Oil Starvation Discussion

[Benckj]

Intending to open up the discussion on a potential engine killing topic which was briefly discussed on the Hampton Down event thread.

http://mr2.org.nz/phpbb3/viewtopic.php? ... 03#p131803

Would like to identify contributing factors and look for possible solutions. It is well known that gen 2's are particulary vunerable to oil starvation which is why 3sge engines are getting hard to come by. Not sure about AW's or Spyder engines

Contributing Factors
1/ Low oil
2/ Oil type
3/ Oil viscosity
4/ Engine condition
5/ Enviromental conditions
6/ Corner geometry (length, speed, camber, radius, direction, etc)
7/ Sump design
8/ Oil circulation design
9/ Engine model (AW, SW, ZZ)
10/ other....

Prevention Measures
1/ Additional oil in sump
2/ Baffled sump
3/ Dry sump
4/ Oil filtration (remote, cooler, etc0
5/ Oil pump flow
6/ Accumulator (ie Accusump or other systems)
7/ other....

[Benckj]

No replies? Would have thought this would be a hot topic.

[chief1eye]

Give us a bit more time Jim!

That's fine, just thought others would have some ideas worth chasing up.

[chief1eye]

Having experienced oil starvation myself with my MR2 at Hampton Downs back in June last year, Yep on the killer sweeper then bang up the main straight. From that experience I have concluded that the risk of this happening to the stock cars would fall into possible to likely. I can only talk about a gen2 3SGTE engine as I'm not familiar with the Gen3 or later. The stock sump is not baffled and lacks capacity for getting serious on the track. If your'e on the track your gonna give it some. The idea of topping up with a little more oil, saying well Ive only got road tires so I wont be cornering that fast, not like I got semi slicks and even limting the revs ( which is hard to do when you got someone up our tailpipe), Doesn't stop the laws of physics and centripetal force. My recommedation is spend the money and get an upgraded baffled racing sump. However, running the big end bearing did start my journey on the new engine build which I'm really stoked how it has ended up. So if you got the money just go hard until you run a big end bearing and go for the rebuild you always wanted!

[Benckj]

So what is the solution apart from rebuilding engine and adding performance dump?

Maybe back up a step and look at the problem first. Certain engines seem to suffer the most failures; 3sge, 3sgte, Subaru boxer........

[GDII]

So what is the solution apart from rebuilding engine and adding performance dump?

Maybe back up a step and look at the problem first. Certain engines seem to suffer the most failures; 3sge, 3sgte, Subaru boxer........

Mitsubishi EVOs suffer the same fate from that corner.

[Benckj]

OK, so what is the common denominator between these engines? Sump position, capacity, oil pick-up, oil return, oil type, viscosity, etc, etc?

I wonder if we can start spreadsheet to capture info on failures. From this info conclusions can be determined rather than guessing at answer.

[GDII]

The only thing I can tell on the MR2 and EVO is the sump position being on the right side of the car forcing oil away from the pick up. This is the same for almost any FWD based 80s and 90s Japanese cars except for Hondas that have the engine and sump on the other side of the car. Not that other engine types and configurations can't experience the problem, it's just more prominent in these engines we know of.
The BRZ/GT86 has a central sump but being a boxer they are different and I have very limited knowledge of them so can't comment. All I know is they are stupid complicated in design. Pulling apart a boxer engine takes so long to do and so many parts.

Other than that I can't be sure of what other factors come into play with these engines. Like Gary mentioned, not reving it too high to allow more oil to stay in the sump during these corners.
Most people know of the Moroso sump for the GEN2 3S with it's wings and baffle plates. This obviously allows for much higher oil capacity and it helps keep the oil centred around the pickup. But for lots of people this is just not practical and for racing series' like the 2K cup you cannot use it.

[Malcolm]

OK, so what is the common denominator between these engines? Sump position, capacity, oil pick-up, oil return, oil type, viscosity, etc, etc?

I wonder if we can start spreadsheet to capture info on failures. From this info conclusions can be determined rather than guessing at answer.

I think you're going to need a lot of data in that spreadsheet before you can draw any proper conclusions, there are just far too many variables to try and control and you're essentially relying on anecdotes (ever heard the saying, "the plural of anecdote is not data"?). If you want a definitive answer as to why 3S engines suffer from oil starvation, I would think it would be better to do some controlled experiments on the track.

It's not really rocket science to figure out the cause(s). Start with a stock setup, a device capable of logging a couple of 0-5V sensors, a good oil pressure sensor, lateral and longitudinal accelerometers, and an engine speed input (ideally this would just be an aftermarket ECU with accel data being input into it). If you have the right software or are good at processing data, you should be able to single out the conditions which cause oil surge. As far as oil surge is concerned, the engine doesn't really care about "5/ Enviromental conditions 6/ Corner geometry (length, speed, camber, radius, direction, etc)". Although those things have an effect, you really need to reduce them down to the basics: oil temperature, engine speed and vehicle accelerations. Oh and time.

If you have a good low oil pressure warning system - not just the idiot light but preferably using the aforementioned logging device to tell you when oil pressure has deviated from the normal value for the engine speed and oil temperature, you should be able to do this without killing an engine

Then once you know the conditions that cause surge, it's just a matter of trying some possible solutions. This is the tricky bit. From my experience, I would make the following notes:
* Oil type/viscosity doesn't make much difference. It'll change the normal oil pressure, but going thicker or thinner doesn't usually have much effect on whether or not it surges
* Adding extra volume is a pretty reasonable and easy solution, but if it works you need to be wary of potential side effects, e.g. excessive windage causing significant aeration and/or heating of the oil
* Accusumps are ok but they're not without problems. They add a lot of inertia to the system, so when you get surge/starvation, instead of an immediate drop in pressure it just slowly drops, but the flip side of that is that after a surge it takes longer for the oil pressure to recover to normal, and when you increase engine speed the oil pressure lags behind a bit. You can tune that behaviour a bit by changing the air pressure (i.e. spring force) in the accumulator, but it's always a compromise. You also really need to use a solenoid to shut them off before the engine stops, otherwise they lose pressure when you stop the engine and significantly extend the time taken for the engine to build up pressure after you start it - this is really not good.
* Well thought-out baffling can be very effective
* Dry sumps are awesome, but they still need to be properly designed to work well

[Benckj]

All good information. Rather than funding a research project I'd rather collate data in this thread in order to understand the issue and derive possible solutions. I agree it will not be a scientific result but may help keep a few engines from destruction.

Keep the ideas flowing.

[Malcolm]

Rather than actually investigating the issue, I'd rather try and solve it entirely while sitting at a desk and speculating wildly while drawing tenuous links based on incomplete information

there I fixed it for you

[Benckj]

Not many of us have an engine with sensors and logging equipment installed along with a dedicated race track to collect data. I'll ask Tony Quinn next time I see him if he wouldn't mind using one of his tracks for the experiment.

[Malcolm]

Don't you have a megasquirt? Can't they do logging? If not you could tack on $100 with of Arduino bits and make something that works. Aren't you an engineer of some type?

[mknz]

Don't you have a mega squirt? Can't they do logging? If not you could tack on $100 with of Arduino bits and make something that works. Aren't you an engineer of some type?

I have an Arduino, if you have a gen 2 in your car and an oil sensor pop round one day before Hamptons and we'll sort it.

[Malcolm]

Don't you have a mega squirt? Can't they do logging? If not you could tack on $100 with of Arduino bits and make something that works. Aren't you an engineer of some type?

I have an Arduino, if you have a gen 2 in your car and an oil sensor pop round one day before Hamptons and we'll sort it.

I'm good thanks, I too have an Arduino (well, at least 8 of different flavours), one of which I've been dicking around with to make a super-cheap logger/display unit for my car for fun. If I was going to undertake anything I suggested above though, I'd probably borrow a Motec logger from work for the ease of use/ability to use the i2 software for analysis which makes life significantly easier. But to be honest, oil starvation isn't much of a problem for me, given the car sits in the garage about 350 days a year and only comes out when I can't be arsed riding to work or I need to get a WOF