Coking the shaft on a turbo?
id fullname
dealer's mechanic says is responsible for the turbo on my '86 Subaru XT
chewing itself to pieces within months of buying it in April '92. 90% of my
driving is the short 4 mile commute to and from work, although I have made
one long trip from Indiana to Wash, DC and back. Anyway, the turbo worked
fine for about three months, generally kicking in once or twice on each
little commute. Then in October, I began to hear a high-pitched whine from
the engine that seemed to coincide with the operation of the turbo. This
changed to a horrible grinding sound in less than a week, so by the time I
got it into the shop, the turbo was shot.
The mech says the turbo exceeded tolerances and contacted its housing due to
coking of the shaft, described to me as a rapid heating or cooling upon
startup and shutdown that weakens the shaft and eventually leaves it bent.
He put in a rebuilt one, and I've been babying it ever since, including
complying with a technique he suggested of letting the car idle for 60
seconds upon startup and shutdown, to allow the oil to properly lubricate
the turbo shaft. The car also has a 10 increment temperature gauge, and I
wait until it stabilizes in the midrange before driving or shutting down.
OK, here's my question? Is this idling procedure helpful? It adds two
minutes to every trip, and uses more fuel than if I didn't do it. Still,
the rebuilt turbo has survived so far, and if it's because of the new
technique, it's worth the hundreds of dollars it would cost to keep buying
replacement turbos. Or has it lasted this long because I don't use the
turbo as often on the road.
Thanks in advance for any advice or opinions, posted here or e-mailed to me
at james.m....@nd.edu.
Jim Jenista
bl...@inland.com
Coking of the shaft is not the same as a bent shaft due to some
mysterious weakening. Coking is the oil getting baked on the shaft
by heat soaking back from the turbine to the oil cooled shaft after
shutdown. When you shut your engine down, there is no cooling oil
flow to remove the heat from the shaft, so the remaining oil
on the shaft gets burnt to carbon, which is not a good bearing.
Your mechanic is correct in advising you to idle your engine
for a minute or two before you shut it off. This extra minute
provides cooling oil to the shaft while the hot turbine can get
rid of the heat during a no load idling condition. The heat of
the turbine increases with engine load and is greatest at long
durations of wide open throttle. The worst thing you can do is
shut your engine down quickly after a screaming blast down the
highway. Not only do you get heat soak to coke the oil,
but you have no oil pressure to support the shaft bearing while
the shaft is still spinning at relatively high rpm. I had a
Saab 900 turbo with no problems after 110,000 miles; just idled
it for a minute on starting a cold engine, drove off slowly under
low loads 'til engine was warm, and idled it a minute before
shutdown. I've got an air cooled Yamaha 650 turbo bike with
20,000 miles also that I use the same procedure on.
Alex D Rodriguez
>I would like some info on a phenomenon called "coking the shaft" that my
>dealer's mechanic says is responsible for the turbo on my '86 Subaru XT
>chewing itself to pieces within months of buying it in April '92. 90% of my
>driving is the short 4 mile commute to and from work, although I have made
>one long trip from Indiana to Wash, DC and back. Anyway, the turbo worked
>fine for about three months, generally kicking in once or twice on each
>little commute. Then in October, I began to hear a high-pitched whine from
>the engine that seemed to coincide with the operation of the turbo. This
Turbos are suposed to have a high pitched whine when they are working. Most
large trucks have turbos, next time you are next to one at a stop light
listen and you will hear the turbo whine when he takes off.
>changed to a horrible grinding sound in less than a week, so by the time I
>got it into the shop, the turbo was shot.
>
>The mech says the turbo exceeded tolerances and contacted its housing due to
>coking of the shaft, described to me as a rapid heating or cooling upon
>startup and shutdown that weakens the shaft and eventually leaves it bent.
That does not describe what coking is at all. Coking occurs to oil when it
is heated beyond a certain point. Synthetics have a high heat tolerance
before they coke and that is why they are recommended for turbo cars. Coking
turns your nice slippery oil into what I would describe as black sand. So
when you get this in your turbo the shaft is riding on sand instead of oil
and it eats up the shaft in a hurry. After a while there is enough play in
the shaft that the vanes of the turbo come in contact with the housing causing
lots of damage.
>He put in a rebuilt one, and I've been babying it ever since, including
>complying with a technique he suggested of letting the car idle for 60
>seconds upon startup and shutdown, to allow the oil to properly lubricate
>the turbo shaft. The car also has a 10 increment temperature gauge, and I
>wait until it stabilizes in the midrange before driving or shutting down.
>
>OK, here's my question? Is this idling procedure helpful? It adds two
>minutes to every trip, and uses more fuel than if I didn't do it. Still,
>the rebuilt turbo has survived so far, and if it's because of the new
>technique, it's worth the hundreds of dollars it would cost to keep buying
>replacement turbos. Or has it lasted this long because I don't use the
>turbo as often on the road.
Yes the idling is very helpfull. It keeps the oil running through the turbo
until it spools down to a reasonable speed. This in turn helps keep the
oil temperature below the oils coking point. I don't think that not using the
turbo is what made it last longer that your previous one. I've owned two
turbo cars. The first a Renault Fuego had its turbo die at about 60k miles.
I believe it was for two reasons. The first is that the previous owner did
not let the car idle, just as suggested by your mechanic, and that the turbo
was not water cooled. My current car is a Dodge Omni GLH Turbo. It has
over 100,000 miles and the turbo is running fine. This is probably attributed
to the fact that both the previous owner and I let the car idle and that the
turbo has water cooled bearings. The coolant running throught the turbo
core helps keep the oil below it coking point. If you are really interested
in a hassle free, but expensive, way to make your turbo last longer check out
the advertising section in the back of Auto Week a company called HKS sells
a device that will let your car idle for a as long as necessary to get the
turbo temperature below a certain point. I have also heard of another
product that was basically an oil reservoir that stored a certain amount of
oil under pressure and when you shut your car off released it past the turbo
bearings to cool them down. Hope that helps.
>
>Thanks in advance for any advice or opinions, posted here or e-mailed to me
>at james.m....@nd.edu.
>
>Jim Jenista
----------------
Alex Rodriguez
Sean Reifschneider
>dealer's mechanic says is responsible for the turbo on my '86 Subaru XT
>chewing itself to pieces within months of buying it in April '92. 90% of my
Coking is what happens to oil when it gets real hot. The turbo, espeically
when driving at highway speeds, gets REAL hot. If you just shut down the
engine after hard driving the oil doesn't have a chance to cool down the turbo.
It then cokes (turns into a sludge like substance).
I have the same engine (well, mine is a '91) and Subaru recommends running it
for 60 seconds after highway driving. Driving it gently in town keeps the
turbo cooler, and I only let it cool for 10-30 seconds in town. As far as start
up time, I only let it idle for 30 seconds the first time of the day I drive it
or if it's real cold. Then I just drive it gently until it warms up. I've not
had a turbo problem in the year that I've had the car.
Also, I use Synthetic (Mobil 1, but switching to Valvoline) oil which doesn't
tend to coke in the turbo, it sort of flakes from what I've heard.
Sean
--
"If you were happy every day of your life, you wouldn't be a human being...
You'd be a gameshow host." -- Heathers
Sean Reifschneider, Supreme hack
Tommy Reingold
about idling a minute before driving? My manual recommends ten seconds
of warm up.
--
Tommy Reingold
AT&T Bell Labs, Holmdel, NJ
to...@boole.att.com or att!boole!tommy
bl...@inland.com
> I understand the benefit of idling after driving but what's all this
> about idling a minute before driving? My manual recommends ten seconds
> of warm up.
> --
> Tommy Reingold
Oil has high viscosity when cold and doesn't flow readily thru
tiny oil passages. You need good oil pressure and good flow
to support the turbo shaft. Using a lot of throttle on a
stone cold engine is not a good idea, unless you have a big
bag of money q:}
Silver spoon, and a paper plate. How sad.
>I understand the benefit of idling after driving but what's all this
>about idling a minute before driving? My manual recommends ten seconds
>of warm up.
Depends on the oil presssure, really. Mine recoomends at least 20s
of warm-up, which is to allow the oil to reach the turbo at full
pressure, as there's quite a big pipe going up to the turbo, and
I also seem to have about a 1 minute oil pressure resevoir. It then
recommends not to kick in the turbo (low-load) until the engine warms
up. I also let it idle on a cool down cycle before stopping the engine
for anywhere between 20s and 2 minutes, depending on previous
driving.
Ron