I have been following the discussion about the location of the turbocharger 
oil return line in vanagon installations for some time but have been too busy 
to join the discussion. My interest devolves from my effort to develop my 
Turbocharged 3A Audi 2.0 L engine for vanagon installation. I have a number 
of key documents on the subject from the Motortechnische Zeitschrift (MTZ) 
and Automobiltechnische Zeitschrift (AZT) journals. As part of this post, I 
am attaching a set of drawings covering the end on and transverse views of 
the vanagon version of the normally aspirated and the European turbo diesel 
engine. I also included the Vanagon TDi drawings posted on Michael Sullivan’s 
web site, and drawings of the transmission and engine package as mounted in 
the ’82 4 speed 2WD version. These drawings are in JPG format and have been 
zipped for the computationally challenged still using IBM style computers. I 
realize that the attachment will be strippede from the list posting, so 
should anyone want to see these, send an email request.

In addition to the drawings, I have several spare diesel (603A) oil pans, oil 
input spouts, dipstick assemblies, oil pumps, oil pump pickup assemblies, 
etc, etc. I have therefore measured and re-measured the whole system, 
resolved several apparent inconsistencies in those drawings. I have imported 
the scanned drawings into my CAD programs and added several notations to show 
actual static oil levels, in order to properly address the oil return 
placement issue. This topic should be of critical import to any of you who 
are contemplating adding a turbo charger to a diesel or inline 4 engine 
mounted at the standard 50 degree from vertical vanagon diesel angle.

Now for some discussion:

The first thing that these drawings show is that the engine and transmission 
package for the 4 speed early vanagon diesels is mounted at a downward 
sloping angle of 4 degrees (from the transmission mount to the crank pulley). 
The engine package is also displaced towards the front of the vehicle canting 
the axle half shafts forward relative to the WB half shaft position. 

This downward slope is also shown in the contour and shape of the diesel oil 
pan. The base of the pan is designed to be level (horizontal) and therefore 
the front (pulley side for me) of the pan is more shallow than the rear 
(flywheel) side. This means that the static oil level will be higher into the 
block at the front than at the rear. In the drawing of the normally aspirated 
diesel, I have shown the static oil level corresponding to the max and min 
marks on the dipstick. Measurements show that the static oil level is just 
52.4 mm (Max) above the oil pan base. Therefore on the front of the engine 
the static oil level is well above the oil pan gasket and intersects the 
center of the lowest crankshaft seal retaining bolt. At the rear of the 
engine, the max oil level just gets to the top of the oil pan seal. Moral: 
loose pan with full oil gives slime on the floor (leaks always!). This front 
to back pan depth variation is also shown in the drawings of the oil pan in 
the transverse views. Here though, the drawing suggests that the front of the 
pan is deep and the rear is shallow.

This leads to some confusing 2D drawings. A first look at the end on drawings 
also suggests that the pan drawing is incorrect. (too deep at the front). 
Upon closer inspection one realizes that the drawn edge of the pan refers to 
the rearmost edge (in the end on drawings). In the vanagon NA-D drawing, I 
have also indicated the location of the front pan edge. Consequently, the oil 
level is further below the crank centerline as one moves from front to rear 
(pulley to flywheel). 

Looking at the vanagon euro TD drawings, the transverse view shows this depth 
variation now goes from shallow in front to deep in the rear. This symmetric 
reversal suggests that the turbo diesel is mounted at 4 degrees elevating 
from transmission mount to crankshaft pulley (exact opposite of the NA case). 

BTW, the actual volume in the TD and NA pans are quite similar. By VW 
documentation, the oil capacity of the ’82 models was 4 liters with filter 
and 4.5 liters for the later TD models.

As to the location of the turbo oil drain return line, as Mike Snow had 
noted, the vanagon TD drawing shows a nearly horizontal oil line going 
smoothly from the turbo to the oil pan and entering well under the static oil 
level at that point. Indeed, the euro vanagon TD turbo oil return flange is 
just below the static oil level. So, the tube is full of oil and represents a 
static head pressure of at least 30 mm of hot 30 weight oil (mean viscosity) 
resisting the flow of oil to the pan. 

Studing the drawings of Mike Sullivan’s vanagon TDi installation, shows that 
the turbo is now significantly higher than the euro TD install, and the oil 
outlet flange is now above the static oil line. But, the outlet tube first 
takes a 90 degree bend going up and away from the pan base, then another 90 
degree jog down, ending in a boss entry into the block sidewall that is well 
above the crankcase oil level. The entire section of this tube up to the exit 
point in the block sidewall must be filled with oil representing a larger 
static head back pressure than the euro TD case. 

Finally, in this discussion of factory solutions, last weekend I had the 
opportunity to study Mark Drillock’s 1.9 L Turbodiesel installation in his 
’82 Westfalia. Here, the engine uses the exhaust manifold (and intake) from 
the Golf/Jetta installations. There is plenty of room in the engine bay. He 
also has the engine mounted at the 50 degree angle. In this version, the oil 
drain entry to the block is also in the sidewall (appears to be at the same 
point as the TDi install), and therefore above the static oil level. However, 
because the turbo is mounted significantly higher, the turbo oil drain line 
runs smoothly downward to the block entry point. This is the only 
installation that actually follows the traditional turbocharger mounting 
wisdom. 

Returning to my 3A block, I have mounted the Garrett T3 turbo on an adapter, 
then to the Quantum TD exhaust manifold. The oil drain exit flange on the 
turbo is at nearly the same level as the euro vanagon TD installation. I have 
indicated the turbo mount position on the vanagon NA-D drawing. Again, the 
static oil level is above the gasket to the turbo return flange. So the 
dilemma, where to mount the return line?

According to Corky Bell’s book “Maximum Boost”, the return line must never 
enter below the static oil level, and there should be no bends in the line to 
enable trapped oil. Trapped oil or a high backpressure in the return is 
supposed to lead to loss of the turbo shaft seal, and bearing coking problems 
(oil in contact with very hot shaft decomposes to sludge, then coke). He 
matter-of-factly says that the low turbo mount situation calls for a 
scavenger pump, as used by Porsche on the 911 turbo’s and a variety of racing 
and motorcycle applications. 12 Volt scavenger pumps abound but are a bit 
pricey ($300 to $800). 

I am considering several options:

1). Mount the return line to the pan in the VW factory TD way, get on with 
life, and begin saving to buy remanufactured turbo cartridges from ebay in 
bulk.

2). Fly the Michael Sullivan TDi way with a high block entry and a looping 
hose. Life also proceeds and the saving club for steady turbo replacement 
begins as well.

3). Fabricate an intermediate oil chamber with two large 1 inch ID pipes 
going into the oil pan sump at front and back for good circulation. Add a 
large vertical line from this side chamber to the crankcase ventilation 
circuit to relieve any pressure build-up. Then run a 0.75 inch ID line to the 
turbo minimizing the path. According to Bell, the turbo axis can be tilted 
+/- 30 degrees from vertical with no problems. This would help shorten the 
path and slightly reduce the height differential.

4). Mechanical scavenger pump. I have several VW/Audi rotary vane power 
steering pumps sitting unused, complete with mounting brackets and pulleys. I 
am now wondering if I could hook up one of these units as a hydraulic 
scavenging oil pump. Connect the suction line to the turbo oil outlet return 
and the outlet of the PS pump would go to the unused oil dipstik tube. If the 
output pressure is too high I could change the orifice in the pressure 
control valve thereby increasing the amount of bypass flow and lowering the 
output pressure. Presumably, the oil input line will always be supplying oil 
flow through the bearing during operation. The PS pump should not quite run 
dry, but would generally be inhaling heavily aerated oil (foam). The flow 
rate on the PS pump is probably much higher than the oil return flow, so I 
worry about running it dry. Squealing PS pump is not good. 

5). Buy 12 volt scavenging dry sump pump from Summit racing. Probably bad. 
Admits defeat and may trigger uncontrollable urge to convert engine to a dry 
sump system. 

I will gratefully entertain comments and alternative suggestions. Anyone know 
of other factory engines using belt driven oil scavenging pumps?

My apologies to the disinterested on the length. But you saw the poster ID 
and therefore knew it was long and verbose! Remember the delete key.

Frank Grunthaner