EFR 6758 leaking into compressor housing
 

I have around 2000 miles on my kraken 6758 setup, maybe 500 miles on the forged engine on a previous setup. Photo below. Most of this driving was done with a stock PCV setup, although after noticing the oil, I cleaned the charge piping and deleted the PCV. Stock breathers in the valve cover, PCV open to atmo. As you can see, that’s a temp setup while I wait for the proper fitting from radium, all the hoses are ready once that shows up.

I let it idle like this long enough for the thermostat to open (~10+ min) after wiping the turbo clean. After sitting, theres a small build up of oil below the compressor wheel, you can see similar oil in the pic of the compressor housing.

Oil feed is -4 from VVT housing, goes through a Turbosmart turbo oil filter, pressure at it's source (the block) reads around 45psi cold, 17psi hot, usually 45-60psi when driving. Running Total 5-30 synthetic. Exhaust breather has always just been open, never seen any evidence of oil in it to cause concern, although I may put a small catch can on it for fun. Leak down was 3-4% when I checked it a few weeks ago. It is an aluminum super core, looks like the compressor housing was built in 2020, super core in 2016.

Any idea what could be causing it? I'm hesitant to blame the super core after reading up on this thread:
https://www.miataturbo.net/engine-pe...-please-90835/
and this one:
https://www.miataturbo.net/build-thr...1/#post1276378

There is some shaft play I'll post a video of, but only when hot. Cold there's barely any, similar to what @psyber_0ptix describes in his thread.

https://cimg4.ibsrv.net/gimg/www.mia...c563c39cd.jpeg
excuse the mess, heater core sprung a leak, waiting on catch can fittings, and a few things were ziptied out of the way for compressor housingless running.
https://cimg7.ibsrv.net/gimg/www.mia...aa1fa3471.jpeg

Here's the promised video. I had to use a rag it was so hot, prissy mechanic hands. I start by rocking up/down, end of the video I'm pushing in/out.


I'll also say I've done most of my driving lately at 18psi, which is a little north of 325hp. 30+ miles round trip commute to work, usually with multiple pulls to red line. RIP 6-speed.

It's enough oil that the BOV is coating my sway bar below it.

Pic and specs on drain?

Sadly don't have much that comes to mind that might help. I couldn't find any shaft play specs but I think you're justified in thinking if it's tight while cold you're probably fine. It's not much help, but from my reading of this guy it looks like the nominal behavior (under some load at least) would be boost blowing by into the cartridge and not the other way around: https://www.borgwarner.com/docs/defa...ining_book.pdf . That said if the oil accumulation is only at idle there would be an expectation of sooome oil leakage, but certainly not coating your sway bar amounts. Maybe there's thinning due to something like fuel contamination? How old is the oil/were you using a thicker weight previously? Only other thing that would come to mind is an oil drain restriction (or drain clocking angle so drain location is higher than bottom of the oil seal)?

https://cimg8.ibsrv.net/gimg/www.mia...aecbebd5c2.png
https://cimg9.ibsrv.net/gimg/www.mia...afe2848212.png
https://cimg6.ibsrv.net/gimg/www.mia...cca5eff48c.png

Ah, yes. It’s a -10 off of fullrace’s EFR adapter. I believe it’s 1/2npt, but I’ll get a pic

From your second link in the first post it seems that your drain is too small. I'm not sure why they tapped the oil drain in the cartridge but I think I read somewhere that if you use the tapered threads you absolutely have to use a scavenge pump. I bet fullrace adapter isn't any better...

https://www.miataturbo.net/diy-turbo...l-drain-93944/

There's the scavenge pump post.

I think AN on the drain is a hell of a lot more trouble than it's worth.

Good link deezums.

I’ve got a lathe, I can open up my adapter and/or the flange. Chra is easy enough to remove at this point

Pics of drain. line is -10, so .625 ID. The turbo drain is about .575" ID, the steel 3/8NPT to -10 adapter is much smaller, a little over .500" on the -10 side, and .400 on the NPT side. I have access to a lathe, I'll try opening it up and see if it has enough strength to tighten and seal properly. If not, I can weld a -10 bung to the aluminum adapter to skip the NPT portion.

https://cimg9.ibsrv.net/gimg/www.mia...359175ca6.jpeg
https://cimg0.ibsrv.net/gimg/www.mia...3ca635c72.jpeg
https://cimg1.ibsrv.net/gimg/www.mia...84e315cc2.jpeg

Yeah, those threaded adapters are known to be problematic. The pan side fitting looks like a 90 degree too, so that is not recommended. A welded flange fitting is my favorite way to go. I like to extend it enough to clear the compressor housing so I can get a wrench on it.

Edit: pic of 12AN


https://cimg5.ibsrv.net/gimg/www.mia...e6ef76e494.jpg

Location of 1/2NPT bung welded to pan requires a 90 to meet the turbo. I can extend down and maybe make it a 45/45 instead of 90/30 fitting.

That's a better drain than most have. Could be better but hard to believe that's the cause of your issue.

Nothing really stands out to me, but my guesses would be:
1. Excessive shaft play? I think mine too has a tiny bit when hot, but not much.
2. Excessive backpressure? I had to mod my breather on mine as I had way too much backpressure.

Clean looking setup, sir! Hope you get it figured out quick.

I agree with Pat, your drain does already look better than most. The observation that sticks with me is leaking at idle. If you had enough blow-by at idle to cause a problem, I would think it would launch the dipstick at 325hp. I think it would absolutely puke out that ex side port there too.

Did you confirm that you are clear of the pick-up tube where your pan fitting is?

Turbo seals may be the only other reasonable culprit.

Hard to think there would be enough blow-by or back pressure to cause serious oil accumulation with just ~10-15 min of idling? I mean it's not a labyrinth seal or anything, there will always be some leakage, but other than for cooling the ball bearings really don't need much oil to function (hence the tiny built in restrictor from what I can tell). From what I can logic, the difference between something like a 0.5" and 0.6" ID drain in terms of flow restriction should not matter significantly at the pressure/flow rate I'm guestimating BW designed for from reading their docs and the relative size of the supply restriction.

Is the oil pump boundary/shimmed? I know its a no-no according to BW, but two things to try would be a small (magnitude wise, so large orifice wise) restrictor on the oil feed and a test to verify overall flow rate (to make sure you're not reducing the cooling power significantly with the restrictor/see if it's excessive with the current setup). Would it be difficult to blank off the drain on the pan and idle the car (already warm) in the air with the turbo drain going into a small pail for a minute or two? Could measure vol & time to get an idea of what might be going on.

Well, I haven't posted here since I started a build thread way back when and then didn't update it ever, but that's usually how things go for me. A friend pointed me here after seeing some issues I was having with my own 6758 setup, so I thought I'd add in my own experience as another datapoint:

https://cimg5.ibsrv.net/gimg/www.mia...9c08a6d94c.jpg
I got my Kraken high mount manifold installed along with the 6758 from Turbokits and built all the plumbing. Decided to run a -6AN oil line via advice from the BW tech docs saying that runs over 18" should use that size line. Hooked up a drain with a Vibrant bolt-on -10AN flange to the CHRA and a welded-in -10AN bung on the pan.
From the first few days I was running it, I noticed oil collecting in the turbo intake (actually inside the plumbing to the air filter) and got concerned. Did a bunch of research while doing some light tuning and the oil persisted. The intercooler was also collecting oil.

https://cimg0.ibsrv.net/gimg/www.mia...319472ea28.jpghttps://cimg1.ibsrv.net/gimg/www.mia...7edea18609.jpghttps://cimg9.ibsrv.net/gimg/www.mia...6368a32eae.jpg

After reading a lot about oil drains and how critical they were, I decided to change from the Vibrant flange (which admittedly was a little on the small side) to the flange Kraken provided in their plumbing kit (welded pipe to a wide open flange bore). At the same time I switched from the braided line I'd been using to the Kraken-provided line, still into a 90-degree bend into the pan. I measured the diameter of the AN fitting, and it was .485", which seems to be large enough compared to other similar setups.
https://cimg5.ibsrv.net/gimg/www.mia...f51c51d4cb.jpg
I pulled the compressor housing and it had oil in it, but I didn't start and idle it without the compressor housing installed as earlier in this thread, so not sure if it was leaking at idle. I'd guess it may be acting similar though.
Since I had to pull the CHRA to swap over the flange, I took a look at the turbine. It didn't appear to me that there was any oil leakage from that side, but I'm not very experienced in the ways of turbos, so not sure if the dark soot in the sealing area is from oil or just rich burning that hasn't gotten hot enough to burn off.

https://cimg4.ibsrv.net/gimg/www.mia...bdd7c3d135.jpg
After changing my oil drain out, I still had the same amount of oil gathering in the air filter and also in the intake piping, and so decided that either the changes I'd made to the oil drain hadn't worked, or that wasn't the source of the issue.

I saw Ryan_G's build thread, in which he was having similar issues with an EFR leaking oil out of the compressor side: https://www.miataturbo.net/build-thr...5/#post1382850
I'm unsure if it was ever resolved, as I think the car ended up getting sold after the CHRA was rebuilt.

A couple days ago I resolved to try changing my oil feed out from -6AN to -4AN to act as a bit of a restrictor (without actually installing a restrictor) and I found that it actually worked somewhat.
Oil in the filter has been reduced to several drops instead of pouring out. There's still oil getting through though, which leads me to believe that it may still be over-oiled. The next option would be to put an even larger oil drain in, but I'm not sure about going to a 3/4" as Ryan_G had done in an effort to exhaust every available option. It seems extreme, but if it's what's required, then I'll pull the engine and put another hole in the pan.

More info not included in the saga above:
Running a BE VVT pump with 70 PSI shims
Current line routing Block -> BSP to NPT adapter -> NPT to -4AN adapter -> -4AN line (18" length) -> -4AN to Turbo adapter
My oil pressure gauge is out of the loop while I test this smaller line to see what effect it has.
Watching the oil pressures on the -6AN line, pressure would hit 60-80 PSI on cold start, but during warmup to running it would drop back down, and the turbo wouldn't be fed over 40 PSI warm.
I have some shaft play while cold, i can push the shaft around maybe 0.25 to 0.5mm, nothing much but just noticeable. I haven't tested while hot.
I have a genuine Mazda GTX PCV valve in there (from the dealer) and that's routed appropriately, and this is a fresh rebuild, so there shouldn't be much blowby, though it's possible.



The next steps for me are to test with a Turbosmart oil pressure regulator I have, and then move to a larger drain line. I have a feeling after seeing the pressures it was being fed with, the regulator would do little more than act as another restriction in the system and possibly provide lower flow (and may mask the actual issue by reducing oil flow). Drilling a bigger drain hole in the pan may help too, but that's a significant effort, especially at that size (I'd be worried about getting chips in the pan, so I'd probably pull the engine to do it).
I'm a bit hesitant on installing the regulator, as it's a drain-back regulator, and thus would require it's own oil feed-back into the pan. I have plans to do a temporary test setup using a replacement oil cap with a hole drilled/tapped to accommodate the return line.
Maybe I'll try pulling the PCV and running a vent for a short bit to see if that has any effect on the amount of oil to validate if there's crankcase issues as well. Thread a pressure sensor into the oil cap, even!

Interested to see that this issue has happened among a few turbos here, and curious if it's just a single problem or the result of several things (too much oil, too little drain, too much crankcase pressure) all coming together to cause this. I really don't know how much oil usually leaks from turbos, but seems like more than there should be.

Turbosmart OPR with a VVT actuator feed , restrictor, and -4 lines

https://cimg9.ibsrv.net/gimg/www.mia...0c32cb8802.jpg

That's really clever, using the VVT actuator and welding the opr flange to the valve cover. Much cleaner than how I did mine. I ran hardlines from the oil pressure switch and a softline to the pan. I wish I thought of doing it your way.

https://cimg2.ibsrv.net/gimg/www.mia...240dca5ba4.jpg
https://cimg3.ibsrv.net/gimg/www.mia...6a977c2fd7.jpg

I do have a -6 bung welded to the pan for a catch can drain, un used currently. Could easily use that for a opr return.

ted, that’s where I source my oil too, but I took the housing off/apart, and put a NPT tap into it, same thread pitch as BSPT, no need to drill it. Less fittings for you.

to be clear, I am not advocating for the OPR. I have several iterations on my design, early versions suffered from compressor side oil leaks, hence the giant in-manufacturer specification drain and the OPR. I also did quite a bit of work on the valve cover and actively monitor both crankcase pressure and turbo oil pressure. Suffice to say, I no longer have oil leaks from the turbo.

There are many builds out there that don't have reports of compressor side oil leaks, so I have to think the OPR is not necessary. I also don't know of any other -12 drains.

If you do decide to go OPR from the VVT actuator, I recommend a moderate restrictor, as it allows the VVT mechanism to get the stock oil pressures.

Honestly, my first step would be to get rid of the pan side 90 degree if possible and ditch the AN-NPT adapter flange (or bore to 0.5"). This assumes you confirmed that the pan drain port isn't blocked by the pick-up tube.

P.S. you don't want the stuff that comes out of a catch can going back to the sump anyway, it is very polluted.

PPS (Edit) - The fittings at the actuator facilitate the restrictor

For sure, I opened up the NPT portion of the adapter today, took the ID from .400" to .490". Sounds small, but that's almost a 50% gain in area. It was free and fairly easy, I'll give that a try before anything else drastic. And to clarify on the pan fitting, it was meant for an AOS drain back before I got a free radium catch can from a friend.

The -10 fitting at the pan was a 90* fitting, I opened it up a bit, and I haven't leaked any oil from the car sitting with the line off, so it's well above the oil level. The 1/2NPT bung was welded at a very slight upwards angle

I have fuel and oil pressure sensors wired into the MS, I could easily repopulate one for some sort of crank case pressure monitor temporarily, I had already thought of doing that when this issue popped up. Can you explain your tests any more and what you found? Although, looking back, I see the sensor on your valve cover now.