credible evidence of the affect on
cooling. I believe it helps, you don't believe it helps, but I don't
think either of us has any data to prove one way or the other.


Matt

Mike Rapoport wrote:

I thought Jay's experience of no change in temp was enough data.

Well, one data point is hardly compelling evidence. And the affect
might be too small to measure on the typical aircraft gauge which may
have a resolution of 3-5 degres if you are lucky. Also, you'd have to
carefully control conditions to be sure that the OAT was identical, etc.
For all I know, he may have been flying on two different days with
fairly different flight conditions.


Matt

I thought Jay's experience of no change in temp was enough data.

I don't see how you can conclude anything from the fact that the temperature
gauge reads the same.

If the oil cooler is working properly, it should keep the oil at a
steady-state temperature. In order to achieve that steady-state, it my have
to work much harder (I.E.: The thermostat may have to keep more oil flowing
through the cooler in order to maintain that steady temperature.) with less
oil on board to provide cooling.

This seems intuitive, but I honestly don't know enough about thermal
dynamics and engine design to conclusively say anything one way or the
other.

What we're not able to measure is how much "easier" it is for the engine to
remain at the a relatively cool 180 degrees, now that I've got 12 quarts
flowing through it instead of only 8.

I suppose the way to prove (or disprove) this theory would be to push the
engine to the limit, first with 8 and then with 12 quarts of oil on board,
and see if it overheats more quickly with less oil on board.

I suggest we try that with one of *your* engines first...

;-)
--
Jay Honeck
Iowa City, IA
Pathfinder N56993
www.AlexisParkInn.com
"Your Aviation Destination"

There are two terms in this temperature equation.

One is steady state thermal transfer rate. If you generate a
quantity of heat say one BTU then it will raise one quart of oil X
degrees F. If you add two times as much heat to the same oil it will
raise the temperature of the quart of oil 2X degrees F.
At any given RPM the oil pump will pump "Y" quarts of oil per minute
whether you have 2 quarts or 10 quarts in the sump since the pump is a
gear positive displacement pump. This means that the oil will carry
of the SAME amount of heat per minute if the temperature delta is the
same. If you want to get rid of more heat then you have to pump the
oil faster or heat the oil hotter to get a larger delta temperature
difference. This is the steady state condition. This is the condition
the engine is in when the temperature gage quits moving up.

The other part of the equation is the transient part that occurs on
warm up of the engine If one BTU will rase one quart of oil X degrees
then it will take ten BTUs to raise ten quarts of oil X degrees. It
might take an extra two or three minutes to raise the extra 8 quarts
of oil to the final steady state temperature. Once this temperature
was reached the extra 8 quarts does nothing for you unless you have a
big leak and start dumping oil overboard in which case it gives you
some more time before the oil all runs out.

If you put too much oil in the engine such that the crank shaft hits
the surface of the oil in the sump then a lot of mechanical energy is
transferred into the oil which heats the oil excessively due to the
excessive splashing.

The end result is the oil and engine temperature will be the same in
less than 30 minutes whether you have 2 quarts or 10 quarts as long as
the crank is not hitting the surface of the oil or you are not sucking
air into the pump due to low oil level at the pump inlet.

On Thu, 02 Sep 2004 01:29:49 GMT, "Jay Honeck"
wrote:

I thought Jay's experience of no change in temp was enough data.

I don't see how you can conclude anything from the fact that the temperature
gauge reads the same.

If the oil cooler is working properly, it should keep the oil at a
steady-state temperature. In order to achieve that steady-state, it my have
to work much harder (I.E.: The thermostat may have to keep more oil flowing
through the cooler in order to maintain that steady temperature.) with less
oil on board to provide cooling.

This seems intuitive, but I honestly don't know enough about thermal
dynamics and engine design to conclusively say anything one way or the
other.

What we're not able to measure is how much "easier" it is for the engine to
remain at the a relatively cool 180 degrees, now that I've got 12 quarts
flowing through it instead of only 8.

I suppose the way to prove (or disprove) this theory would be to push the
engine to the limit, first with 8 and then with 12 quarts of oil on board,
and see if it overheats more quickly with less oil on board.

I suggest we try that with one of *your* engines first...

;-)

Jay Honeck wrote:

I suppose the way to prove (or disprove) this theory would be to push the
engine to the limit, first with 8 and then with 12 quarts of oil on board,
and see if it overheats more quickly with less oil on board.

I suggest we try that with one of *your* engines first...

I think John Deakin of Avwebs Pelican Perch did this but in the opposite
direction...

He had an engine in his plane that was due for an overhaul. I don't
remember the situation, but he was willing to destroy it for the
information. Anyway, it was an Continental IO-550 or a -520 that normally
uses 8 to 10 quarts in the sump. He ran it on 2. He ran it hard. Not
only did the engine not self-destruct, he didn't even see an increase in
oil temperature.

I thought he wrote about this in one of his Pelican Perch articals, but I
can't seem to find it.

--
Frank Stutzman
Bonanza N494B "Hula Girl"
Hood River, OR

Jay Honeck wrote:


I suppose the way to prove (or disprove) this theory would be to push the
engine to the limit, first with 8 and then with 12 quarts of oil on board,
and see if it overheats more quickly with less oil on board.

I suggest we try that with one of *your* engines first...

;-)

The only way to control things would be to use an engine on a dyno.
Unfortunately, I don't have a dyno in my garage. :-)


Matt

John_F wrote:

There are two terms in this temperature equation.

One is steady state thermal transfer rate. If you generate a
quantity of heat say one BTU then it will raise one quart of oil X
degrees F. If you add two times as much heat to the same oil it will
raise the temperature of the quart of oil 2X degrees F.
At any given RPM the oil pump will pump "Y" quarts of oil per minute
whether you have 2 quarts or 10 quarts in the sump since the pump is a
gear positive displacement pump. This means that the oil will carry
of the SAME amount of heat per minute if the temperature delta is the
same. If you want to get rid of more heat then you have to pump the
oil faster or heat the oil hotter to get a larger delta temperature
difference. This is the steady state condition. This is the condition
the engine is in when the temperature gage quits moving up.

Or you start with cooler oil in the sump. This is what, I believe, will
happen when you have more oil in the engine. The oil has a longer
residence time in the sump and contacts more surface area of the sump
through which it may dissipate heat.

Let's run a "thought" experiment at the limits. Let's assume that the
oil level is so low that no oil is ever in the sump. The oil pump pulls
it out just as fast as it comes in, just short of the point of sucking
air. I realize this isn't possible in the real world, but that is why
this is a thought experiment. In this case, the oil will get very hot
as it is constantly being circulated through the heads which are one of
the hottest parts of most engines. The oil has very little opportunity
to reject heat in the coolest part of the engine, the sump. The
equilibrium temperature will be rather high.

Now take the other extreme. The oil sump has infinite capacity so the
oil starts out at the same temperature regardless of how hot the hot
parts of the engine are. The oil will enter the oil pump relatively
cool and pick up heat, but will never again get circulated through the
engine so it has "forever" to dissipate its heat.

A real engine is somewhere in between these to limit cases, therefore it
is reasonably logical to expect some slope that connects the
steady-state oil temperature of the one limit with the other. I don't
think it reasonable to believe that both steady-state temperatures will
be the same and thus have a zero-slope line in between. This is what
would have to be the case for the oil temperature to be completely
independent of the amount of oil in the engine.


Matt


Matt

It doesn't matter how long the oil is in the sump, it only matters how much
heat is lost from the outside of the sump to the air. Do we agree that the
only variable is how great the temperature diffference is between the
surface of the sump and the air? If so, it doesnt matter how long the oil
is in the sump. For your theory to be correct, doesn't the surface of the
sump have to be hotter? Also have you considered that on many engines the
sump is picking up a lot of radiated heat from the exhaust and that there
isn't a lot of airflow over the sump? The outer surface of the sump may
actually be hotter than the oil.

Mike
MU-2


"Matt Whiting" wrote in message
...
John_F wrote:

There are two terms in this temperature equation.

One is steady state thermal transfer rate. If you generate a
quantity of heat say one BTU then it will raise one quart of oil X
degrees F. If you add two times as much heat to the same oil it will
raise the temperature of the quart of oil 2X degrees F.
At any given RPM the oil pump will pump "Y" quarts of oil per minute
whether you have 2 quarts or 10 quarts in the sump since the pump is a
gear positive displacement pump. This means that the oil will carry
of the SAME amount of heat per minute if the temperature delta is the
same. If you want to get rid of more heat then you have to pump the
oil faster or heat the oil hotter to get a larger delta temperature
difference. This is the steady state condition. This is the condition
the engine is in when the temperature gage quits moving up.

Or you start with cooler oil in the sump. This is what, I believe, will
happen when you have more oil in the engine. The oil has a longer
residence time in the sump and contacts more surface area of the sump
through which it may dissipate heat.

Let's run a "thought" experiment at the limits. Let's assume that the
oil level is so low that no oil is ever in the sump. The oil pump pulls
it out just as fast as it comes in, just short of the point of sucking
air. I realize this isn't possible in the real world, but that is why
this is a thought experiment. In this case, the oil will get very hot
as it is constantly being circulated through the heads which are one of
the hottest parts of most engines. The oil has very little opportunity
to reject heat in the coolest part of the engine, the sump. The
equilibrium temperature will be rather high.

Now take the other extreme. The oil sump has infinite capacity so the
oil starts out at the same temperature regardless of how hot the hot
parts of the engine are. The oil will enter the oil pump relatively
cool and pick up heat, but will never again get circulated through the
engine so it has "forever" to dissipate its heat.

A real engine is somewhere in between these to limit cases, therefore it
is reasonably logical to expect some slope that connects the
steady-state oil temperature of the one limit with the other. I don't
think it reasonable to believe that both steady-state temperatures will
be the same and thus have a zero-slope line in between. This is what
would have to be the case for the oil temperature to be completely
independent of the amount of oil in the engine.


Matt


Matt

On Thu, 02 Sep 2004 16:56:03 -0400, Matt Whiting
wrote:

John_F wrote:

There are two terms in this temperature equation.

One is steady state thermal transfer rate. If you generate a
quantity of heat say one BTU then it will raise one quart of oil X
degrees F. If you add two times as much heat to the same oil it will
raise the temperature of the quart of oil 2X degrees F.
At any given RPM the oil pump will pump "Y" quarts of oil per minute
whether you have 2 quarts or 10 quarts in the sump since the pump is a
gear positive displacement pump. This means that the oil will carry
of the SAME amount of heat per minute if the temperature delta is the
same. If you want to get rid of more heat then you have to pump the
oil faster or heat the oil hotter to get a larger delta temperature
difference. This is the steady state condition. This is the condition
the engine is in when the temperature gage quits moving up.

Or you start with cooler oil in the sump. This is what, I believe, will
happen when you have more oil in the engine. The oil has a longer
residence time in the sump and contacts more surface area of the sump
through which it may dissipate heat.

You always START with COOL oil. It just does not stay that way long.

Let's run a "thought" experiment at the limits. Let's assume that the
oil level is so low that no oil is ever in the sump. The oil pump pulls
it out just as fast as it comes in, just short of the point of sucking
air. I realize this isn't possible in the real world, but that is why
this is a thought experiment.

This is done all the time. It is called a DRY sump and is used on many
aircraft. As far as I know ALL radial use a dry sump.

In this case, the oil will get very hot
as it is constantly being circulated through the heads which are one of
the hottest parts of most engines. The oil has very little opportunity
to reject heat in the coolest part of the engine, the sump. The
equilibrium temperature will be rather high.

Nope! That is what the oil cooler is for.

Most oil sumps on Lycoming engines are flat bottomed with vertical
sides. Look at the ratio of the surface area of the bottom to the
sides. It is at least two to one. Aluminum conducts heat MUCH
better than oil so any oil that runs across the bottom of the sump
will get cooling via the bottom and by conduction through the aluminum
up the sides. Most of the heat does NOT go out through the sump but
through the oil COOLER. If you have enough oil in the sump to cover
the oil intake it will be cooled by the bottom surface of the sump.
If you do not believe that cover up your oil cooler and see what will
happen to the oil temperature.

Now take the other extreme. The oil sump has infinite capacity so the
oil starts out at the same temperature regardless of how hot the hot
parts of the engine are. The oil will enter the oil pump relatively
cool and pick up heat, but will never again get circulated through the
engine so it has "forever" to dissipate its heat.

If you have an infinite oil cooler that will cool the oil back to the
original temperature it still does not matter how much oil is in the
sump.

A real engine is somewhere in between these to limit cases, therefore it
is reasonably logical to expect some slope that connects the
steady-state oil temperature of the one limit with the other. I don't
think it reasonable to believe that both steady-state temperatures will
be the same and thus have a zero-slope line in between. This is what
would have to be the case for the oil temperature to be completely
independent of the amount of oil in the engine.

There may be a small slope but it may be positive not negative since
the extra oil gets tossed around more and heats up due to mechanical
splashing.

Applying some simple math:
Since 90 % of the heat that is carried by the oil is lost in the oil
cooler not the sump you would expect that if the quantity of heat
carried away by the sump doubled the oil temperature would only drop
50% of 10% which is 5%.




Matt


Matt

Here is an actual thing that happened to me.
I loaned one of my lawn mowers to a neighbor. He misread the oil
level dip stick and added half a quart too much oil. The extra oil
covered the bottom of the crank shaft and the bottom of the piston.
The extra splashing caused the oil to get so hot that the crank case
melted the ignition kill wire that went to the key switch grounding
out the ignition which killed the engine before more damage was done.
This wire had lain on same spot of the crankcase for hundreds of
hours and never melted until the oil level was over filled.

More oil caused the engine to get hotter not cooler.

John

Mike Rapoport wrote:

It doesn't matter how long the oil is in the sump, it only matters how much
heat is lost from the outside of the sump to the air. Do we agree that the
only variable is how great the temperature diffference is between the
surface of the sump and the air? If so, it doesnt matter how long the oil
is in the sump. For your theory to be correct, doesn't the surface of the
sump have to be hotter? Also have you considered that on many engines the
sump is picking up a lot of radiated heat from the exhaust and that there
isn't a lot of airflow over the sump? The outer surface of the sump may
actually be hotter than the oil.

Yes, we agree that what matters is the heat lost from the sump to the
surrounding air, although I'd say more specific to our discussion is the
heat lost from the oil to the sump to the outside air.

Again, I'd add a second variable and that is the difference between the
temperature of the oil and that of the sump and then the sump vs. that
of the air.

Yes, my theory is dependent on the assumption that the temperature of
the oil entering the sump exceeds the temperature of the sump walls
themselves. If not, then the oil will not contribute at all to cooling
the engine, except in the obvious and trivial case of preventing the
friction that would otherwise heat the engine to failure within seconds.
It would likely still contribute to equalizing the temperature of the
engine by moving heat around as it flows, but as you say, it would not
contribute to any net cooling of the engine if your assumption is correct.

I honestly don't know if the sump is losing or gaining heat in net in
the engine compartment. I suspect the sump is the coolest part of the
engine, but I have no data either confirm or deny that premise. I did
some searching the other night and couldn't find anything substantial
regarding the temperatures of relative locations of an aircraft engine
or anything about the effect of more of less oil on engine cooling
capability.


Matt