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.
a lot of people have stated out of hand that accusumps don't work. i think some of that is internet lore due to initial misapplication of the product. there was some good real data-based analysis on 986forum measuring pressures at different points of the engine with/without accusump, with accusump introduced at different places, with/without the automatic pressure switch, etc. i think if implemented properly an accusump is a viable solution IF oil QUANTITY is the issue (ie pickup tubes can't find oil because of lat-g or pooling in heads). if however it is not quantity but rather the ability of the oil pump to get the oil to the far reaches of the engine (ie, oil pump is going into bypass) then another soln needs to be found.
So, the investigation continues. I had one of the scavenge pumps out today for inspection/measurement. It's properly worn(177k km) so a new is on order. So, what would the effect be of a worn scavenge pump?
Your AOS will be swamped. Oil will start collecting in the head. The drivers side bank will be worst as due to the location of the scavenge pump and oil return and an AOS vent line which gets flooded with too much oil there. It will start slowly. But every right (or left hand for bank 2) hand turn more oil will start flooding the AOS and eventually the in the crankcase will eject the oil out via the AOS outlet. Your engine will become an oil injected engine. I have been there and seen it. Luckily my AOS is rerouted via the air filter box and saved my engine. The oil came into the air filer boy (no catch tank at that time) and caused some nice BBQ smoke, This AOS mod is maybe not the most perfect. But it is a very safe one.
You on the other hand go out of your way to say things like" the oil pump drive is designed to break on purpose to save the engine" and "the Scavenge pump snout is designed to break on purpose" and " the sump baffles are not there to help keep oil around the pick-up".. I know your wife told you the last one, but who told you the other two? Oh and" the IMSB problem is made up, the factory bearing is best !!" Did you dream it up?
I have explained all of that in detail during discussion but you are too stupid to understand(or actually don't read my explanations), that's all.
The things that are meant to break: Well, when you'r an engineer you see certain designs reoccur over and over, this is one of them. I've used them myself many times. It's there to have a controlled break down, you know what will break if something goes wrong. It's good design practise. It's like using fuses in an electrical circuit, you protect stuff. The oil pumps on this engine bolts to the outside of the engine and connects to the inside via a weak link, protecting the inside if something goes wrong. Get it?
This is a cool diagram. Your basic assumption is wrong though. There isn't 100mm of oil level in the sump under track conditions. I don't know the exact number, but I expect it is closer to 2 cm, which is why modifying the sump or pickup is a waste of time.
Chris Cervelli Cervelli Technical Service
Thanks for your input. I pulled the scavenge pump to investigate the oil return as I suspect the running level is too low.
I also mentioned this modification in other threads on this forum where it is commented on. You can find it probably easily looking for posts with my name and ‘AOS’ ‘modif+’ as keywords
Below is a photo how it looks like.
The modification can be further extended with an overpressure relief hose going from the oil filler cap to the catchtank. That is the capped hose on the right you see. In the event of an AOS vent block, crankcase overpressure can get relieved via this connection. The catch tank is not compulsory but it helps a bit with keeping the air filter box clean.
As you can see the weakest point at this stage is the material of the tubes which doesn’t like the tight bends, it tends to crack. When taking the photo just now I see it cracked again. Need to replace it
The use of a rubber-like flexible hose is also not ideal as they tend to fold/collapse in a bend due to the extra elasticity from the heat of the engine. I notice I need to clean my engine bay I more thoroughly. But it is my daily car, outside sleeper.
The AOS modification I have is probably not the most ideal following the expert’s opinions here.
Check also PorscheTech’s thread about the ultimate AOS modification where my setup has also been discussed and conceptually been reviewed.
There are advantages compared to other modifications/solutions:
the AOS will never fail. I mean that the weakest link, the diaphragm is not subject anymore to the harsh vacuum application.
If any oil comes out from whatever other failure it will not go into the intake.
Your intake stays free from oil dirt
Idle is much smoother as there is no more 'unmeasured' air form the crank int he intake
Disadvantages:
Your air filter gets a bit dirtier quicker, and you need to clean out the lower part of the air filter box once a while.
There might be some dirt get past the air filter onto the MAF
the experts says the crankcase pressure is not ideal.
Related to disadvantage 1: never had any problem. I change air filter more often.
Related to disadvantage 2: no issue for me. I check the Hot Film Voltage regularly. If it comes near to 1.05 or 0.95 I change it anyway. For me the MAF is just a wear item. It is one of the most important sensors in the car. Till now I replaced it every around 40.000km
I think when you are tracking your car it is a nice solution to get rid off ALL the the risk of the most common AOS failure.
I have this AOS modification on my car for 80k kms / 50k miles now.
It has been further done on +300 cars by now.
From what I heard no problems so far.
At least mine is still going very strong. Total 135k miles now
Okay lengthy post. Deviating form the original topic
It seems very unlikely that you could run oil lines to the front of the car, install an air-to-oil heat exchanger there, runs lines back and have that arrangement be an improvement over the stock oil-to-water heat exchanger, which works very well.
Chris Cervelli
Cervelli Technical Service
I was thinking NACA in rear bumper behind wheel feeding into a shrouded heat exchanger mounted off the old muffler hanger. Shroud would be simple CF surround (prob NASCAR surplus ) with heat reflective foil on the outside. Exhaust vents cut into back of rear bumper.
I was thinking NACA in rear bumper behind wheel feeding into a shrouded heat exchanger mounted off the old muffler hanger. Shroud would be simple CF surround (prob NASCAR surplus ) with heat reflective foil on the outside. Exhaust vents cut into back of rear bumper.
Because of the shape of the car and the proximity to the wheel that's a low pressure area, which is why the Turbo's intercooler exhausts are located there. I don't think you would get much if any airflow into a duct located there, even a NACA duct. You could exhaust there though I don't see an obvious way to supply intake air.
Were it me I would locate an oil/air heat exchanger on the driver's side of the transmission with fan forced airflow. That is where the oil/air heat exchanger for transmission cooling is placed when adding pumped transmission cooling to the G50 gearbox.
Another choice would be an oil/water heat exchanger with a separate water circuit.
A third possibility would be keep the existing oil/coolant heat exchanger and improve the cooling system. When I ran Lime Rock in my Turbo on a 90 degree day I had coolant temps max out at 195 degrees and oil temps at 220 degrees. That's with the stock Turbo heat exchangers - and the Tiptronic ATF cooler is also in the water cooling system.
It does seem that a classic air/oil heat exchanger would be the simplest if it could be made to work.
Had the new scavenge pump installed this morning and I'm just back from the test drive. I test drive the same route all the time when testing this issue, it'a 10km road climbing around 500m up a mountain side.
The pressure drops are now very much muted and some of the turns where it used to drop it stayed steady. Still not perfect but then again I have only replaced one side and seeing the condition of that one I think the other side is probably equally bad. Bank 1-3 seams a bit tricky to change tho, not enough space to pull out...or is there a trick to it?
At the top of my test drive I stopped on a level surface and checked the oil when idling, it was reading at the bottom tip of the dip stick. When checking according to the manual (hot and level) it's at 4/5 full. Thus, just at idle on a level surface there is roughly 1.5L in circulation.
Well, when you'r an engineer The oil pumps on this engine bolts to the outside of the engine and connects to the inside via a weak link, protecting the inside if something goes wrong. Get it?
That is the perfect example of your preposterous statements !!
1. If the main oil pump drive breaks, what exactly are you protecting? Nothing !!!
2. If the Scavenge pump snout was designed to break, it does a poor job !! Many many times when the scavenge pump ingests a piece of valve spring or lifter debris it will lock up and instead of the snout breaking the timing chain will break !!
This is not Porsche Engineers intention, this is your imagination !!
Give it a rest....
As Mr Raby says, cooler oil is better. He does not say what temperature he prefers, but I would like to see a normal operating oil temp of 200 degrees.
This is not going to happen unless you are racing in 30 degree weather and the cooling system remains in the control range of the coolant thermostat. Maybe in Canada, but I doubt it.
It seems very unlikely that you could run oil lines to the front of the car, install an air-to-oil heat exchanger there, runs lines back and have that arrangement be an improvement over the stock oil-to-water heat exchanger, which works very well. With the stock cooler and with a 3.4 engine, expect to see an oil temp 40 degrees F higher than the coolant temp when you are racing, which means the oil is likely to be over 250 on a hot day. So the engine and everything else has to be able to withstand that temperature, and all you can do is try to drop the coolant temp and drag the oil down with it. It would be great if the stock cooler were more effective and could shrink that 40 degree delta, but it is what it is.
It is not going to happen. On a hot day, the coolant will be 210 and the oil 250. Plan accordingly. I have raced these engines many hours with the coolant temp at 235, and they will sustain that without damage. The resultant high oil temp is a much greater concern, even with a dry sump.
Chris Cervelli
Cervelli Technical Service
When I ran the m96 engine, I had the Mantis 2qt sump and plumbed some -8 lines to the front where an oil cooler lived. The oil was being fed up there through a Mocal electric pump. The result was oil temp staying around 220-230* at the track (with 70-80* ambient temps). Water temps were roughly 210-215 regardless of what I did (3 CFS radiators). I used JoeGibbs racing oil in part due to the advise on this board (which didnt change temps or pressure). What i noticed is oil pressure dropping down to 1bar (sometimes 0 bar) with anything above 1.1G turns (so 80% of the turns on my local tracks). This was with 200tw 225/265 Hankooks. I was taking temp from the sump so it was pretty accurate. I tracked the m96 for 2 years before I realized nothing short of a dry sump would solve my issue. An accusump will help with most tracks, but some will still be an issue.
I see tons of people track their 996's over the years and some do fine, some blow up. To be fair, of all the cars I've seen come off the track with issues, the 996 is up there (boxster, cayman 996...any m96 engine). I gave up trying to mess with the M96 and sold it. A common mind set I've seen shared between people who track these things is: "I'm running on borrowed time". People are willing to accept that these engines can let go at anytime. If you can stomach that, I see no issue. I can't...there are significantly better platforms for racing (chassis and money wise). These cars are terrific for the road though
When I ran the m96 engine, I had the Mantis 2qt sump and plumbed some -8 lines to the front where an oil cooler lived. The oil was being fed up there through a Mocal electric pump. The result was oil temp staying around 220-230* at the track (with 70-80* ambient temps). Water temps were roughly 210-215 regardless of what I did (3 CFS radiators). I used JoeGibbs racing oil in part due to the advise on this board (which didnt change temps or pressure). What i noticed is oil pressure dropping down to 1bar (sometimes 0 bar) with anything above 1.1G turns (so 80% of the turns on my local tracks). This was with 200tw 225/265 Hankooks. I was taking temp from the sump so it was pretty accurate. I tracked the m96 for 2 years before I realized nothing short of a dry sump would solve my issue. An accusump will help with most tracks, but some will still be an issue.
I see tons of people track their 996's over the years and some do fine, some blow up. To be fair, of all the cars I've seen come off the track with issues, the 996 is up there (boxster, cayman 996...any m96 engine). I gave up trying to mess with the M96 and sold it. A common mind set I've seen shared between people who track these things is: "I'm running on borrowed time". People are willing to accept that these engines can let go at anytime. If you can stomach that, I see no issue. I can't...there are significantly better platforms for racing (chassis and money wise). These cars are terrific for the road though
+1. If folks want a real race car, they should get a real race car. The stock 996 is not that.
So you used a discrete pump to move oil from the sump to the front of the car ?
How did you ensure that when the sump volume was compromised due to cornering that you weren’t exacerbating the issue by pumping what oil was left in the sump to the front mounted cooler ?
Where did the oil re enter the oiling system after leaving the cooler ?
So you used a discrete pump to move oil from the sump to the front of the car ?
How did you ensure that when the sump volume was compromised due to cornering that you weren’t exacerbating the issue by pumping what oil was left in the sump to the front mounted cooler ?
Where did the oil re enter the oiling system after leaving the cooler ?
good question. Oil cooler pick up was farthest away from the actual oil pick up (engine oil pump) and return was closer. Sadly, only real data I can share is that oil drop happened quicker and more severe without the oil cooler. Prior to the cooler, oil temp would get to 250-260 in similar conditions and oil pressure would never get above 2.5-3.5 bar in the straights. After the cooler, it would take longer for oil pressure drops to get to 1bar (they would usually only drop to 1.5-2bar) but it would always eventually get there...I was just buying time. At the end, I may have made some things better while others worse, which is probably what your question was eluding to. Mantis recommended this set up claiming that it solved some of their issues. Never solved mine.
This may not have any relevance on the current thread, but I now run an LS3 engine in the 996. I use the same oil cooler set up (just on a ls3 pan). at the track, temps are managed within reason.
+1. If folks want a real race car, they should get a real race car. The stock 996 is not that.
I don’t get what the $hit that has to do with anything.
Thread is about addressing a weakness with the platform, Some really smart guys have provided their thoughts - that is a benefit to the community regardless purpose of use of the owner.
If only race cars were used on track, Track attendance would be pretty low huh...
Jason
Originally Posted by kondata17
good question. Oil cooler pick up was farthest away from the actual oil pick up (engine oil pump) and return was closer. Sadly, only real data I can share is that oil drop happened quicker and more severe without the oil cooler. Prior to the cooler, oil temp would get to 250-260 in similar conditions and oil pressure would never get above 2.5-3.5 bar in the straights. After the cooler, it would take longer for oil pressure drops to get to 1bar (they would usually only drop to 1.5-2bar) but it would always eventually get there...I was just buying time. At the end, I may have made some things better while others worse, which is probably what your question was eluding to. Mantis recommended this set up claiming that it solved some of their issues. Never solved mine.
This may not have any relevance on the current thread, but I now run an LS3 engine in the 996. I use the same oil cooler set up (just on a ls3 pan). at the track, temps are managed within reason.
I applaud the ingenuity but I would. not run setup a system like that.
1. I would never want to rob the sump of oil
2. your dumping cooled oil back in with hot instead of letting it soak heat by running through motor first
3. Long lines/external pump
Good input though as not a lot of folks have seemingly tried the external cooler method.
I don’t get what the $hit that has to do with anything.
Thread is about addressing a weakness with the platform, Some really smart guys have provided their thoughts - that is a benefit to the community regardless purpose of use of the owner.
If only race cars were used on track, Track attendance would be pretty low huh...
Jason
I applaud the ingenuity but I would. It setup a system like that.
1. I would never want to rob the sump of oil
2. your dumping cooled oil back in with hot instead of letting it soak heat by running through motor first
3. Long lines/external pump
Good input though as not a lot of folks have seemingly tried the external cooler method.
you're not really robbing the sump since the lines are never empty to being with (and if they are, its because they failed to pick up oil from the sump, so there is extra oil in the sump). If you're continuously running the pump, the oil in the sump balances out temperature wise. I agree with your third point on extra stuff to break/fail but that was the only way I could make it work. Going on 4 years without any issues.