GR86 RTV in oil pickup causes engine failure.

Keith Tanner said:

In reply to z31maniac :

I understand that. Makes no difference how it's constructed. The oil pickup is partially obstructed. This will show up when there is maximum demand for flow. The construction of the pickup filter isn't really a factor, the important thing is that it's clogged to some extent. Subaru felt it needed X square inches of surface area to meet requirements, it does not have that anymore.

 

Oh I agree, I just shared that since you said you weren't familiar with it's construction. 

In reply to Slippery :

Even so, it's a known risk. But it's ironic that too much RTV is being applied by a computer, since they were introduced to specifically avoid these kinds of problems. 

It's likely that robots are applying the RTV. I'm wondering how the programming error happened, or if a different nozzle size or something was used.

In reply to alfadriver/Brett_Murphy :

Dont discount a problem with the RTV itself as Keith mentioned. If it was applied correctly, but it did not cure properly or it was not flowing properly due to its condition, then that could be causing these problems. 

In reply to Slippery :

In theory, a cnc machine should be capable of detecting material problems like that. Otherwise, it's not doing a big part of its intention. 

I have seen plenty of overly globbed on form in place gasket material and it has always just sort of stayed there.

i wonder too if it is a formulation error.  The OE level materials should be tenacious enough that you could lift the engine by the oil pan with no bolts in it.

alfadriver said:

In reply to Slippery :

In theory, a cnc machine should be capable of detecting material problems like that. Otherwise, it's not doing a big part of its intention. 

It's pretty clear that something isn't going as intended :)

I was really interested in these for a while. I took a test drive and had contacted every dealer in a 100 mile radius. I came *this* close to buying one, but a few posts about the reliability of the earlier motors had me concerned. I'm really glad I decided to wait. 

Pete. (l33t FS) said:

I have seen plenty of overly globbed on form in place gasket material and it has always just sort of stayed there.

 

i wonder too if it is a formulation error.  The OE level materials should be tenacious enough that you could lift the engine by the oil pan with no bolts in it.

A friend and I were just brainstorming about this. We were wondering if it was something like they're firing up the cars at the end of the assembly line before the sealant is fully cured. Because the pieces that didn't stay stuck to the surface they were originally stuck to seem to do a pretty good job of sticking to the pickup screen. We were also trying to figure out if a more traditional pickup design would have been better or worse for this.

In reply to dps214 :

The stuff I am familiar with is ready to use in 15 minutes, fully cured in a little longer.  Just-in-time manufacturing is efficient at keeping inventory low but I do not think it is that efficient 

I think Toyota's official stuff says do not run engine for a half hour.  Which is an amusing thing to read in the procedure for putting the timing cover on a 2GR-FE from an Avalon, which is an engine out job, and about four hours away from having an engine ready to go back in.

The oil pickup (specifically its restrictiveness at high RPMs) was one of the main weak points of the FA20, would be ironic if the new design contributed to this problem...

alfadriver said:

In reply to Slippery :

In theory, a cnc machine should be capable of detecting material problems like that. Otherwise, it's not doing a big part of its intention. 

Everything is perfect in a theoretic world ... but something failed. 

I know the CNC machines that I have experience with have the capability for the operator to set an amperage draw upper limit on the spindle for every tool. As the tool wears, the draw is rises, letting you know its time for a tool change or the machine can do it for you. 

With that being said, the upper alarm limit is determined by someone, a person usually. Who knows, maybe their math/testing was not correct. 
 

In reply to Matthew Kennedy :

As a cheap sports car enjoyer, I am sincerely empathetic about the situation and wish for the 86 twins to continue to succeed. But also as a Miata owner...whew!

As a person who would like to buy more cheap first generation Frisbees, I am disappointed because I was hoping the new ones would be perfect so people would give the old ones away for nothing because "those had the bad engine."

I wonder if it would be possible to flush the entire oil system, preferably flowing backwards, with some sort of solvent that can at least loosen any sealant that might be lodged in the wrong place.

In reply to ¯\_(ツ)_/¯ :

I'd think you'd have to run the oil pump backwards to get anything to flow the wrong way through the pickup tube, wouldn't you?  I've only taken a couple oil pumps apart, but none of them looked like much would pass through them if the gears weren't actively turning.

And I'd be concerned that whatever would eat through the misplaced sealant would also eat up the placed sealant.

In reply to wae :

Spin the engine backwards and just let it all flow out the drain plug?  I don't know, I'm just guessing at potential half measures when really it seems like pulling the pickup tube and cleaning thoroughly is probably the answer.

In reply to tremm :

That does look like a first gen pan.

I did this on my first gen- the original 2013 engine in it had died due to RTV ingestion by the look of the screen, and the 2016 engine I put in it I checked the screen (pretty much clear) and installed a Killer B oil pickup.  It's not dead yet after around 3-4k road miles and approaching 10 stage rallies, but I'm planning to check it again this winter.

It's possible they used the wrong formulation - Toyota got in trouble with the 2GR-FE 3.5L V6 when it was first introduced. The oil line in Toyota applications for the VVT was half metal, half rubber. Turns out the rubber section wasn't rated for engine oil use. They would develop pinholes over time, and eventually split. They recalled everything and the replacement part was again half metal, half rubber. The Lexus applications had a full metal oil line, so that's the one I put in our Rav4.

Keith Tanner said:

It's pretty clear that something isn't going as intended :)

Everything, everywhere. 

I took the car to the dealership to have the pan pulled.  2,400 miles, oil was changed by dealership at 1,000 miles.  Looks like about 15% to 20% coverage of upper screen.

Car runs perfect no issues whatsoever.  I had it pulled for piece of mind.

Waiting to see if Subaru is going to cover the cost of pulling and reinstalling the pan.

Glad I checked...

In reply to pimpm3 (Forum Supporter) :

Even if you have to pay, which you shouldn't, it was worth it. 

Are you saying 20% coverage is good or bad? Are you trying to get subaru to cover pulling the whole pan to clean the pickup, or just the inspection?

pimpm3 (Forum Supporter) said:

I took the car to the dealership to have the pan pulled.  2,400 miles, oil was changed by dealership at 1,000 miles.  Looks like about 15% to 20% coverage of upper screen.

Waiting to see if Subaru is going to cover the cost of pulling and reinstalling the pan.

Glad I checked...

Build date?