Custom, small, simple oil/air separater in the scavenge line

[RussMcB]

I'd like your opinions regarding a small custom oil/air separator can I'm considering making.

My scavenge pump is -10AN and my oil cooler inlet is -12AN. I think I'd like to introduce a simple, small tank that will separate some of the vapors out of the mix, while taking care of the hose size mismatch.

First question: Is this a worthwhile project? I've had a few people tell me not to bother, just use an adapter to address the size difference.

Second question is regarding design. I plan to have the -10AN inlet port enter near the top of a small can (maybe 2-3" diameter, 4" tall) tangentially to encourage swirl. There will be a -12AN outlet at the bottom to send the oil to the cooler, and a pretty small vent hose at the top, plumbed to the oil tank. The vent needs to be small, otherwise too much oil will take that route, bypassing the oil cooler.

I've got a buddy who's good with Tig.

What do you think? Is it as simple as I think? Any suggestions regarding the design or if it will help cooling?

on edit: FormulaSuper (Scott) has a similar car/engine/oil cooler combination and he does not have any oil cooling issues. In fact, he says he may start blocking off some of his cooler. So, this mod I'm considering isn't to address a cooling problem.

I've got a buddy who's good with Tig.

Thanks very much for your input.

[Jon Jeffords]

In order to allow the entrapped air to rise to the top, you will need the oil velocity in the tank to be slow, and the residence time in the tank high. I have a suspicion that the tank you described above will be too small to allow either.

Any objection to plumbing the oil cooler on the discharge side of the feed pump, instead of the scavenge pump?

[RussMcB]

Thanks, Jon. Good info.

Plumbing the oil cooler after my main pump isn't an option because my dry sump system uses the internal motorcycle engine's oil pump (which directly feeds the bearings, etc.).

[Lotus7]

Russ, your intentions are correct, since oil coolers are terribly inefficient when asked to deal with aerated oil, but I fear the device you're describing won't work well.

The separator can't be upstream of the scavenge pump; if it were, the pump would simply pull air in through the vent and not scavenge any oil.

If the separator is downstream of the pump, in front of the cooler, the back pressure of the cooler will likely cause the separator to fill completely with pressured oil (and squirt small amounts out the vent as well), preventing any meaningful de-aeration effect.
(A good oil de-aerator needs an air column up its center to function with any efficiency).

Unless I've pictured your setup wrong, about the only way to achieve aeration is to scavenge the sump directly into a vented dry-sump tank of suitably large volume (or bellhousing in most of our cars), get whatever de-aeration you can out of that tank design, then cool that oil as it comes out of the tank on its way to the engine.

[Jon Jeffords]

Plumbing the oil cooler after my main pump isn't an option because my dry sump system uses the internal motorcycle engine's oil pump (which directly feeds the bearings, etc.).

I have seen some sexy 'sandwich plates' that install under the oil filter that will allow you to divert the feed pump flow to an oil cooler, then back to the filter:

https://www.pegasusautoracing.com/ke...ywordButton=Go

[romoman]

This is basically what you want, but they are expensive and moderately big http://www.spintric.com/

[RussMcB]

Thanks guys. Good points and suggestions.

I saw the Spintric last week while Googling, then saw the price!

<snip> ... If the separator is downstream of the pump, ...

Yes - After the scavenge pump and before the oil cooler.

I was thinking, yes, there will be stuff going out the vent (because of the backpressure caused by the cooler), but not a big deal - It might be mostly a frothy mix, and it would be vented to the big oil tank (not to atmosphere).


It's starting to sound like more trouble than it's worth. I certainly have plenty of other things I can/should be working on in place of this one.

[Lotus7]

I was thinking, yes, there will be stuff going out the vent (because of the backpressure caused by the cooler), but not a big deal - It might be mostly a frothy mix, and it would be vented to the big oil tank (not to atmosphere).
.

Yup, I realized it was venting to tank, I'm just not convinced that a full, pressurized "swirl pot" will be effective enough to be worth the trouble.
(and of course, every additional oil connection is another possible leak) :-)

[S Lathrop]

On my cars, I run an inline screen between the scavenge side of the pump and the oil cooler. The screen will help separate the oil and air, reduces the foamy oil, and improves the cooler performance a bit before the oil is returned to the tank.

What I rely on is an oil/water heat exchanger between the oil pump and the engine. Modine made a oil/water heat exchanger that fit between the oil filter and the block. I have used Modine exchangers with remote mounted filters. The water source is a continuous flow of water from the engine before the thermostat. The case of FF and FC engines, I took water from the rear of the cylinder heads.

The other thing I have done is use a oil/water heat exchanger in the cross over pipe between the 2 radiators on my cars. For the exchangers I started with a oil cooler and built a box around the cooler to force water through the core. This system worked the best. It has the advantage that it helps heat the oil; as well as cool it when it gets above the water temps.

Cooling the pressure oil is way more efficient than cooling the scavenge oil.

[RobLav]

Steve wrote what I was about to post. An inline screen. You want one anyway! And I used a Marston Aerospace oil/water heat exchanger on mine to great effect. I found it on EBay years ago for like $100 bucks.

[HayesCages]

What's your projected gpm of oil flow at race rpm?

[RussMcB]

What's your projected gpm of oil flow at race rpm?

No clue. The engine is a stock Suzuki Hayabusa with a stock internal oil pump. I think redline is 10,500 RPM.

The SBD dry sump system's scavenge pump is two stage. I don't know its specs. I'm not sure how much oil (and air) it will be sucking/pushing.

Good suggestion about the screen. Maybe I can find one with changeable ends so I can have -10AN inlet and -12AN outlet.

[HayesCages]

https://www.summitracing.com/parts/ctr-22-750

https://www.summitracing.com/parts/aer-fbe2954

[RussMcB]

https://www.summitracing.com/parts/ctr-22-750

https://www.summitracing.com/parts/aer-fbe2954

Thanks, but (if I'm reading the specs right), both in/out would end up being -12AN (and I need a -10AN inlet).

This is still good, though. Might be very helpful for searching for the right combination.

Thank you.

[HayesCages]

If you look you will see that the adapter is -12 x -10 although they don't make it clear which end is which.

Pegasus sells one as well. Look up part #3603-12-10

Summit parts, your choice of color: https://www.summitracing.com/search/...ibanner=SREPD4

Screw adapter into -12 female end of inline filter and there you go.

[dereklola]

........but both in/out would end up being -12AN (and I need a -10AN inlet).

Russ - since you've got a friendly welder just chop off the -12 end and weld on a -10 fitting. I do it all the time, saves stocking or buying all kinds of odd-ball or odd-angle fittings. I use my lathe to 'chop off' but you can do it with saw and file quite OK. While in the lathe I also open up the internal passages a couple of drill sizes - got to be worthwhile on oil flow although I've never tried to measure it.

Regarding the air separator - I'd copy the Spindrift idea - start with a tube just add end fittings to suit - kind of what you suggested to begin with. Based on my Cosworth pumps (think Indy Lights and F1/F3000 etc) the air separator is quite short - just a swirl chamber on the end of the scavenge pump (in the same housing) - maybe 2" long. It does however have a -12 'air' hose back to the top of the oil tank.

Final thought - on my old Atlantic (Lola T460, also on many others like T342 FFs) the air separation was achieved by directing the oil/air mix inside the oil tank on an angle on to a flat surface - ie. the inside of the tank. Same scientific theory as a swirl pot - just flat instead of cylindrical. Worth think about!

Good luck and Merry Christmas - Derek

[10rmotor]

Russ, when I ran the old carbureted ZX10 we carried a huge capacity dry sump tank (about 2 gal volume) and the tank had an internal swirl pot and vertical baffles with trap doors in the bottom. It was a tal rectangular tank. That motor had a two stage pump. One stage sucked out of the sump and fed the tank with the oil going into the swirl pot first before dropping into the tank. Second stage sucked out of the tank, pushed the oil through the filter, then the cooler, and then to the bearings and such. Perfect solution.

I still have one of those tanks. Don't know how much space you have, if you're interested I can give you more details.

[RussMcB]

Thanks, Ted. I'm content with Plan B. I bought a fairly big Canton tube screen filter with -12AN ends (pn 22-660), swapped one end for a -10AN (Earl's 985011ERL) and I think it'll be fine. The RT-4 has a decent oil tank.

I've moved on to the cooling and fuel systems. It's all slowly coming together. It's nice working on a project without a deadline.

[10rmotor]

It is nice to not have a deadline, but in my experience you need to have a goal in mind that you take at least somewhat seriously. For proof of this, just look at how many partially finished race cars there are for sale.

In our area, if you don't at least bet lunch on when the car will take the green flag, you're not serious. I call it the Bob Fox factor, as usually the bet is with him!

[dereklola]

Ralt oil tanks are different. They have nothing inside - no special swirl pot, no baffles, no deflector plates and they are large diameter and squat (low height). However hundreds were made so they must work.

Several years ago I bought on old beaten up one just to cut open and inspect. Pics below.

1 - Tank is 8" diam and 11" tall.
2 - nothing inside
3 - top of tank - tangential inflow hits the 8" diameter - much larger than most "swirl pots" - air extracted from the oil just escapes through side tube vent in the central filler which only goes down 2"
4 - bottom is V shaped with small sump - 1.5" diam tube for oil out has 1"sq. hole on bottom (ie. as low as possible) - all to ensure only oil is fed to the pressure pump.

Hope this is useful to some. Russ - if you are planning to use the Ralt oil cooler (Mocal type) and the Ralt tank it should work fine - unless of course the Suzuki oil flow rate and air inclusion are vastly different from the Cosworth and Toyota engines.

[RussMcB]

That's fantastic. Thank you very much, Derek.

Mine looks almost the same, but the vents might look a little different? See pic below.

I think mine might have had some Frankenstein surgery at some point. :-)

Thanks, again!