Sale ends todayGet 30% off any course (excluding packages)

Ends in --- --- ---

Sealed, vacuumed Dry Sump Systems

Off Topic Discussion

Forum Posts



Tech Articles

If it's not really about tuning or wiring. Then it belongs in here.

= Resolved threads


Not tuning really tuning related although thought id try my luck,

Anyone here with experience with a sealed engine, vacuumed dry sump setups?

Currently plumbing up a dry sump system and using a Peterson Vacuum regulator to control crankcase vacuum for a better ring seal. The plumbing of the pump, cooler and in/out of the tank etc is sorted. Question is related to the breather side. How are you plumbing your breather, if you're running one?

Obviously you can't run a breather straight from the engine, or no vacuum will be created. However, I have seen some systems where the engine has no breather line to the tank, although the tank goes to a breather, creating vacuum in the engine but not the tank. Or other setups where the engine has a breather to the tank and the tank has none, effectively creating a vacuum in the tank and making it equal with the engine.

First time I have tried running a vacuumed system so hoping someone out there has some tips or advice.

Thanks in advance

My experience with dry sump setups has been that with an N/A engine the dry sump can pull a vacuum in the sump which is beneficial in reducing windage and can show a modest improvement in power - So we definitely want that vacuum in the crank case. To achieve this, the engine must be sealed - ie no rocker cover breathers. Even an N/A engine will create positive crankcase pressure under high load/rpm but the dry sump pump should be able to over come this (provided you are using enough scavenge stages). The vacuum regulator is necessary to prevent the dry sump pump creating excessive crankcase vacuum if you jump off the throttle at very high rpm - Theoretically without the vac reg you could end up with all sorts of nasty stuff happening like crank seals being sucked in etc.

The tank on the other hand needs to be vented as this is where the evacuated blow by gasses end up. Also a side effect of the dry sump system is that it removes aeration from the engine oil and this needs to vent somewhere.

On a turbocharged engine this system won't work as (depending on boost level) the blow by past the rings will still create positive crank case pressure. In turbocharged dry sump systems you still need a breather from the rocker cover. Your option is to run this back to the dry sump tank and vent from there or some people prefer to run a separate breather and catch can system in the engine bay from the rocker covers. You should find that with a dry sump system even a high boost turbo engine doesn't put much oil into the catch can - Breathing should be greatly reduced.

I can't see that running the tank sealed would work very well. As stated, the blow by gas needs to be evacuated somehow.

On my old drag car I ran a 4 stage peterson dry sump pump. I ran rocker cover breathers but used one way valves in them so the dry sump pump could still pull a vacuum at idle and low boost. With 54 psi I still saw positive pressure in the crank case but in this condition the one way valves would open in the breathers allowing them to vent. The one way valves were probably unnecessary but it did allow me to monitor vacuum and gave me some idea bout the ring seal.

Thanks for the reply.

So im thinking I will run a line from the tank to a breather to vent the tank, also ill run a line from the engine to the tank with a one way valve.

This should allow the crankcase to hold vacuum although allowing any pressure which may form to vent to the tank.

so a breather and a oil catch can are the same ? or are they two different things?

A breather vents to atmosphere while a catch can is, usually, a sealed unit that catches oil from PCV or other emission systems stopping it from entering the intake air stream while allowing the crankcase pressure to be evacuated by the engine's vacuum.

http://rehermorrison.com/tech-talk-84-dry-sumps-dave-lives/ “In a wet-sump engine, the rotating assembly whips the oil like a milkshake in a blender.”

http://nutterracingengines.com/racing_oil_pumps/crankcase_vacuum_facts.html “Reducing down to dash-10 scavenge lines will help achieve a higher level of vacuum.” Interesting because virtually all pans and pumps use AN-12 lines. Do any -10 scavenge filters even exist?

The biggest Barnes pumps have 2 -16 return lines...unless there is a piston sealing problem can engine blow more than that into the crankcase? Wet sump Supras typically run 2 AN-12 lines to a catch can. http://barnessystems.com/the-barnes-difference/2730101

The old Peterson pumps were georotor on all stages...not ideal for scavenging air. The current Peterson pumps have 4-lobe twisted rotors.

On the 2JZ, Dailey specs a stage just for scavenging the turbo. Many just continue to drain the turbo to the pan. Lots of discussions about this online.

Update: an on-point article specifically about creating crankcase vacuum on forced induction engines: https://www.dragzine.com/tech-stories/engine/tech-boost-vacuum-finding-hidden-horsepower/

A bit late but possibly of interest -

On my 240BHP 10,000RPM 4AGE I has a mid sized 3 scavenge/1 pressure dry sump pump.

It pulled a partial vacuum once above 2500RPM (it ticked over at 200RPM).

There were two "breathers" -

One was a -6 line with an inline restrictor to control the vacuum level (by drilling out the restrictor hole size).

The other was an unrestricted -6 line but with a one way flap fitted. If there was positive pressure (so above atmospheric) then the engine could still vent...rather than blow oil out of every seal and really annoy the rest of the competitors.

Both the vents went to the dry sump tank. The dry sump tank had a rather pointless oil catch tank attached to it to meet regs (and a breather for the axle and gearbox...another long story).

Apart from the rear main seal constantly collapsing under the partial vacuum (big wobbly thing, eventually resolved by some welding/machining and a much smaller OD seal) we never had a single issue with the setup.

Bit late, and i hope you got it sorted your satisfaction, but some thoughts on it (and I may be wrong... ;-) ).

1/ Scavenge pumps need some means of getting the oil to the pumps - or, rather, having a pressure differential between the pump gears and crankcase to blow/suck the oil/air mix from the crankcase (and head/valley if used) to the pump mechanism - with too high a crankcase vacuum scavenging properly can be a problem. For best results, sufficent airflow to carry the oil as a mist or droplets is much better than trying to pump oil alone - th8ink of it like a vacuum cleaner - it uses the airflow to carry the dust, it cannot just suck dust alone. You can see what i mean if you have a container of, say, dry sand - if you hold the tip a little above the surface the air being sucked into the vacuum carries the sand with it, but if you stick the tip deep into the sand it will cease working*.

2/ To allow enough air into the crankcase (and disregarding blowby as it 'should' be minimised) some form of breather should be used - there are commercially available filtered breathers that can be set to a desired vacuum level.

3/ The receiver tank/reservoir must be vented, to void the air sucked from the crankcase and prevent pressurisation and in order for the air/oil separator to work properly.

4/ As has been mentioned, with higher levels seals may not function properly as the pressure gradient is the opposite of what they are designed for. I have known some instances where folks have tried fitting them backwards - lip outwards - but is not generally recommended. There are specialist seal companies that can offer seals specifically for these applications.

*Slightly different results, in detail, as one is a centrifugal and one a positive displacement pump, but similar principle.

Hi, sorry for a late reply but this topic seems to be hard to find on the internet.

I have the Honda K20 turbo dry sump setup build ongoing in my 1994 Eunos Roadster and in the near future I have to start making the turbo oil feed and the return lines. I have had this setup working on the NA K24 engine before for 10000kms without issues. My concern is now the turbo oil return line, how to route oil back? In the naturally aspirated versions the oil is returned to the timing cover from the customized chain tensioner lid. However, this way the oil is basically going almost straight to the crankcase where the vacuum is sucking. Is this a bad idea to implement the oil return from the ball bearing turbo to the timing chain cover or does it need some sort of tray in the middle where to drop the oil and then suck it to the circulation?

I can't see it being an issue, the primary reasons for the dry sump are to minimise the windage/oil in the air around the crank' and rods, and to ensure a reliable oil supply to the engine, and you seem to be returning the oil from the turbo' to an area clear of the crank' where it has easy access to being drawn out by the scavenge pumps - but if you could post the make/details of the front cover?

The Turbo is on the left side of the pic on the level of black VTEC solenoid (about) and the oil drain will go to the black lid of the timing chain tensioner cover. Is that a good place to route the turbo oil return considering the scavenge plate is on the bottom of the block a quite distance away?

TC cover

Have you considered just returning the oil to the dry sump tank?

Returning the oil back to the oil tank is out of consideration since the tank is located at the back of the car.

I have considered to make the return to the timing cover (where wet sump guys put it) or to one of the scavenge stages with a tee.

Front cover or drilled and tapped block or the dry sump should work well - just make sure the material is thick enough.

What's behind the black plate at the front? Might be viable to braze/weld a fitting onto it, if you can keep it flat, or maybe machine an alloy cover with a inlet ready for a fitting?

What you need to bear in mind is the drain is reliant on gravity, so needs to be below the turbo' outlet with a continuous drop to the engine's return point, the greater the drop the better, and it needs to be a large enough internal diameter as practical to ensure it has an easy return. Get it wrong and the oil can gather in the bearing housing to the point where it overcomes the seals and leaks into the turbine and/or impellor housing.

Behind the black cover/lid is the timing chain tensioner. Yes, I have an aluminum one existing already (10mm thick) to weld a fitting there and most probably I will but the oil return there as the drop will be sufficient for it.

I have a dry sumped N/A BMW motor and I am trying to work out how best to achieve vacuum and pressure relief from the rocker cover. My initial thoughts were to run the line from the rocker cover to a catch can that has a morosso vacuum relief valve at the top with a filter fitted which would allow the vacuum to be adjusted but if there is positive pressure it has a relief valve integrated into it which would vent if there was positive pressure. My thoughts were to mount the catch can above the rocker cover with the line connecting it routed to the bottom of the catch can thinking that if there is positive pressure oil or air can enter the can and when there is vacuum the oil collected will then get sucked back into the rocker cover. What i am worried about however is that air entering the catch can whilst it is partially fill with oil that the bubbles will cause oli to exit the can via the breather..How have others achieved this. Also the dry sump tank in the boot and are trying to avoid another return line through the cabin if that were another option...

Any help would be appreciated


Any real reason you can't run a drain from the bottom of the catch tank to the sump, or some other point, the scavenge pumps will be applying a partial vacuum to pull the oil out and the tank's breather would be a good place to have additional, controlled air available to balance the scavenging's air removal?

It might be a good idea to also re-visit the oil separator(s) in the cam' cover, as they 'should' just be letting oil mist through.

We usually reply within 12hrs (often sooner)

Need Help?

Need help choosing a course?

Experiencing website difficulties?

Or need to contact us for any other reason?