Question about oil screen box & RTV
 

I’ve been thinking more about the issues surrounding RTV pieces clogging the oil screen box. From what I understand, Subaru had the RTV problem with gen 1 86/BRZ models, which utilized a flat disc of metal screening to catch pieces of RTV in the oil. This flat screen could quickly became clogged and cause oil starvation.

Given Subaru’s manufacturing change that substituted RTV for conventional rubber gaskets, it seems likely Subaru knew pieces of RTV would inevitably find their way into the oil supply, and thus Subaru knew they would have to account for this issue more effectively in future engine designs.

So Subaru redesigned the oil screen for the current gen 2 GR86/BRZ, changing to a three-dimensional screen box, the closed end of the box at the top (closest to the oil pump) and the open end of the box pointed down towards the bottom of the oil pan.

The advantage of this approach is that the closed top of the screen box can be significantly clogged and the oil pump can still draw an adequate supply of oil through the holes in the four sides of the oil screen box. Unless the oil screen box became becomes virtually filled solid with RTV, there should be enough holes exposed to allow sufficient oil flow.

Note: The above theory assumes the oil pump has enough suction power to pull a sufficient quantity of oil through the sides of the oil screen box. Given that Subaru specifically designed this oil screen box to address the RTV issue, it seems likely they designed the oil pump to maintain sufficient flow when the closed top of the oil screen box is blocked.

Whew…still with me?

So my question is, why do we assume that chunks at the top of the oil strainer box could have caused anyone’s engine failure? The possibility that makes the most sense to me is that the oiling system is not maintaining pressure under high-g cornering (autocrossing, drifting, etc), because there is no baffling in the oil pan to ensure the oil screen box is always fully immersed in oil and thus always enabling the oil pump to draw oil up from the pan.

What am I missing? If the above analysis is reasonably correct (ha!), that would suggest the main problem is oil starvation under extended high-g cornering, and not so much the presence of RTV in the oil supply. So for average GR86/BRZ drivers who don’t engage in extended high-g cornering activities, perhaps there is nothing to worry about.

Thoughts?

If the engine relies on at least the surface area of the face of the screen to maintain sufficient pressure drop, then if the area on the raised sides is not equivalent then it may be insufficient to keep up under high flow conditions.


If the head pressure ahead of the pump is too low, the oil starts to cavitate. You can't push as much foamy oil as you can hard oil.

just remember, who tells the rtv to only clog the bottom of the screen?

I was thinking about this as well. This is not Suby's first rodeo with RTV and their challenges with sealing a boxer engine. I feel the redesigned pick up is a balancing act between oil flow and sealing the engine. I don't think the partially blocked pick up that we have been seeing is the only reason why these engine are popping. Having said that, if other mentioned is true, RTV can still block individual journals causing starvation situation.

The VAST majority of Gen 1 problems with RTV were after Toyota dealers did the valve spring recall and reassembled the engine poorly.

The rest of your post, it's just, I don't even know where to begin.

I don't buy this whole "they redesigned to a larger pickup because they knew it would get plugged" line of thought. That just is not a rational direction that design engineers would head in. They are not amateur's that need to find workarounds without using resources. It is in all probability just a simple case that they found the FA24 series of engine needed more pickup surface in general.
They managed to fix whatever the sealant issue was on the early FA20s so I fully expect them to do that again. If it wasn't fixed already that is.

There were a number that went well before the recall was ever even announced. When you take the known causes out of the mix there were still about 30 failures (reported on just this forum) that could, and in most cases were, directly related to sealant.
Of course we don't know the total number then or now but at this point there are only about 10 reported GR failures here so we haven't even reached to point we were at in 2015.

I have deliberately left the recall debacle out of the conversations and tallies because although the failure mode may be the same the root cause of the issue was not manufacturing.

From our best guesses, we are looking at what, less than 1% afflicted by that? So I would say that's still pretty rare.

Yep. I have no idea what Subaru/Toyota's acceptable failure margin is but I bet it is well over 1%.
If it hadn't been for the whole denied because of "racing" publicity this whole situation would never have even made the press the way it did.
Same as there was never one word from anybody at Subaru or Toyota back in 2015 when the first failures happened. They just quietly fixed it and carried on.

Yep. That's why I'm laughing at guys that are cancelling orders and such.

Any engine failure that happens with RTV is going to be warrantied and get a new engine. Even though it's exceedingly rare. I have my supsension here for it, the duckbill should be here today, going to order wheels in the next day or two.

And I'm going to drive it. If something happens that's what the warranty is for. I don't plan on tracking this one. Just a fun car since I haven't had a car the last 18 months.

We both work from home, so it's tough to justify more than one car.

I'm seeing the term oil starvation being thrown around a lot in these posts about the RTV issues in the pickup. I'm sure it has contributed to some failures, but have the actual starvation failures been documented? What I mean by that is, where are the pics of clogged oil ports, overheated bearings, over heated rods, etc... Something like this has a cause and effect. I'm seeing a lot on the cause, but next to nothing on the effect, which makes me question some of these failure modes.

We're just getting started on this and I'm looking for as much hard data as possible, to assure the true issues are properly dealt with. Tail chasing is no fun for anyone :p

As far as suspected lateral load starvation goes. Is anyone logging telemetry? Stock and lightly modified setups should be well within the capabilities of the stock setup. The FA setup is better than the EJs, and we know those are good to ~1.2 Gs sustained.

The pictures and info we have so far all include just one bearing spinning. The scoring and heat discoloration are perfectly consistent with oil starvation on those individual parts. Nobody has shown any damage to any other bearing or part other than those that actually failed. May be just a case of "Oh there is your problem we can stop looking now" but more data is certainly needed.
If it is a single bearing spinning and the rest look perfect it is likely a single channel blocked.
If it is an overall drop in pressure/flow yes a single bearing my fail FIRST but there should be other visible signs of damage or impending failure in other areas.

At this point, these are all theories. And here's mine:
If the top of the filter box is clogged, the oil would have to make two 90 degree turn through the sides. Very unstreamlined. Now this could cause several extremely low pressure points between the holes, and possible cavitation as well. I don't know how this engine differs from an Ascent's engine in terms of oil passages, but you would think (and hope) that they designed is for much higher RPM than an Ascent. The old school Japanese reliability is a thing of the 90s/pre-2008. So I don't expect Subaru to make sacrifices in cost to make reliability the topmost priority.

Oil going around 90-degree corners happens all over the place in your engine. Even when you break the corners and open the passages, you only see marginal changes from that. At some level (like building a $30K engine) it makes sense to commit that kind of time/effort. I don't think it adds up to much here.

ALTHOUGH, I'll cut a pickup apart soon and measure the mesh's total open area. Blocking off say 25% of that COULD have a pretty profound effect on oil flow characteristics. A combination of the two only compounds things.