Turbo Engine Blowby, High Load, Sustained High RPM

[Chris1]

Oil/Air separator tanks vertically located (drain back) to each vent should help a great deal. In steady state testing I've seen simple systems like this reduce blowby accumulation on a well sealed engine 20 fold and more versus a direct path breather system to the accumulation tank. You're situation has the added difficulty of best case breather location as well... gdlk!

[Buddyworm]

You guys are awesome. I'm sorry for not being able to reply sooner! I have been short on time to give proper reply to all who responded.

The thought-per-word ratio on this forum is definitely up there.

If you even have .5% blowby on a 500hp engine, that's 2.5hp worth of exhaust blowing past the rings.

I've seen some remarkable wind blowing from the breathers on otherwise tight sealing engines running at high boost.

Good point! Is there some known minimum percentage of possible blowby leakage during a power stroke? I'd imagine on many engines the total blowby % is much higher than .5%.

What engine are you working with and what rpm is "high rpm"?

I think you could be on the right track, crank case pressure is building quicker than you can vent it and drawing oil along for the ride. What size are the breathers and how's the baffling? There also might be some clues in sump design.

It is a CA18DET engine. My redline is set at 7500rpm and during drifting I tend to keep it between 6-7000 over the longer sections with the occasional bounce off the limiter.

It has two valve covers, one each for the intake and exhaust cam with one 5/8" breather on each cover. Both sides are connected through passages in the head casting. The engine sits canted to the exhaust side and I have blocked off the exhaust side breather as it is low on the valvecover and at the rear. The intake side breather is high on the cover and at the front. It is connected to an AOS/catch can.

There is a 7/8" breather port right off the block, more or less adjacent to the crank with a baffle plate. This is also vented to the can.

Sump is nothing fancy so far as I know, basically stamped steel with a small amount of baffling below the crank girdle.

It's late and I'm really tired, so I'm going to unlurk a bit and suggest something off the wall - the sustained rpms when drifting fills the valve cover with oil and you are getting blow-by past the intake valves that is bubbling through that oil and sweeping it out the breathers. As theories go, it's got some good points. It's a leakage path that won't show on a blowby test, and you aren't getting any detonation which I think you would be if the blowby was going past the rings because some oil would surely be going upwards some of the time. On a drag strip it's all over in a few seconds, and on a road course you are on and off the throttle and cycling the rpms up and down constantly so there's much more time for drainback compared to drifting. Would be interesting to put separate lines to separate catch cans on the valve cover and anywhere lower on the engine that you can, to see where the oil is coming from. Even if the airflow isn't coming past the intake valve stems it must be some kind of windage and/or lack of drainback issue that gets the oil up to the breather lines so it can be swept out. Well, hopefully it's at least an amusing idea, I'm going back to lurking and learning.

Good ideas. I'm positive this engine suffers from this to a certain degree on the exhaust side due to the canting of the engine mentioned above. I do wonder if it's significant enough to cause what I'm seeing.

You probably need a more effective breather system.
Most aftermarket "systems" I see are very crude, and not very good.
Venting the engine:
Simple form, using a breathing location on the engine with the least oil mist naturally, your are using the engine itself as a mist filter. It can work, but as internal windage gets greater that may not be enough, then next logical step you put a baffle system in the valve cover or wherever there is room for it, and some of those baffle systems are quite impressive, and go unnoticed, until some fool puts on a aftermarket valve cover LOL, but if that isn't enough then what?
Then you make an external mist filter. Just using a "big box" is good to keep oil off the track in the event you ventilate a piston, and that's good for everybody else, but not good for you.
The difference between a barf box and a mist separator would be the separator is made to remove the mist and then drain back sto teh sump, not HOLD it, so it will have a drain, and it will have some scheme to do this. Centrifugal is a simple but useful idea. Simple form you use a tube vertically, have the vent lines from engine come in at a tangent in the middle(so it swirls), at the bottom is a drain line, at the top is the vent. There is a specific corner of engineering about cyclone filters, to work best the proportions and velocities need to be looked at, but good news is even if you are way off it still works suprisiingly well, but not as good as dyson vaccum LOL

The funny thing is, building WET FLOW intake manifolds, we want to keep the mist SUSPENDED, then build a breather we want the mist to fall out, and many of the same rules apply. The worst intake manifold ever would be a good breather box, and the worse breather box would be a good intake
Mist flow is also an issue with stills, which we use, you know, for "fuel"

I had considered this in fact! Although ventilating through the valve covers doesn't seem to trap any oil, thing still pushed oil into the can before I connected the 7/8" port directly to the can.

Cyclones definitely make sense and I wish I'd seen the design peejay had posted before I started building my AOS!

My AOS is divided into two chambers, one side for the breather inlets and the drain, the other for the vent to atmosphere. The divider does not go all the way to the bottom of the can. My thought was to create a longer path, with directional changes to encourage the entrained oil to fall out of suspension. My divider is a curved section of pipe wall with one wall intersecting the wall right next to the breather inlet ports so I'm hoping that will induce some swirl as well.

You'll know what's wrong with it when you pull the pistons out.

To me it's just the usual turbo problem.......knock monitors that don't work and broken ringlands as a result of it.

Catch cans are only for broken engines and compression tests don't pick up on broken ringlands, if anything the numbers are a bit higher
Checking the plugs ALWAYS works.

Plugs look a-ok to me! No aluminum flecks or other signs of detonation, no detectable oil deposits on the plugs either. Haven't torn the current block down yet, but the last one that I took down exhibited similar blowby flow and the pistons that came out of that engine were fairly pristine.

We'd expect signifcant oil consumption with a cracked land, yes? With my barf box (great term) draining back into the pan the engine doesn't seem to consume oil.

So, high sustained G forces and RPM cause your engine to puke oil? Do you know about the history of racing and the evolution of engines and lubrication systems in land based racing? You have the greatest of opportunities for complete loss of oil control under your operating conditions: a wet sump and breathers as a means of "control." We learned all of this from WWI and WWII aircraft engine designs done long ago by genius designers. Study at their feet for a long while and prosper. It's all in books.

They made upside down V-12s work just fine. Get your study on. Your engine will benefit.

We made CanAm unlimited race cars with clones of those WWII engine lubrication systems and boy howdy they worked just fine thank you.

I definitely don't know enough! Any suggested titles? I'd be very interested in learning about those WWI and II era systems.

Was there ever a time when the engine did not do this , like when it was making far less HP?
I think a clue resides in there !
Anyway a motor like this needs a custom made crank scraper that is lined with Teflon.
It should be set up to rub lightly and it will ware in with no issue and provide less then ,010" clearance when that has taken place!

Yes! When it made far less power! That being said I've never had one of these engines that didn't like pushing the dipstick partway out the tube. The breather system is a common topic of discussion on the forums for my engine.

...
Interestingly, the engine does not push oil out the breathers when I'm just rowing through the gears at the road course or on the street. Only sustained high rpm and load. ...

While drifting?

Yes, while drifting. I have not tested this steady-state on the dyno as I'm not confident I'll be able to keep temps under control long enough to make an informative test.

Controlling oil flow and controlling gas flow are two related but separate problems.

One thing to consider is where to vent the crankcase gas out of the engine. With high g-forces, I think you have to go thru all operating modes one by one. That's what we tried to do (with a street car, no drifting there, or inverted flight!), and had some success. For example, when accelerating hard under boost, you probably want to vent the crankcase gasses out of the front of the engine. When braking hard and under manifold vacuum, you probably want to vent the engine from the back of the engine. This is because the oil sloshes to the back of the engine under acceleration and to the front of the engine when braking. Under lateral G's, you want to vent from the inside side of the turn, maybe with a two-inlet one-outlet gravity ball valve like in the piston airplane inverted flight systems.

I wrote a couple of posts on our system here. It would be completely inadequate for your use, but maybe you'll get some ideas from there:

http://rennlist.com/forums/928-forum/56 ... st13127650
http://rennlist.com/forums/928-forum/56 ... st13127650
http://rennlist.com/forums/928-forum/56 ... st13128484
http://rennlist.com/forums/928-forum/56 ... st13128486

Thank you! I will have to spend some time reading these!

This passage caught my eye:

"The lateral accelerations are relevant for the following reasons. In a 90-degree V8 engine, the oil drains from the cylinder heads are in approximately 45-degree angle relative to the vertical plane. In a 1g lateral acceleration, from the oil drain perspective the engine is effectively turned sideways with the oil drains horizontal on the outside turn bank of the engine. In this situation, there is no gravity drain, and the pressure differential between the crankcase and the head mostly determines in which direction the oil flows. If the crankcase pressurizes and the heads are relieved, it is logical to assume that the oil will flow into the heads instead of the sump."

I would imagine a canted inline engine might suffer from similar effects! Although the tilt of my engine wouldn't be as significant as one bank of a V8.

A simple system to achieve such 'switching' was credited to the Petty operation ~50 years ago. He/they built a U tube that straddled the L&R rocker covers with a tee fitting in the straight transverse portion, which was a very loose fit over a large (1" or so) ball bearing, but the curved-down ends were 1" OD so on a left turn the ball rolled to the right, sealing off the outboard cover and vice-versa.

It would be a bit Rube Goldberg, but if necessary, such a rig could be plumbed not into the rocker covers, but into a pair of similar longitudinal breathers joining the front and rear of each cover.

I'm not sure if this will solve my problem. The valve covers do have baffling over the vents and even labyrinth passages as well but they haven't proved to be very effective so far. The ball valve may just determine which route the oil takes, hehe.

Found it.

http://www.bufkinengineering.com/Catch%20Can.htm

An interesting approach to crankcase breathing that evolved through trial and error and analyzing the situation. Went from filling a catchcan in less than one track session to nothing but a little oily water vapor.

Thank you! This is awesome. I wish I'd seen this design earlier. Looks like it would be absurdly effective!

I may have to just go ahead and build myself one anyway...

What's the largest amount of oil vapour we can hope to remove from the blowby gas? If I routed such an AOS into my pre-turbo intake pipe I could set up a nice one way flow path in the valve covers and out the crankcase but I'm very reluctant to contaminate my intake charge with oil. And also coat my intercooler system in it...

On water cooled VW Vanagon we stuffed stainless scrub pads in the vertical oil breather tube to slow down the velocity, sorta like the materiel in an old bath air cleaner, figured it turned little drops into big drops which would then run down
do you have a baffle under your breather hose hole

I've considered this type of thing as well. I still may stuff my own AOS with them but haven't settled on the final design yet. (Not to mention everywhere I go seems to be sold out of SS scouring pads right now.)

Oil/Air separator tanks vertically located (drain back) to each vent should help a great deal. In steady state testing I've seen simple systems like this reduce blowby accumulation on a well sealed engine 20 fold and more versus a direct path breather system to the accumulation tank. You're situation has the added difficulty of best case breather location as well... gdlk!

This is good news for me! So are we in agreement then that this is a relatively common problem and that, short of me having damaged a piston, we're bound to run into it with this kind of turbocharged engine?

[MadBill]

What are you using for ring gaps? Is there an accumulator groove between the top and second ring and is the 2nd at least 0.004" more than the top, to prevent pressure build up and ring flutter? Similarly, I don't know how much boost and thus bore distortion might be present in your engine, but JIC, I'd try Gapless™ top rings in the next rebuild. Maintaining an effectively zero gap despite possible cylinder panting might do the trick.

By way of diagnosis, you could run the final vent line through a blow by/flowmeter and video it during a run to get a handle on what you're dealing with. A lot of really good race engines produce less than 2 CFM at 700 HP or more.

Also, here's a good article on separators: http://www.enginelabs.com/engine-tech/e ... r-control/

[Kevin Johnson]

The reaction wall force of the RWD pan in oversteer is likely causing the rear of the sump to become a de facto second reservoir and the oil is being churned by the rotating assembly and from this is overwhelming your system.

Try adding angled ribs to the floor of the pan to actively transport this oil back to the sump contra the prevailing lateral acceleration to the vehicle. Porsche, Mercedes and Ford all employ this technology.

Also, if you have not already done so, get rid of the hump at the rear of the floor. It will launch oil into the rotating assembly.

[Cubic_Cleveland]

Ahh CA18! Yep, those have a lot of problems venting crankcase pressure without taking oil along for the ride. The block vent seems to be most problematic. Is the factory deflector definitely in place? I can't remember, but aren't they a single stage baffle with no gauze? If so, I'd be making a better baffled system that incorporates gauze.

Also, have you enlarged the breather baffles/circuit in the rocker covers? Generally you can make factory systems larger and still very effective to help with the RPM you're turning. I would also put another breather on the exhaust cover, either in the factory location, or at the front and don't be afraid to make them large. Just make sure they are well baffled!

[Buddyworm]

Well my new separator didn't perform as well as I'd hoped. A longer drift run at max power was enough to overwhelm the system. It wasn't a complete failure though, the volume of my new can is smaller than my previous unbaffled can and coped as well as before.

Short of pursuing Kevin Johnson's suggestion of further baffling the oil pan, which I may have to delve into next time I have it apart, my quest for a solution continues!

Upon examining a spare set of valve covers I found fairly extensive baffling almost the whole length of both covers. The baffles are open at both ends. Since there's likely oil pooling at the rear of the head at high revs and the rearward G's under drift I should benefit from blocking off the rear off the passage created by the baffle, forcing the blowby gas to take a path only through the front of the cover. Am I on the right track here?

As well, how important is a check valve in the oil return drain? I would imagine that there's lower pressure in the can than in the crankcase so the valve will tend to be closed against outflow under max power and RPM. To what extent can I count on the column of oil standing on top of the check valve combined with the pressure in the can overcoming the crankcase pressure and allowing drainage? Is it even possible to have the can drain as quickly as it fills?

What are you using for ring gaps? Is there an accumulator groove between the top and second ring and is the 2nd at least 0.004" more than the top, to prevent pressure build up and ring flutter? Similarly, I don't know how much boost and thus bore distortion might be present in your engine, but JIC, I'd try Gapless™ top rings in the next rebuild. Maintaining an effectively zero gap despite possible cylinder panting might do the trick.

By way of diagnosis, you could run the final vent line through a blow by/flowmeter and video it during a run to get a handle on what you're dealing with. A lot of really good race engines produce less than 2 CFM at 700 HP or more.

Also, here's a good article on separators: http://www.enginelabs.com/engine-tech/e ... r-control/

I don't know my ring gaps off-hand as I did not build the engine but if I remember correctly they are within OE spec. This engine is absolutely not what I would call a "really good race engine". Miniscule budget and all that... This is more of a, "push the stock package as far as it'll go with bolt-ons and tuning and see how it holds up," engine.

[benno318]

im not familiar with the ca18 but i built a mazda bp turbo engine and the place doing the tune was reporting excessive oil use/smoking and telling the customer it was poor ring seal/honing/bedding in procedure. they ended up putting a temporary catch can in instead of letting it vent to the turbo inlet and measured it at about 100ml of oil "puke" for only 2 or 3 dyno pulls.

i asked the customer to bring the car to me and checked a few things out. this particular car was a factory turbo effort (se mx5) but they only ever ran about 5 or 6 psi in stock form, this one was now running a little over 20psi. so in looking at the breathing system there was a pcv valve (which obviously shuts off under any boost) a breather from rocker cover to a separator, the breather from the separator to the turbo inlet piping, then a drain tube from the separator to the sump.

the breather in the rocker cover used only small hose, as in 3/8 OD. checking things out, i removed the hose from the separator to the sump and blew into it - it "bubbled" in other words it returned the oil UNDER the oil level in the sump. i was intending to increase the rocker cover breather outlet size, so i removed the rocker cover, removed the plate covering the chambers that separate oil from mist, and looked at the passages in it too. just terrible. one side camshaft bank vented into a separation chamber with just a single hole approx 4-5mm diameter eventually joining up to the other side where the main breather outlet was. i could not believe the tiny size of the holes in the rocker cover for blowby.

my thoughts were that under boost, blowby was building up due to the tiny connecting holes in the rocker cover passages, and forcing oil backwards up the drain into the separator and from there into the turbo inlet. my solution was to drill extra and larger holes in the rocker cover allowing the blowby to reach the breather and do its job properly. PROBLEM SOLVED. no more smoke, no more oil usage, a happy customer and no-one needed to pull an engine down unnecessarily.

so the breathing passages is one of the first things i would advise to check.

[ptuomov]

Just a quick update. Take a look at the linked video at about 10:45 mark:

https://youtu.be/LnavQHFTmk8?t=10m45s

This is out (positive) experience with the crankcase breather routing in the links of this post: posting.php?mode=reply&f=1&t=46588#pr607608

[jlbayes]

Found it.

http://www.bufkinengineering.com/Catch%20Can.htm

An interesting approach to crankcase breathing that evolved through trial and error and analyzing the situation. Went from filling a catchcan in less than one track session to nothing but a little oily water vapor.

This is excellent. Will be changing my crank case vent system around to mirror the setup in the link. I have similar issues with oil control due to the 4v layout as well as extremely high boost pressure.

[n2xlr8n]

Anecdotal, I know, but...

I'm a big fan of vacuum pumps, but I replaced my vacuum pump with one of these in 2011 and it worked flawlessly.

https://crawfordperformance.com/collect ... ion-savers

I wish I had cut it apart (pricey experiment), but you guys with knowledge of such things can tell what they are doing inside by viewing the installation instructions.

Subaru EJs are notorious for poor oil control in stock form.

[CamKing]

I would bet that the head only has drain holes on one side, to drain the oil back down to the crankcase, and that the G-forces while drifting, are pushing all the oil away from those drain holes, keeping it from draining back down, and that's why it's coming out the breather.

[ptuomov]

I would bet that the head only has drain holes on one side, to drain the oil back down to the crankcase, and that the G-forces while drifting, are pushing all the oil away from those drain holes, keeping it from draining back down, and that's why it's coming out the breather.

I've got a 90-degree V8 engine, and I've tried to solve this potential problem by connecting the valve covers with a hose.

[CamKing]

I would bet that the head only has drain holes on one side, to drain the oil back down to the crankcase, and that the G-forces while drifting, are pushing all the oil away from those drain holes, keeping it from draining back down, and that's why it's coming out the breather.

I've got a 90-degree V8 engine, and I've tried to solve this potential problem by connecting the valve covers with a hose.

Get a belt drive, single stage "return" pump, and have it pull from the valve covers(or right below them), back to the crank case.

[ptuomov]

I would bet that the head only has drain holes on one side, to drain the oil back down to the crankcase, and that the G-forces while drifting, are pushing all the oil away from those drain holes, keeping it from draining back down, and that's why it's coming out the breather.

I've got a 90-degree V8 engine, and I've tried to solve this potential problem by connecting the valve covers with a hose.

Get a belt drive, single stage "return" pump, and have it pull from the valve covers(or right below them), back to the crank case.

So far, it looks like the passive methods we've put together for this engine have in fact solved the problem completely (on street tires and with low 6000's rpms).

[bentvalves]

MY GOD, speedtalks new format is AWSOME.

look at this gem that I just happened to bump into god knows how.