At 10:50 AM 5/22/02 -0600, Bullwinkle wrote:
><<<snip>>>
>my MGA lost oil pressure completely at about 7000 rpm because the oil pump
>was cavitating.
><<<snip>>>
>
>Okay, that comment has me thinking, terrible thought that is.
>
>Where does the excess oil from the relief valve go to? Into the sump or
>back into the pump? On the B series engine where does it go? On the XPAG
>engine where does it go?
Any oil not going through the engine oilways goes past the pressure relief
valve and directly back into the sump. An MG B-series engine in tip top
condition with all new bearings will only pass about 12 quarts per minute
through the oil ways at 60 psi, which can happen at engine speed as low as
1500 rpm (when it hits relief pressure). After that any and all additional
flow from the pump at higher speed goes over the relief valve back to the
sump, and the energy used to pump that excess oil is instantly turned into
heat. An engine with worn bearings and more clearance in the bearing
journals may be able to pass twice as much oil through the engine, and
therefore will not get to relief pressure until about 3000 rpm.
>Was the effect of raising the oil pressure on racing engines actually to
>decrease the flow rate at high RPM's and therefore reduce the oil
>cavitation problem in the pump? ....
No. Increasing the flow can actually result in cavitation of the oil pump,
which happens from flow restrictions on the inlet side of the
pump. Increasing the pressure setting of the relief valve on the output
side of teh pump has nothing to do with the cavitation problem.
And I don't know why anyone thinks that oil pressure any higher than about
80 psi can be beneficial. As the supply pressure increases the oil flow
through the engine will increase slightly, but not much. The pressure
increases exponentially with the flow rate. In laymans terms that means it
takes a large increase in pressure to induce a small increase in flow. And
above 80 psi the increased pressure begins to have some detrimantal
effects, like eroding worm holes through the soft crankshaft bearings from
excess oil velocity, and increasing wear on the oil pump drive gears, and
increasing heating of the oil. For these reasons, in a stock engine you
should not boost the pressure too high, and you should not install a high
volume oil pump.
Certain specially prepared race engines may have intentionally increased
clearance in the crankshaft bearings to promote increased oil flow for
better cooling of the bearings. In this case the high volume oil pump is
desireable to maintain sufficient oil pressure at low engine speed. At
higher speed the excess flow over the relief valve will still generate more
heat in the oil, but a larger oil cooler may take care of that
situation. Racers are also not terribly concerned if the oil pump drive
gears wear out after 20,000 miles of hard use, and they are also likely to
install new bearing shells every few thousand miles, or in some cases after
every race.
For general information, the actual instantaneous pressure on the oil film
in the connecting rod big end bearings at high speed is far in excess of
the pumping pressure. The mass of the connecting rod whipping around at
high speed creates very high centrifugal force on the big end
bearing. Think in terms of a few hundred psi here rather than tens of
psi. It is the thin oil film itself that supports this high load on the
bearings, not the pumping pressure. The pumping pressure effectively keeps
the oil circulating so that the oil does not overheat inside of the bearings.
When the oil presure fails at high speed (several thousand rpm), the oil
stops flowing, followed by immediate overheating of the oil inside of the
bearing journals, followed by breakdown of the oil film, followed by high
friction heating of the bearing surfaces, followed by melting of the white
metal facing material on the bearing shell, followed by the white metal
running out of the sides of the bearing journal like melted candle
wax. This can happen in a matter of a few seconds at very high engine
speed. BTDT, too many times. If the oil pressure and flow then returns as
the engine slows down, the white metal can cool and solidify before the
engine comes to full stop, which results in a continuing running engine
with a very bad knock.
In cases when I was lucky and got off the throttle quick enough there was
still a little white metal left in the bearings, and the crankshaft wasn't
damaged, and a new set of bearing shells put it back in order. In cases
when I didn't get off the throttle quick enough there was bare steel from
the base of the bearing shells in contact with the journals on the
crankshaft, which makes for some very nasty scoring of the crankshaft
journals. Now I have not done this next thing personally, but in cases
where you keep your foot in the throttle a little longer at high speed
after oil pressure failure you can have spun bearing shells and/or
connecting rods exiting through the sides of the engine block.
A more common occurence is the case of the DCD (Dumb Current Driver)
failing to notice a low oil level and a subsequent loss of oil pressure at
moderate engine speed. The first wake up call is the knocking noise coming
from the engine, and a possible rise in engine temperature. If the DCD
fails to heed this warning and continues to drive at moderate speed for
just a minute or two more with no oil pressure (just wanted to get the car
off the next exit ramp), then another interesting thing will happen. When
the engine finally comes to a stop, either through loss of power of from
manual shut down, there will still be no oil in the bearings for cooling,
and as the engine cools down gradually the white metal in the bearings
(what little may be left) will effectively solder itself to the crankshaft
journals, resulting in a firmly siezed engine, in which case it is likely
that no one would ever be able to extricate the crankshaft. So don't do
that. If you should ever notice a loss of oil pressure, shut the engine
off immediately and walk if you have to.
Barney Gaylord
1958 MGA with an attitude
http://www.ntsource.com/~barneymg
/// or try http://www.team.net/cgi-bin/majorcool
/// Archives at http://www.team.net/archive
|