single dry sump, twin engines?

[bettonracing17]

Please scrutinize the following conceptual oil system setup

Setup:
Two engines arranged longitudonally with a dry sump pump mounted inbetween both engines tied to both cranks using separate belts (1 aft of the other on the front pulley). i.e. a single pump used for both engines. I've thought of two scavenge setups that might work: 1) all scavenge stages of the pump split and run to both engine oil pans and 2) enough scavenge stages to distribute enough scavenge sections to each engine (e.g. 6 stage, 2 scavenge for each engine).

Engines are mildly modded 'busa engines (~11krpm redline)
Application: circuit-inspired tube chassis.


Problems already thought of:
1) The redistribution of lateral loading on the pump shaft. Shouldn't be a major problem as I doubt the shafts/seals/bearing are designed directionally (unless a specific mountng side is specified).

2) Synchronizing both engines (including startup and shut down) to avoid belt slippage or even rotational resistance (i.e. one engine spinning, the other not. Belt not slipping). I think this can be minimized or even eliminated with the incorporation of a 1-way bearing (1-way sprag comes to mind). Please give recommendations concerning engine sync'ing for the pumps if u have any.

Any suggestions/ scrutiny concerning the oil pump setup is welcome. Please minimize scrutiny focused on engine arrangement and 'dry sump vs. wet sump' as there are well written (& responded) posts on these forums concerning those.

Thanks in advance
Kurt Betton

 

[magnograil]

You did not state if the engine cranks were tied to each other in any way than the pump drive. Engines do not turn a constant angular velocity when running a "constant" RPM. The crank accelerates with each power pulse and slows during the rest.
Any reason the pump cannot be driven by one engine?
A "circuit inspired tube chassis" does not make clear if it is still a motorcycle or four wheel vehicle but with 1300cc each, presumably a four wheel. How will you handle the lateral acceleration on the pickup placement? If using a single scavenge, you need some way to keep from pulling air from one engine while oil is in the other.
Why not add a second stage to the existing pump? The Virago has a two stage Eaton style pump chain driven from the crank.

 

[bettonracing17]

Thanks for your help. I was under the impression that the engine does turn constant angular velocity but now that u mention the power pulses, I'm kicking my ass for being ignorant. Esp now that I'm visulaizing a drive belt on an idling racecar (such as a scharger belt).

In my original setup, the cranks were not tied to each other, apart from the drive system (chain/belt) used to drive the oil pump. Considering the angular accel/decel, some sort of one way sprag would become a necessity. Any recommendations? (Bicycle sprocket type pulley on pump comes to mind)

I did plan to use multiple stages - at least 4 (1 pressure/engine, 1 or more scavenge/engine) but I'm still weighing the pros/cons of each scavenge split to both engines or each scavenge designated to one engine only. I think your question about handling air in one engine to oil oil in the other is answered (or at least the effects of the problem reduced) by using multiple scavenge stages.

Sorry about the vagueness of the application. It is a 4 wheel chassis (Hoping to obtain </= 1 tonne). Lateral accel was part of the deciding factor for considering the dry sump. The pickups will be on the side <of each respective engine> inbetween the two engines. This would emphasize the point u brought up earlier about oil/air being pulled up (particularly w/ lateral loads), but I'm still under the assumption that the multiple stages would alleviate any associated problems to negligible.

When u say "...driven by one engine" I'm assuming u're referring to each engine having it's own pump as I'd expect a load imbalance on identical engines if arranged otherwise. To be honest, I can't completely justify not doing pump/engine other than potential cost reduction (single pump) and/or engineering masturbation. I can see few (<-- place emphasis here) applications for "rotation synchronization" but the engineering principle of simplicity usually kicks in just when it starts getting exciting. The setup should be goood to kick off engineering convos at the track tho

> 4 stages are not out of the question either (I guess it's time to start crunching those numbers eh...)

Thanks again.
Kurt

 

[magnograil]

Probably the easiest sprag is a one way roller clutch needle bearing such as a Timkin. It does not have the dead space between reversal and drive that a bicycle sprag would have.
Without windage trays and trap doors, multiple pick-ups are necessary unless the lateral acceleration does not last too long and the oil supply is big enough.
I believe F1 V10 engines use 5 scavenge pumps, one for each V pair which are isolated from each other to cut pumping loss.
I was thinking of the Virago pump as a possibility since it is a complete bolt-on unit with the chain tensioner built into the cover. Not sure about the capacity.
The power pulses are why most engines cannot be converted to Big Bang firing order. The output drive is cushioned by a damper but the cam drive is not. The valves have to work with those angular velocity variations. People design pretty cam profiles but do not take into account that the cam is not rotating at a constant RPM which throws the smooth acceleration curves out the window.
1 tonne is a lot of weight. My Lotus Europa is 1530 lbs. with no carbon fibre in the shell.

 

[bobqzzi04]

Since it seems you are considering comingling the oil, you may wish to consider the effects if one engine were to blow up. I'm not really seeing the potential savings here-you'll need twice as many stages for 2 engines as one (or very close) so the pump will be more expensive. Engineering a sprag clutch will probably consume more time and money than buying 2 drives.

Neat project.

 

[bettonracing17]

Magnograil: Thanks for the recommendation of the Timkin and the Virago. From the little reading I've done on the Virago thus far (coupled with your recommendation), it would appear as an add on pump with the primary objective being to alleviate an oiling deficiency (or increase margin of safety). I'll look into it more but it seems as if I'd still end up doing some fabrication (I'm looking to use 'busa/GSXR 1100 engines) and not have the <even tho somewhat minimal> performance advantages of the dry sump. I'll do some more research tho.

Concerning the weight, I'm aiming for minimum but add in the 3rd engine in the front, all the accessories, the 220lbs driver and the 20 gallons of gas that I'll need to make it to the supermarket, I think 2200lbs is optimistic. I hope to prove myself pessimistic...

===========================================================

Bobqzzi04: Thanks for the reply. I had contemplated the effects of engine failure, but not from the oil perspective. I am under the assumption that scavenge screens as well a "large" capacity oil tank (i.e. = 2* stock oil capacity), in addition to a kill switch <system> aimed at minimizing (or eliminating) one engine running w/o the other, would eliminate (minimize) problems associated w/ oil contamination. Miniscule debris scoring the scavenge section walls would encourage using sperate sections for each engine.

From the minimal internet/catalog shopping I've done, it seems as if multiple stages (e.g. 8stg) cost a little less (10-20%) than 2 pumps equalling to the same total amount of stages (e.g. 2 * 4 stages), but I'm sure once I start buying parts I'll be kicking myself again. I was also contemplating the advantages of adding an electrical "pre-start" motor into the mix. Individual pumps would make this more difficult. I guess I'll make a few calls and get a more accurate estimate of the prices.

Thanks guys. Keep it coming!

 

[bobqzzi04]

Rather than a "pre-start" motor, you may wish to consoder an accusump- a quite simple hydraulic accumulator. I believe Moroso sells them now.

 

[bettonracing17]

Where's the engineering fun in that? j.k.
Depending on my final decision on the arrangement, I will definitely have either the accusump or the "pre-start" motor. My ideas on that were based on the central mounting location (betweene engines), I would have to fab up a plate similar to a 'motor plate' to minimize belt tightness variations. I'm thinking with little to no brainpower I should be able to adapt a motor (starter comes to mind) flipped around that disengages after the momentary button is released. An external (starter gear) to internal (pump pulley) gear arrangement should suffice (with appropriate teeth for engagement/disengagement). The required one way sprags take care of engine pulleys.

Who knows, u might see one way pulleys and pre-lube motors on dry sumps become mainplace in the near future... Or u might just see a bunch of ppl shaking their heads and saying "what a waste of time" looking at my car...

Thanks for your advice. Keep it coming.