Power Steering Pump as Turbo Oil Scavenge Pump

[FrankGRUN]

I'm trying to put together a belt driven oil pump to take (scavenge) aerated
oil from a Garrett T3 turbocharger oil return line and dump it back into the
crankcase. Because the turbo oil outlet is mounted about 1 inch below the
static crankcase oil sump level, this requires the scavenge pump approach.
While gear pumps for dry sump systems are available, the cost is high (excess
of $600) and many custom fittings and mountings would have to fabricated.
Electrical pumps are available, but are not rated to operate at the elevated
temperatures (as high as 275 F or 135 C). BTW, this is for a turbocharged 2.0L
Audi 3A Inline-4 engine mounted at 50 degrees from the vertical.

The Question: I have several VW power steering pumps on my bench complete with
mounting brackets. These pumps are vane pumps made to work with remotely
mounted reservoirs and have a pressure limiting assembly set at around 1200
psi. I need to know if I could use the power steering pump as a mechanically
driven hydraulic pump. The required flow rate is around 0.5 gpm, and the
viscosity will be equivalent to 10W-40 at 130 F/ 55 C. Any input would be
greatly appreciated.

My current plan is to build a 1 liter auxiliary sump with a fitting vertically
draining through a 3/4 inch tube into it from the turbo return flange. I will
include a deaerating screen in the top half of the auxiliary sump and provide
for a 3/4 inch vent line to go to the crankcase ventilation system. On the side
of the auxiliary sump near the bottom, I will have a 3/4 inch fitting, plumbing
the oil to the inlet of the power steering pump. I plan to send the output from
the power steering pump into the block through a fitting mounted below the
crankcase oil sump level to minimize aeration.

I also had two alternate options in mind. In the first, I would direct the oil
from the power steering pump to an external oil cooler, then to the crankcase
sump. In the second (possibly concerned about letting the pump go dry), I could
change the pressure relief spring, setting it to 35 psi and/or I could add
another line to the auxiliary sump connecting it to the main sump. If this line
would be around 0.25 inches in diameter, I thought the pump would always have
oil independent of the flow through the turbo.

Hair-brained scheme or? Appreciate any inputs.

Frank Grunthaner

[Turbodave]

Frank
You are on the right track I would say, I did mine a little different
though, I actually used a 3500psi "capacity" hydraulic pump driven off of
the crank with a gilmer belt [ to keep the side load down on the pump
bearings ], I used the this type of pump because it has to pull oil away
from two TO4's at the same time.
I did think of the power steering pump also but it was my second choice if
the hydraulic pump was going to be to much aggravation to build.
I built it [ just under the aggravation threshold :) ], but I think your
pump would be fine for what you are doing [ one turbo ] .
You could do some bench testing, it would need to pull about two liquid
gallons per min. to be able to keep up with the outlet side of the turbo.
You may want to build a tank under the turbo to catch the oil/foam so it
has a place to liquify again so the pump can pull it easier.
lowering the output psi to equal the input max psi would also be good to
keep things as even as possible on the pressure/flow side of things.
Keep in mind excessive vacuum on the turbo outlet could cause bearing
problems as well, if you think [ during your test ] that it has to much
vacuum you can put in a bleeder valve from a filtered open air source.
Good luck on your project.
let us know how you made out.
Dave.
ps: if you need a pic of what I have done for reference I can post it .

FrankGRUN wrote in message <20010828190951...@mb-cg.aol.com>...

[FrankGRUN]

Dave,

Really appreciate your comments. Would really like to see pics of what you have
done. I had figured the output of the T3 as around 0.5 gpm. I assumed that
number corresponded to a non aerated flow rate, but didn't know what to assume
for a conversion factor. Henceforth I'll use a factor of 4!

My plan was to put a small auxiliary sump underneath the T3. I was going to put
a screen separating the two halves of the sump chamber and fill the upper
portion with steel wool to deaerate the oil. I planned to vent the chamber to
the crankcase vent system to make sure that the PS pump couldn't evacuate the
turbo return circuit.

Thanks again,

Frank

[Turbodave]

I am not sure where to post the pics, is it ok to post on this ng...
Steel wool is a good idea to help re-liquefy the foamy oil, a small amount
in the bottom of the chamber should to it without causing any backup, I
would just have a concern that a small particle would come away and go
through the pump, maybe a flat screen filter in line before the inlet and a
tranni filter after to keep everything pure would be good just in case.
As you will see on mine I have just used large lines right out of the
outlets, but the pump I am using is pretty large as well.
I forget what the actual gmp #'s are on a single turbo, you want to be way
above that on the draw side, so more is definitely better in this case, you
can always do the bleed valve if it is to much.
Dave.

FrankGRUN wrote in message <20010829030329...@mb-mv.aol.com>...

[FrankGRUN]

Dave,

I don't think the NewsGroups can accept pictures, but try. Anyway, you can
email them direct, if convenient. I think the input que will support anything
up to 3 Megs.

As for the pump capacity, it appears that it is between 2 - 3 GPM at idle, and
well over 6 GPM at turbo relevant speeds (3000+ rpm). Interestingly, the best
estimate of the power consumed turning the pump under these conditions is less
than 0.5 HP. I'll add a screen to the input line to the pump.

Thanks,

Frank Grunthaner

[Peter Hill]

Why bother to pump back to the sump? Suck from lower drain tank
though an oil cooler with an electric pump. Feed the oil straight
though a filter and back into the turbo. Maybe put a independent
water cooling system on it too. Engine cooling system is effectively
upgraded as it will no longer have to dump the heat from the turbo.

--
Peter Hill

Can of worms - what every fisherman wants.
Can of worms - what every PC owner gets!

[David Algie]

Hey, the turbo has to have oil pressure as well, not just oil at low or no
pressure. You would have to use a high pressure pump and regulate it also.
Sounds like unreliability waiting to happen to me.
Dave Algie.
Peter Hill wrote in message <8eccqt441pto5v3uf...@4ax.com>...

[FrankGRUN]

Thought I would update the situation with what I've learned so far. First,
several discussions with engineers at Garrett (ne. Honeywell) on flow
requirements. They confirmed how critical it was to have a freeflow drain or a
scavenger pump. Big impact on turbo lifetime. Main issue is oil backup and thin
film coking on the turbo shaft. This problem is acute for the sleeve bearing
approach, much less so for the ball bearing designs. Garrett confirmed a flow
rate of 1.5 gallons/minute at 35 psi and engine speeds of around 4000 rpm
(specifically for a T3 running on a inline 4 with a 2.0 liter displacement).
This is in contrast with the generic statement from Corky Bell's book giving a
minimum flow rate of 0.5 gallons/min. Several engine builders emphasized that
"Garretts flow more oil than other turbos".

This 1.5 GPM is for the input unaerated oil. The actual output is anywhere from
2 to 4 times greater volume.

Conclusion, I need a pump with 3 to 6 GPM capacity. The addition of a small,
low capacity minisump to take the turbo return oil flow, a vent system to
eliminate pressure surges, and a screen before the inlet to the scavenge pump
will minimize the required flow rate.

As for the flow of VW power steering pumps, the picture is not good. According
to ZF, the standard power steering pump fitted for the VW only flows 0.7 liters
per minute (just about 0.2 GPM). The larger pumps used on the Audi 4 cylinder
have a rating of 7 liters per minute at idle and 4 liters per minute at speeds
above 1000 rpm. So, at least for VW power steering pumps, the capacity is
inadequate for for scavenger pump applications. Also a bit pricy - junkyard
pumps run about $35, plus $20 for the mounting bracket.

I have gone to the JS Barnes hydraulic pump (rated for 3.0 GPM at 1800 rpm).
It is mounted using the VW PS bracket and a v-belt pulley at 1:1. I'm also
fabricating a 1.5 quart minisump just below the turbo using a VW air-cooled
Beetle aluminum extended cast sump. This is drained through a screen and pumped
by the hydraulic pump. Output from the pump goes to an external oil cooler then
to the main sump.

Frank Grunthaner