Mode of failure for dry sump systems?

[Racer-X]

We are thinking about coming up with a dry sump system for the 2014 car. I know that's a long way out but I figure that gives us a lot of time to work out the bugs.

I have heard that teams have had issues with dry sump failures putting them out of the race. What are the usual ways for a dry sump to fail? From a bit of searching it looks like if you make your own pumps you are asking for trouble other than that what goes wrong? Maintenance, not using enough oil, leaks? I feel like a pressure light would solve most of the engine failures, and testing would find the weak spots.

[Bemo]

The problem with finding the weak spots during testing in this case is, that usually it costs you an engine to find a weak spot http://fsae.com/groupee_common/emoticons/icon_biggrin.gif.

I'm not an expert on this topic as I used to be a chassis guy, but in my experience the critical issues are having enough oil in the system and number and location of spots from where oil is sucked from the oil pan.
As far as I know usually you have one or more pumps which are sucking oil from the engine into the oil tank and one which supplies the engine with oil from the tank. The critical thing is that this pump mustn't suck air at any time. That means your oil tank mustn't get empty at any time. The simple solution is to have a huge tank and running with quite a lot of oil, but that of course makes the system heavy. You can see, where this is going...

Another problem with dry sump systems is that the sucking pumps usually suck a lot of air out the engines which results in a lot of bubbles in the oil making some sort of "oil foam" (sorry I'm not a native English speaker - I don't know how to explain it better in English). That again shouldn't be pumped into the engine...
Getting the air out of the oil can be done by an intelligent oil tank design.

That's pretty much all I can tell you about it from my experience, as I said, I used to be a chassis guy...

[Racer-X]

That makes sense, people just trying to push things to the point they fail.

If you are interested the bubbles and foam usually are called just that. In the case of a motor it is sometimes called windage or aeration.

[jlangholzj]

two pumps, typically a scavenge pump that pulls the oil from the tray and into a tank which the internal pump pulls from to pump the oil through the block.

One thing that you need to keep in mind are the flow rates of your pumps. If those aren't right, you can end up in a bad situation.

[Jan_Dressler]

Originally posted by Bemo:
Getting the air out of the oil can be done by an intelligent oil tank design. ...or by a centrifugal air/oil-separator, attached to the scavenge pumps for example. Might be a little bit heavier, but you have more freedom in oil tank design (and therefore oil tank location in your car) as you don't have to have the de-aeration part in the tank.

Another point not mentioned yet is hose sizes (from scavenge points to scavenge pump, pump to oil tank, oil tank back to the engine...). And to find a clever (and working!) solution for the PRV.

[Dad]

Check out Dailey Engineering. He works with fsae teams. We are working with him on a 1000cc midget project. It is not one of those things you can do without some research, you may loose a few motors in development work.

Honest Dad himself

[Rex Chan]

You got the main failure modes:

1. Oil tank running dry due to not enough oil in the tank/not enough scavenge
2. Leaks - these won't hurt your engine in testing, but will put you out of enduro/driver change. We live with leaks in the knowledge that they *must* be fixed by comp. And if you want to keep your track sponsor happy.

You are correct that an automated shutdown of the engine in low oil pressure situations will prevent any really nasty incidents. We run oil pressure into the ecu (M400) and then via CAN to PDM15, which shuts power to the engine if oil pressure is below acceptable for a given RPM. However, this doesn't mean you can't get accumulated engine wear issues over the long run (our 2011 car engine needed new pistons, which Honda MPE Australia kindly replaced for us).

I don't trust drivers (inexperienced, busy with driving the car drivers) to see the light and turn the engine off. We had this happen in the 2010 car, and it took the driver 40s to see the light and another 30s to drive back the pits. I like automation - some will say it can cause an accident, but I think it's safer and the engine can cut out for lots of reasons anyway on an FSAE car; drivers learn to deal.

If I had to design a dry sump system from scratch, I'd still go with a Dailey scav pump (but with oil-air sep as Jan pointed out). Then your oil tank can be any shape (or just follow the Carrol Smith design - we did and it worked "fine" for us). Then all you have to do is figure out how to route the pickups to the scav pump. And what to do with the pressure releif valve, but it's not that hard.

Feel free to fb msg me to discuss dry sumps more. Though my track record with dry sump pan design kind of sucks....

[Tilman]

Originally posted by Rex Chan:
I don't trust drivers (inexperienced, busy with driving the car drivers) to see the light and turn the engine off. We had this happen in the 2010 car, and it took the driver 40s to see the light and another 30s to drive back the pits. I like automation - some will say it can cause an accident, but I think it's safer and the engine can cut out for lots of reasons anyway on an FSAE car; drivers learn to deal.
Looks like you need a remotely controlled master switch http://fsae.com/groupee_common/emoticons/icon_smile.gif I wanted to have this in some situations, too. And it is not very unsafe if you have a slipper clutch.