When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
I've been looking for what way the oil that goes to the heads can travel inside the engine. The pictures below have been to great help. What is obvious is that there isn't any good way for gravity(without the scavenge pumps) to drain oil from the heads back into the sump. The green arrows(four small holes) are the main way for gravity to get it back there.( I think there is also bigger holes, where the top of the swirl pots are located, but they are located above the cassette drain holes. )This can be compared to the big cut outs just above which drains the cassette (rod and main bearings, sprayers etc). There is, as you can see marked by the red line, a small passage where oil can travel between left and right head, meaning that both pumps kind of work to scavenge the same oil, in parallel. If you are constantly turning hard there is only one scavenge pump working at a time as the other will run dry.
One idea can be to increasing the gravity drain back holes. I believe this will help keeping the sump level higher(avoid pooling) as this off-load/help the work of the scavenge pumps. The cavities for the cam chains is the only big place I've found so far that could pool any significant amount of oil.
Correct me if I'm wrong, I might be, I've just spent a few hours on this.
I've been looking for what way the oil that goes to the heads can travel inside the engine. The pictures below have been to great help. What is obvious is that there isn't any good way for gravity(without the scavenge pumps) to drain oil from the heads back into the sump. The green arrows(four small holes) are the main way for gravity to get it back there.( I think there is also bigger holes, where the top of the swirl pots are located, but they are located above the cassette drain holes. )This can be compared to the big cut outs just above which drains the cassette (rod and main bearings, sprayers etc). There is, as you can see marked by the red line, a small passage where oil can travel between left and right head, meaning that both pumps kind of work to scavenge the same oil, in parallel. If you are constantly turning hard there is only one scavenge pump working at a time as the other will run dry.
One idea can be to increasing the gravity drain back holes. I believe this will help keeping the sump level higher(avoid pooling) as this off-load/help the work of the scavenge pumps. The cavities for the cam chains is the only big place I've found so far that could pool any significant amount of oil.
Correct me if I'm wrong, I might be, I've just spent a few hours on this.
Nick, you are on the right track. The intermediate shaft channel, along with the chain areas (I call this the "Z") is full of oil, all the time. You need to consider what happens to this oil with respect to those drain holes under acceleration, which you will be doing 90% of the time on when you are on track.
You are wasting time and money replacing the scavenge pumps. Yours were functioning fine. They don't wear out. Any data you think you are gathering by watching the OP gauge is garbage and is worse than useless.
You can skip right to the end of this exercise by roughly calculating the internal air volume inside the engine. That's how much oil you need inside the engine to guarantee that there is sufficient oil in the sump for the pickup to remain submerged in all conditions. When the engine is running at high rpm, it can hide 15 qts of oil in its internal volume. I have data that proves this.
The crankcase environment is dominated by the air moving around, and how the chains are moving the oil, not so much by gravity. In other words, there is no guarantee that oil flows downward.
^^^^^^That's a pretty good idea , I wonder where you got that? Do that and lets see how it works on your car, mine is fine.
Originally Posted by Porschetech3
and there is some oil drain back passages that are too small ( about .250) IMO, these needs to be enlarged so the more oil flow can get back to the sump quicker.
Nick, you are on the right track. The intermediate shaft channel, along with the chain areas (I call this the "Z") is full of oil, all the time. You need to consider what happens to this oil with respect to those drain holes under acceleration, which you will be doing 90% of the time on when you are on track.
You are wasting time and money replacing the scavenge pumps. Yours were functioning fine. They don't wear out. Any data you think you are gathering by watching the OP gauge is garbage and is worse than useless.
You can skip right to the end of this exercise by roughly calculating the internal air volume inside the engine. That's how much oil you need inside the engine to guarantee that there is sufficient oil in the sump for the pickup to remain submerged in all conditions. When the engine is running at high rpm, it can hide 15 qts of oil in its internal volume. I have data that proves this.
The crankcase environment is dominated by the air moving around, and how the chains are moving the oil, not so much by gravity. In other words, there is no guarantee that oil flows downward.
Chris Cervelli
Cervelli Technical Service
Did you solve the drain back on your dry sump systems or just accounted for 15qts pooling? Why is the OP gauge garbage? And, what should we be measuring?
I measured my old pump, it is most definitely worn. In the pictures above you can also see the wear in the main pump body(alu piece on the right), it's worn away almost 0.1mm. Normally this kind of wear reduces pumping capacity.
Did you solve the drain back on your dry sump systems or just accounted for 15qts pooling? Why is the OP gauge garbage? And, what should we be measuring?
My data shows that with the current system, which includes some small mods to the engine itself to improve oil scavenging, the oil tank level (average level over the course of a lap) is around 1.5 qts lower than the level in the tank when the engine is idling. The system is doing a good job of scavenging the oil out of the engine and returning it to the tank. There are 3 additional scavenge points and the stock pump is repurposed to be a scavenge pump so there are a total of 4 scavenge pumps. The two stock scavenge pumps are still in place also, but I don't count those directly because their output has to be re-scavenged by the stock pump.
The stock OP sensor is very slow to react to pressure changes. The pressure drops are likely to occur when your eyes are on the road ahead of you, not watching the gauge. If you see pressure drops on the gauge while driving, you can be sure there are bad pressure drops happening. Also, the oil pump will create about .5 bar air pressure at high rpm even with NO oil in the sump, where guarantees that the idiot light will never illuminate at high rpm.
You should redraw the above diagrams with the horizontal lines tilted to show the influence of .25G in acceleration and 1G in braking. Show the lines for mid engine and rear engine applications. The drain holes there are not significant.
Regarding the movement of the oil within the engine, consider the case of a mid engine car in an accelerating right turn:
On the left side of the engine, the scavenge pump on the front of the head is dry due to the accel G forcing the oil into the rear of the head. No return to the sump is occurring.
On the right side of the engine, the cornering G pulls the oil into the intermediate shaft area, away from the scavenge pump on the rear of the head. No return to the sump is occurring.
There is another factor at work here. The stock oil pump is really big, and runs at 66% of crankshaft rpm. It pumps much more oil than the pair of scavenge pumps, which runs at 50% crankshaft rpm. Because of this disparity, the engine does rely on gravity to return the oil to the sump.
The sump shape is really bad too. I think they had to do this to get the necessary volume without reducing the ground clearance.
There is another factor at work here. The stock oil pump is really big, and runs at 66% of crankshaft rpm. It pumps much more oil than the pair of scavenge pumps, which runs at 50% crankshaft rpm. Because of this disparity, the engine does rely on gravity to return the oil to the sump.
I found this hard to believe. If the scavenge pumps was pumping slower than the rate that the oil arrives, we should all be pooling oil and running the sump dry. The gravity drain is as mentioned earlier, rather insignificant. Not plausible.
I measured the scavenge pump and it will pump around 42L/min at 7000RPM(crank RPM). I assume (I'm lacking any exact number here, please fill in if you have) that the main oil pump pumps around 70l/min at 7000RPM(crank RPM).
If we assume 2/5 of the oil goes to the heads(slightly more if running an oil feed IMSB) that is a flow of 28L/min. So, if both scavenge pumps work properly we have a scavenge rate of 84L/min which is an over capacity of 3 times.
I wrote earlier that my pump was worn but after further investigation it is badly worn out. Numbers I found would indicate a reduction of 30-40% of the pumping capacity. If we calculate that both pumps are toast(-40%) we still have 1.8 times the scavenge capacity, well enough if driving straight.
If we further look in to driving, we don’t drive straight all the time. And here is the interesting thing. If we assume a very twisty road where basically you never drive straight but corner hard all the time, then only one pump will scavenge at any time. If we only have 1.8 times over capacity(both pumps toast) that means we are running in to pooling since one pump is not able to scavenge all the oil by itself.
After running the numbers I’m more certain this is an issues of worn scavenge pumps. EDIT: If you have this problem when driving a stock car on road tires.
I've now put the car to test with both scavange pumps replaced. The right side pump (passenger side) is very easy to replace but the drivers side is very tight. It is possible to replace it with just taking the wheel off. The old pump can be disassembled in pieces and easily taken out but the new is built different and needs to go in in one piece. The trick is to file away some material as shown in this picture, then it's possible to get it in without removing anything:
Removed material
On my normal test drive I had 4 turns in which I usually could provoke oil pressure drops. They all had in common that the precursor was several right/left turn and then a strong right i.e no straight driving just before. I guess this caused pooling of oil because of the old/worn scavenge pumps. After changing both pumps I went for a test drive and only in one turn showed a slight drop. After the drive I checked the oil and it was only half full. Then came the second corona lock down here. Last week I was able to get out again, this time with full oil and now there is zero pressure drops. Rock steady.
I guess not everyone will have worn scavenge pumps but it's dead easy to take off the passenger side pump and check. My passenger pump had scoring and very loose tolerances, in the area of 0.12-0.15mm on the internal clearances and >0.25mm on the rotor-to-case clearances. On top of this scoring/pitting. The driver side was significantly worse compared to the passenger side. this is what my driver side pump case looked like:
Driver side scavenge pump scoring driver side scavenge pump scoring
Folks, this is an incredible thread that I hope will become active again and updated with additional info. For those of us that don't have the engineering/mechanical knowledge and experience of the contributors, we truly appreciate your insight.
Thanks to Nick in referencing it in De Jeeper's "Another One Bites The Dust" thread.
My next step is to implement more cooling... as can be seen below - hotter oil = more pressure instability.
Jason,
After reading the complete thread, it looks like it may be a an important preventative step that most of us haven't taken yet. I know with the AIM system you integrated, you have been measuring and ultimately traveling down this path.
At this point I certainly wish I had an engineering background to fully understand the details. But at least I can follow along with the concepts and potential application.
My next step is to implement more cooling... as can be seen below - hotter oil = more pressure instability.
I've come to the same conclusion, and purchased the parts for an external oil cooler, but haven't had the time or energy to install it.
And while running the heat exchanger, dropping coolant temps with better flow, center rad, better rads like CSF, low temp thermo, etc, will make an appreciate difference to oil temps.
Zbomb, looking at your graph the oil pressure is constant from say 4500rpm and up. This indicates the oil pump by-pass is open and the flow of oil is constant. It will reduce the efficiency of the oil cooler. You probably want to run a thinner oil to increase the flow at high RPMs, this should bring your oil temp down.
I've come to the same conclusion, and purchased the parts for an external oil cooler, but haven't had the time or energy to install it.
And while running the heat exchanger, dropping coolant temps with better flow, center rad, better rads like CSF, low temp thermo, etc, will make an appreciate difference to oil temps.
With whatever i do u will be adding a cfs center rad and remote larger 997 cooler per Charles' advice. I already have a low temp and cfs side rads. Just considering it as part of the upgrade costs.
I would NOT recommend using thinner oil to increase flow,( It reduces pressure )there are other ways to increase flow that are much more desirable, if that is your intension.
Increasing flow by using thinner oil and reducing pressure will cause more "dessolved air" to be released from the oil to form bubbles that can block flow.
Oil always has air in it that can be dissolved in it or entrained in it. The higher the pressure the more air can be dessolved in the oil.
Dessolved air in the oil causes no problem, "entrained air" (bubbles) will be formed if pressure is reduced (thinner oil) and will cause overheating and flow restriction.
The oil pump by-pass does not by-pass untill 5 bar pressure is reached
Last edited by Porschetech3; 05-17-2021 at 03:50 PM.
09-23-2020, 05:45 PM
