Has anyone encountered excessive amounts of oil smoke on entering a higher than normal G turn? I'm actually surprised I haven't found a thread or post in track focused threads, though I did find ones relating to overfilling the oil. And I think my issue's symptoms are different, given that it's only happening right after a long sweeper or quick, sharp turns and not on straightaways.

I've got an on-ramp on my way to work that does a left hand 270° turn to merge with a highway, and whenever I'm pushing the car, upon acceleration out of the turn, I get a magnificent smoke show regardless of RPM band. The first time it happened, I thought a seal had failed or something and immediately pulled the car over and had it towed. Then it happened consistently (albeit somewhat less drastic) at an autocross event, where accelerating hard after making sharp turns would produce a nice cloud out of the tail pipe. My previous worn engine would have blow-by if I accelerated hard going in a straight line and didn't correlate to turns, but a straightline only hard acceleration hasn't happened yet with my current one. It's only happening under conditions that would cause oil sloshing or pooling to one side.

I bought a catch can yet to be installed, and I'm going to try running the car with the oil level a tick mark or two lower than full. I figure those two things done independently should confirm my suspicion that: 1.) The oil is entering through the breather tube line and not somewhere else, and 2.) that the pooling or sloshing must somehow block a pressure escape and results in oil reaching the breather tube.

In case anyone wants to throw out other suggestions, I'll mention my compression is okay, about 105 average ± 3 psi per rotor face, and it's been roughly 7500 miles since the rebuild. I have a Sohn adapter with Idemitsu, an AEM intake, working order intake valves, and a UIM from an 04, though mine is an 07. I'm using 5W-40. As far as I know, the rest of the oil system is stock.

 

LH turn would be away from the breather galley on the center iron, but if you pull the accordion tube at the TB then here should be signs of oil entering from hose at the oil cap breather tube connection if it’s from there. if not then the catch can won’t help. Never had that happen even with oil sump over-fill. I’d suspect it to be something else, but please confirm. Maybe block/remove the Sohn and try to see if it stops because I never ran one of those either (and won’t).

The ‘07 UIM has the additional PCV breather point added to it for the milky oil substance TSB issue, but mine is an ‘05 and never did that. If it is coming from the sump then the next step is to determine why. Because that would have to mean excessive blow-by when cornering, which again is not normal with an engine with good seals. It may be a rotor oil seal issue, though I’m not sure how that would manifest itself when cornering.

.

 

The oil fill breather tube runs right into the LIM ports on mine. The catch can should intercept the liquid oil from the vapors going to the LIM. I shouldn't have the additional PCV point since I'm using the 04 UIM, and I know for sure when I had the original 07 UIM, it was capped because of the AEM intake / VFAD delete. I only have 2 hoses running from the "accordion tube", the third is capped.

I'll verify there's no oil at the TB, but I've never seen evidence of it being there (as there's no hose for oil to get there). I couldn't think of a reason the Sohn would/could ever produce oil enough to cause this condition, but I'll try running it dry and see what happens. The first time this smoke show happened, the dealership the car got towed to diagnosed the plugs as fouled, which I thought could have been from too much 2-stroke (Sohn and was supplementing with mixed gas). But that doesn't explain why it would only do it on cornering.

I appreciate the pointers. I'll see if I can rig up a way to video, as I'm convinced the smoke is oil smoke, but I'm also slightly colorblind...

 

ahh, that was part of the ‘07 UIM breather configuration.

Without that UIM, then that hose assembly should also be deleted with the nipples capped just like the earlier vehicles. That’s likely to be the issue then.

Either that or the later UIM needs to be installed. I don’t think that hose to the LIM nipples is going to function properly without the other breather port and hose that was added where the UIM (not actually the UIM, but the PCV oil galley and intake valve vacuum tank assembly that bolts to the UIM) bolts to the center iron breather assembly. That’s what was different between the earlier and later S1 UIM.
.

 

Early S1 PCV-vacuum tank on the left, later one on the right; note the large breather hose coming off at the lower RH corner and turning 180*. This is the new breather port & hose that was added in addition to other hose going to the nipples on the LIM.






Here’s the TSB for it.

https://www.rx8club.com/attachments/...-dipstick-.pdf

.

 

just for the record, most of the mileage on my car is from street use and to say 99% of RX8 owners had this problem and benefited from this change is just more forum lore not borne out in either truth or factual data …

so what defines most people who have this issue? Living in a cold climate area with high moisture and/or not getting the engine oil temperature high/long enough to burn off moisture from the atmosphere that might accumulate in either the engine sump and/or oil fill/pcv chamber on the UIM.

There are a lot more than 1% of RX8 owners who live in warm or hot climate that didn’t ever have this problem. If anything they were fighting too high of oil and coolant temperatures.

.

 

I know no one wants to discuss this topic further, but if anyone has the patience to explain, I'm still having a hard time trying to understand exactly how the PCV system is supposed to work. Or maybe I do and it's not clicking.

Here's what I've put together. At a basic level, the oil fill neck (or PCV vacuum tank) just funnels oil to the oil pan. It's got one or two ports (depending on model year) built into it to allow for ventilation of oil vapor and condensation. These ports are under vacuum, but not enough that it could pull oil up like a siphon (right?). Supposedly it's strong enough to draw unwanted vapors, but in order for it to do that, there has to be a flow path. It can't just pull on the air mass in the tank and evacuate the gasses inside. Like sucking on glass bottle, you can generate a vacuum, but nothing is flowing. So does that mean there's air/gas that's being provided somewhere in the oil pan? Is there always a miniscule amount of blow-by present that keeps the gasses moving? It is called Positive Crankcase Ventilation, indicating there is always some positive pressure coming from blow-by. So does a healthy rotary not have ANY blow-by, or maybe a miniscule acceptable level? At higher RPM, you'd expect more blow-by than at low RPM, so does that mean there's more airflow through the PCV system at higher RPM?

I'm not seeing how our PCV system is necessary, if, by design, blow-by shouldn't be happening.

As for diagnosing my issue, here's what I'm thinking. Liquid oil should stay in the pan, and should never get past the bottom flange that connects the oil fill vacuum tank to the engine's mating flange, even under high G turns. Sloshing I suppose could get droplets up there, but unless the car was upside down, I can't imagine oil being able to flow all the way up to that point even under high speed cornering. Like Team has mentioned, he's undergone high G turns with an overfilled oil system, and not had oil reach his PCV port. You'd have to seriously overfill your oil to the point it's flowing out of the dipstick tube to get it in the oil fill neck under a high G turn. In other words, if the car was put on a G-force simulator without the engine running, you'd have to pull some insane G's to get the oil to overcome gravity and get up to the PCV oil fill chamber. Does anyone know if this is true?

So, if I have a catch can full of oil, oil is either being drawn up by intake vacuum, pushed up by blow-by positive pressure, or a combination of both unless I vent the catch can and don't have it recirculated. In the latter case, it would have to be positive pressure pushing the oil up, indicating bad blow-by. Does this sound right?

So regardless if I have one or two PCV ports, if I vent the catch can to atmosphere with just the oil fill neck tube connected, and STILL end up with lots of liquid oil inside, I should assume bad blow-by is occurring. I should also get rid of the idea in my head that it's the intake vacuum pulling oil through the PCV system because under a tight turn, the oil that's made it somewhat up the filling canal has blocked whatever normal flow path the gasses take.



Also found a similar thread in doing some more searching that lead me to write this post.
https://www.rx8club.com/series-i-tec...ocross-265710/

 

As an update to anyone viewing this thread, I believe my original issue has been "mitigated". I take the same way to work M-F every week, so there have been plenty of right sweeper possibilities for it to happen. I'm fairly certain this was an issue relating to pressure, and here's why.

When I did my next oil change, I made sure to not fill past the full mark (which I typically overshoot by about 1/8th of an inch on the dipstick). Still had a smoke show event on a sharper on-ramp that it happened on, not my usual commute one. So I concluded it's not overfilling the oil that's causing it. So I threw in a catch can with a breather filter, and capped the LIM ports. Since then, there's been no incidents. I haven't done any autox events with it installed yet, but a month of commuting and taking the same on-ramp sweeper has yet to even deposit oil into the can. So the next test is finding an empty road or parking lot, and giving the car a proper shakedown. I'm convinced now that exhaust gas was finding a way into the oil system, and having the breather filter alleviates that pressure constantly without oil consumption. I know my engine isn't in perfect condition as it's a rebuild with average compression (little over 100psi on all faces), so blowby is bound to happen.

I plan to install an oil filler tank with the 2nd PCV port once I get some time and get it painted to make the bay look pretty. I'm going to try to tee that line into the catch can too and see what happens, though I don't think it'll make a difference as nothing will be under vacuum still. After that, I'm thinking I'll try swapping the breather filter for a line back to the LIM ports (or one of the three intake ports), and see if that draws any oil into the can. If it does, and swapping back to the filter doesn't, that should jive with my theory that it's a combination of positive "crankcase" pressure and intake vacuum that draws oil up and causes the smoke show events for me.

 

Just got back from the Tail of the dragon and had the same issue! Which is surprising for a few reasons:
1) I have autocrossed this car extensively for 3 years and never had this issue on stickier tires than what I ran at DGRR.
2) I have run the tail of the dragon numerous times in the past and never had this issue.
3) I have a catch can installed (pulling vac from the 2 ports on the LIM and drawing from the nipple on the fill tube) -- it is sealed not vented

The issue only occurred when in 2nd gear for prolonged periods above 7k, it would start missing and a 007 style smoke screen would emerge from the exhaust, if I shift into 3rd it went away by the next corner and didn't come back. After the first time it happened I drained the catch can and a good amount of oil (6-8 oz) was inside -- to the point that the low oil light came on the following morning, so I agree we can eliminate "overfilling" from the list of possible culprits. Topped of the oil to the proper level and it happened again multiple times the next day; shifted to 3rd and again it went away. After that I used 2nd gear like a "scramble boost" button and pretty much just kept it in 3rd.

My theory is:
A ) The cheap Amazon catch can I used is too small and of a poor design.
B ) Sustained high rpm operation creates strong vacuum at the LIM ports.
C ) Sustained high rpm operation also creates high crank case pressure.
D) High "angle" (not necessarily high G) cornering causes oil to slosh in such a way that it covers the tube/pickup/nipple completely
E ) Due to the sealed nature of the system the LIM sucks in oil like a straw =

Possible Solutions:
Fist step is a better catch can. A larger can would provide a high margin of safety in general and if there is a can with some kind of anti-suction design that prevents oil from being drawn into the suction port that would see to be ideal. I don't want a can vented to atmosphere because i don't want the LIM potentially sucking in unmetered air but I do want to actively evacuate fumes if possible.

 

Is it possible your engine is tired and having more blowby than before, therefore raising the crankcase pressure at high rpm? Does it also happen on straight lines without cornering?
I'm not sure routing the catch can as you have done is a good idea, they're meant to route to the intake before the TB. The way you've done it, when the throttle is closed you're creating suction in the crankcase which is.. weird, maybe not good for crankcase seals long term. Route it normally (there diagrams on this site) and see if it still happens.

 

Compression tested the day before I left and it's 100-105 psi across all faces, so that is unlikely.
The ONLY time anything similar has ever happened was when I first got the car and slightly over-filled the oil, it let out a smoke screen on deccel from 80 mph down an offramp, and that was in the stock configuration with no catch can. It hasn't happened for 8 years since under any condition.

Appreciate the feedback, The main reason it was routed that way was to eliminate the possibility of oil getting sucked into the intake . Also partly because I didn't have the greatest confidence in the design of the cheap can..... which given the circumstance, seems justified. I will still upgrade the can to a quality unit, AND reroute to draw from the intake pre-tb .... shame it'll take a year to really see if it worked...

 

That's the thing, it IS routed normally....The catch can simply sits between the port on the oil filler neck and the two ports on the LIM which was stock in 2009. The hose on the bottom right corner of the vacuum chamber is still connected to the intake as shown in the diagram.

Still; to your point, the engine is constantly pulling vacuum while running and that only increases with RPM, the problem seems to be worst when transitioning from high rpm WOT to a tight corner, it seems like the closing of the throttle promotes the ingesting of oil that reveals itself on corner exit when I attempt to get back on the gas, and instead get a smoke screen.

I've been looking at options to mitigate this and came across a sealed catch can with a one-way check valve. If the problem is excess crank case pressure forcing oil into the LIM ports then this would seem to be a solution. However if the problem is excessive vacuum being generated by the LIM ports during high RPM & closed/low throttle situations, then I doubt this would help...
https://motionraceworks.com/products...bvDBBI0RrQ47Sb

 

Simpler thought, are you sure you are not just getting oil slosh up the tube? I only have had early S1 cars, but those S2 systems look like they could collect oil in that accordion flex vac tube. Which could then be sucked into the engine when you get back on throttle. Maybe buying an oil pan baffle plate would be more beneficial than a catch can that'll just fill up.

On my greddy turbo early S1 that most certainly has oil consumption/blowby, I dont catch much in my catch can even with hard cornering (offramps, mountain runs)......now that I think about it, I never installed a one way check valve. So the crank case is probably boosting and thats where my oil consumption is coming from lol. NVM i remembered the crank is plumbed pre compressor

 

Oh, I'm 100% sure oil IS sloshing up the tube, that's the only way it could make it to the catch can and then down into the LIM, the question is ... why, specifically is that happening? So far we can eliminate: Low compression (over 100 on all faces) and over-filling (it did this AFTER a 500 mile drive TO the Tail of the Dragon AND after dumping a complete catch can 4-5oz full of oil on Day 1) as possible culprits. Part of the problem is definitely the tail of the dragon itself as this has never happened even at autocross events pulling more lateral G's.

The diagram doesn't do a great job of showing the the inner workings of the S2 (really post-06') vaccum chamber but the entire reason it was re-designed is BECAUSE S1 cars pulled oil into the intake that doesn't happen with series 2. IF oil gets sucked anywhere it's into the service ports on the LIM which pull from the nipple on the oil filler neck. The catch can is supposed to prevent that , but clearly it's not doing it's job. I've never even seen an S2 oil pan baffle plate, and while I'm skeptical one would help, but I'm not completely opposed to the idea.

The catch can is pretty small, like 4-5 oz total, some maybe like 3 oz before oil is above the baffle plate, I suspect that is part of the problem. I could see a mostly filled catch can sloshing oil and covering to outlet, in a completely sealed setup there's only one place for it to go.
Also the open port on the oil filler tube is not ideal, I'd like to cap that and use a vented (and baffled) oil filler cap to feed the catch can, thus reducing the likelihood that liquid oil could make it to the can in the first place, although I've never seen a vented cap for the S2..... (as much as I appreciate the S2-specific improvements Mazda made to the RX-8, it does get annoying how many little changes they made for no apparent reason)

 

You'd think in the casting of the tube they could have just put in baffles easily. A baffle plate wouldnt be to much of a challenge to make, plenty of rx7 examples to look at for reference.

 

I Completely agree, but obviously they let the interns design the ventilation system, (hence why they thought dumping the unbaffled fill tube vent into the intake was a good idea in the first place ). Baffles in the fill tube would have been a better redesign than to simply change where the uncontrolled oil slosh ends up... but alas we're at where we're at... (On that note, I just noticed this is actually the "Series 1" Trouble Shooting sub-forum... lol).

I need to improve the catch can setup anyway, so hopefully this solves the problem. A pan baffle couldn't hurt, but I don't think it's necessary as the series 2 pan is a better design and oil starvation has never been an issue.
Take a look at the Ryan Rotary S1 Oil pan & Baffle: https://www.ryanrotaryperformance.co...es-1-2003-2008
Now take a look at their s2 "pan" : https://www.ryanrotaryperformance.co.../r3_sump_plate

The problem isn't oil not staying in the pan, the problem is that oil is escaping the block altogether. As long as it stays in the block I don't really care if it's in the pan or the fill tube (temporarily) as long as it isn't being sucked out and fouling my plugs. Trying to craft a custom baffle is a legitimate solution, however it's the most high-effort solution. Fixing the port / fill cap / catch can should be able to achieve the desired results with a fraction of the mess and effort, but time will tell...

 

Ehh not many people on the forums these days anyways.

I didnt actually know the S2 pan was so different. I figure just buy a plate and cut some holes in it to make your own vertical baffle to sandwich in place.

 

It was a PITA finding a vented & baffled oil cap or S2 since Mazda arbitrarily decided to switch from the threaded (s1) cap to a 1/4 turn cap on the series two. I grabbed this billet LS oil cap from Summit racing because it looked like it would work, but it was still a few mm too large and had to be turned down on a lathe. The O-rings were also too large, but those are easily swappable from the OEM cap. The nipple on the fill tube was capped in order to make liquid oil travel the most difficult possible path to ending up in the catch can/intake. I needed to be able to spin the cap without changing the orientation of the lines, so the Radium -10 ORB fitting was a necessity. The Catch can is an Improved racing CCS-7 and it is an absolute work of art! The high flow one-way check valve and Filter are also Improved racing, and probably not necessary, but in the unlikely event that positive crankcase pressure tries to build, it will be vented to atmosphere until vacuum is restored. Tested it out on a 1 mile long AutoX / Track Sprint this past weekend and it works flawlessly.