Oil pressure drop on hard braking. My answer with Pics

[cstreit]

A common problem on early 911's, especially race cars is that the oil pickup runs dry under hard braking and left hand turns. The oil pump pickup is on the rear left of the tank. As such, it can run dry during these maneuvers.

Some vendors offer remote oil tanks that cost thousands of $$$ for remote tank systems, others ask $400 for modification of your existing tank.

Rather than just shell out a pile of dough and learn nothing, I embarked to take that $400 and do it myself. This has not been tested yet, but I will post my results next weekend.

Here is the new MIG welder I bought. It runs on 110 and has plenty of current. For most sheet metal it runs at 100% duty cycle. I'm running a 75%/25% mixture of Argon and Co2. The welding cart was my first project.



So the goal was to slow down the oil passage to the front, without restricting too much flow. I made a small baffle out of thin gauge steel. It sits about 1/2-3/4" off the bottom of the oil tank, and is about 2" high. The notch in the top is for the baffle to clear the oil return tube inside the tank. The holes will most likely be non-functional and are not really necessary as plenty of oil will flow over and under the baffle.




So I notched the tank right about where I wanted it, with some quick and dirty calculations as to how much oil the pump would use under the 3-4 seconds of hard braking, leaving some extra capacity.



I then tack welded the baffle in place on a low amperage setting. I increased the setting slightly to get good metal flow and penetration. This weld and protruding sheet metal was ground down to about 1/8", then a final bead for a little more structure.




I then checked it for leaks (none!) and covered it with a layer of high-temp silicone to help prevent rust. THe tank was re-painted and Voila!




A few notes:

1. THe tank was cleaned VERY VERY good after each step to ensure no residue was left inside.

2. The copper coating on the tank must be sanded or ground off in order to get good weld adhesion. I thought I read somewhere that the welds will not stick to copper?


Anway, I hope it works. It cost me the same as the modification offered by other places AND I got to keep the welder. (Already finding it very handy.

Chris

[fogcity]

Did this all come to you in a dream? (I'm only jelous, for I don't have the courage you do.)

[Eugene at Pelican Parts]

Originally posted by cstreit
A common problem on early 911's,

Until which model year was this problem corrected?

Funny how Porsche designed a dry sump system that partially defeats its own purpose.

[Wil Ferch]

how 'bout this...
On a separate thread, it was mentioned that Porsche went with smaller ( more restrictive) cam oil line connections ( on the engine block). The old ones had 6mm holes and the newer version had 2.5 mm holes. Steve Grosskemper of San Diego ( ?) region tested this on various cars and reports good oil flow to the cam lines.
Also, this mod....has the supposed added benefit of less foaming , and keeping the oil level in the tank more constant. I hope to get this mod complete before a DE at Watkins Glen next week, but it may not happen that soon. Any others out there with reports ??
This would be a cheap and simple "first try" on this problem, without resorting to the kind of work explained here.
---Wil Ferch

[cstreit]

Well not so much in a dream but just by thinking about what it would take and talking to Kelley Moss and some other Porsche teams.. Believe me I was pretty darn nervous about it and did about 30-40 practive welds on the same gauge steel before attempting. I literally welded the crap out of some scrap steel!

The smaller orifice may help, but I wonder if it really increases the level that much... Something to try next if this doesn't work though!

I'm not sure what years in total were affected by this, but it's any 911 with the oil pickup in the rear of the tank. I've had several people walk away in disbelief that this could happen... But then when to test it and saw the same thing. Run the car up to a decent speed and get on the brakes as hard as you can for 4-5 seconds. Make sure there are no curves or traffic, then watch your pressure gauge.

[HawgRyder]

Chris...did you only weld on one side of the tank?
Can the piece of metal bend inside?
I agree with the idea...but might have done a "through" piece.
I hope you have good luck with it.
Bob

[Dutchie]

Chris i talked to some guys that track their car about this, their solution was a Carrera 2 oilpump in the block itself. Would be great if your solution works as well

[cstreit]

Bob, it is welded on both ends for just that reason. It probably would haev been okay, but I figured better safe than sorry. I just didn't show both sides.

Ed, the issue is that the oil pump itself sucks air, so I don't think additional efficiency would help.

[ttweed]

Originally posted by Wil Ferch
On a separate thread, it was mentioned that Porsche went with smaller ( more restrictive) cam oil line connections ( on the engine block). The old ones had 6mm holes and the newer version had 2.5 mm holes. Steve Grosskemper of San Diego ( ?) region tested this on various cars

Steve is my mechanic and I have this mod on my '73 street/track car. It works. Article is archived here
Executive summary:
- The new fittings decrease oil to the cam housings and decrease oil foaming.
- The decreased foaming allows the scavenge oil pump to transfer oil out of the case and into the storage tank much faster. This in turn keeps the oil-tank level more consistent and causes the oil level gauge to react quicker.
- Less oil foaming will lead to less consumption of oil through the engine breather system.
- The smaller orifice creates higher oil pressure at the main and rod bearings as well as at the piston squirters. The increase we noticed varied from 10-20 psi.

It especially helps the low oil pressure at idle which most Porsches exhibit. If it keeps the oil-tank level higher, with less foaming, it will also help prevent the low-level condition which can cause cavitation of the oil pump through sloshing in the sump tank under g-loads.

TT

[cstreit]

Got a part number for those?

What are the orifice measurements? I want to check to make sure I don't already have them before i order.

Thanks!

[RickM]

Chris,

You heard right...this type of weld wont stick to copper. In fact copper is used as a protective backing when welding up holes.

BTW, what brand welder? Lookjs like a Miller, but not for that price.

[cstreit]

CHicago Electric....

It's not a Lincoln or anything, but for the amount of work I do, works well and parts are available at the local BOC gas place.

[Chuck Moreland]

Looks like it should work. I might have made the baffle a bit taller though.

Any interference with the oil level sender?

Another option might be to re-locate the oil pickup to the front of the tank.

On the cam oil line mod, I wonder how a reduction in foaming was confirmed?

[cstreit]

Chuck,

In order to make it taller I would have had to do a lot more major surgery. The oil retun line from the front fitting up to the filter runs across the bottom of the tank, about 2.5" up. I would have had to slot the baffle even deeper, then cut even deeper still slots in order to maneuver the sheet metal around.

For the first go-around I wanted to minimize the amount of cutting, etc...

If this doesn't work I will most likely add another piece of sheet on the top of the existing one, an over the oil return effectively making it taller and reducing the amount of fooling with the oil return line.

I figured taking the minimalist approach when embarking into the unknown was best. What I didn't want to do was restrict oil going back to the pickup.

[Wil Ferch]

The orifice plugs ( "cam oil fitting") are easily identifable from the outside. The old version ( with 6 mm hole) looks simply like a fat hex head. The newer version ( with 2.5mm hole) has a narrow band cut along the circumference, which makes it look like two hex nuts stacked one-on-the-other, with a gap in between.
Old number ... 901.105.361.00
New part .......901.105.361.01

About $5-6 each ( two needed). Sure beats all this tank welding if it works.

The newer cars that had this from the factory, around 1991....also used a different banjo bolt that had three, 4.5 mm holes in it, versus the old bolt that had four, 3.5 mm holes in it. There isn't much effect with this change, but if you want to be correct, you'd use the new banjo bolt too ( new is 999.134.018.02..... old is N.021.0563.3). Oddly, the equivalent area of the old four-hole unit is 38.5 mm2, whereas the new, three hole unit is 47.8 mm2. So , as a pair of parts, the new fitting is more restrictive, fed by a less restrictive banjo bolt....and vice versa for the old pairing, using a less restrictive orifice fed by more restrictive banjo bolt. Odd. The fitting is the determining piece, however, by a long shot...because the 6mm hole has an area of 28.2 mm2 ( 2.5mm hole has an area 4.9 mm2).

---Wil Ferch

[cstreit]

Ordered two. Thanks.

[304065]

Chris,

CANT WAIT to see the results of this. I got the idiot light all over the place at Watkins Glen this past week (DE thurs, Club Race Fri-Sat-Sun). I have an adjustable sender that turns on at 20psi.

I too have seen the modified tank, it looks good, but a time-consuming thing to send your tank out.

Did the oil fittings help you?

I'm hauling out to Gingerman for the Club Race over July 4th. Will you be there?

[cstreit]

John,

Initial results look good. I snuck a few brief peeks at the oil pressure guage under braking this weekend and didn't notice a drop that didn't correspond with RPM's, however untill I can get it back to Road America (3-4 weeks) I won't know for sure. The long 5th to 2nd gear braking zones there are where it is really apparent.

I too need to put an oil pressure light in!

Chris

[cstreit]

I did, however, solve the oil blowing problem. Reset the rings after a teardown and get no blowby at all. Since I made multiple changes I couldn't pinpoint a solution, but I'm reasonably sure it was re-seating the rings...

Cam oil lines I suspect make a small incremental difference.

[Joe911]

I can verify that the oil line fitting makes a notable difference - probably 15lbs at 3500 and 10lbs at idle.