'96 850R Very low mpg, very low power.

[bmessina]

Ugh... this thing.

I was testing the harness for the ECT, and noticed oil... on top of the coolant reservoir. Open it up, sure enough... oil in the coolant. Checked the oil via the dipstick and it looked clean. I'm not getting any smoke out the back.

I'm operating under the assumption that the issue is the oil cooler in the radiator, and sourcing a radiator. I was going to do a flush of the system when I put in the new heater core anyway.

I know the mechanic I took it do put some additive into the coolant because he thought it had a coolant leak. Bar's Leak. It certainly didn't have this issue before I brought it to him.

[mercuric]

Personally I'd want to verify the oil cooler was actually leaking by pressure testing it, both to possibly not spend $$ on something I didn't need and to be sure the right problem is getting solved. You're going to have to pull it either way, so..

I wasn't aware the radiator had an engine oil cooler too. Yikes. I'm used to those being separate like on the old bricks.

Good luck, I hope no further damage was done.

[erikv11]

As an aside, I would curse that mechanic for putting in ANY kind of stop leak product. Hopefully just the radiator.

[bmessina]

Personally I'd want to verify the oil cooler was actually leaking by pressure testing it, both to possibly not spend $$ on something I didn't need and to be sure the right problem is getting solved. You're going to have to pull it either way, so..

I wasn't aware the radiator had an engine oil cooler too. Yikes. I'm used to those being separate like on the old bricks.

Good luck, I hope no further damage was done.

You're right... I guess I'll just grab a radiator cap from the junkyard and rig it to the air compressor and see if the system holds pressure.

Also test the compression again, just to be sure on that front.

As an aside, I would curse that mechanic for putting in ANY kind of stop leak product. Hopefully just the radiator.

Seriously. That was such a waste of money.

[precopster]

When I was a wee lad I recall using Bar's Leak to help with sealing radiator holes. Of course it didn't work. What I do recall is that it was an oily slimy substance that takes time to bond with surfaces.

Unless you're trying to seal a poor head gasket I'd get that stuff right out of there with a cooling system flush then remove the ECT and be sure it's clean. We're no longer in the 80s now. That stuff has no place in a modern engine.

[bmessina]

How oily? Easy to mistake for engine oil oily? I wonder if that's what I'm seeing... That would be a relief.

[mercuric]

It's still very striking to me that the car was magically better after being jump-started, then got back to "normal" over the course of a few minutes.

This seems overly simple, but have you checked the system voltages? As in battery voltage with the engine stopped vs running?

Were you ever able to reproduce the "normal" state by jumping it again? I see you tried using an alternate ground, but that didn't help. Did you ever try the exact procedure that worked, to see if it repeated? Repeatability is import in testing, try something enough and the chances of it working once for unrelated reasons increase...

[bmessina]

It's still very striking to me that the car was magically better after being jump-started, then got back to "normal" over the course of a few minutes.

This seems overly simple, but have you checked the system voltages? As in battery voltage with the engine stopped vs running?

I'm taking all comers at this point.

I've checked battery voltage once and it seemed low, 11v w/ the engine stopped. Going off memory here, but I believe running was 13.3. I'll double check tonight. I've checked voltages at a few sensors in testing I did a few pages ago (hah), sensors have been testing within specific ranges (compared to the manual).

Were you ever able to reproduce the "normal" state by jumping it again? I see you tried using an alternate ground, but that didn't help. Did you ever try the exact procedure that worked, to see if it repeated? Repeatability is import in testing, try something enough and the chances of it working once for unrelated reasons increase...

I tried jumping it again, without the battery being dead, and could not reproduce the result using the same ground point on the "dead" care. I'm not super excited to purposefully kill the battery to get a more accurate test, but if it comes to that...

[bmessina]

Make that 12.4v with the car off, 13.1v with the car on.

[mercuric]

Sounds like system voltage isn't a problem, then... Figured I'd ask since that's the most logical thing to check when a jump makes a difference. But as I figured, something more illusive!