Dwayne

I've been chasing down a slight idle surge and acceleration surge on the '84 for some time now. When I disconnect the O2 sensor or trip the WOT switch, it runs great - no surge. So I've been concentrating on vacuum leaks and checked everything out - only one leak remaining at the throttle body butterfly valve where the butterfly pin goes through the throttle body to the throttle linkage outside the throttle body. I couldn't find any source for spare parts (seal) so I put it back in. However, I was able to get it to seal under vacuum when I pushed in on the pin/linkage. When I pulled on the linkage or rotated it, it lost the seal and I lost vacuum. To make a long story short, I put the trottle body back in and ordered a spare used one from Jim to conduct experiments on.
In the meantime, I'm looking elsewhere for the contributor to the surging and decided to take a closer look at my exhaust leak. The leak is coming from the "Y" pipe. It's very noticeable when the car is cold but quiets down after getting up to temperature. So today I decide to jack the car up on stands and run it cold and get under and listen for the source of the leak. I find that it appears to be coming from the test port directly under the O2 sensor. I could feel exhaust escaping from around the port. The sound was concentrated in this area. I listened at the manifold to "Y" pipe connections and could detect no leak. Also listened at the CAT to rear exhaust pipe connection and could detect no leak.
So here's my question: If there is a leak at the test port on the other side of the O2 sensor, could oxygen be leaking in small enough amounts to affect the O2 sensor??

I went ahead and removed the "Y" pipe and CAT and decided to try to take off the test port bolt to see if it could be sealed better. After spraying it with WD-40 and attempting to un-screw it, the head broke off. Here's a pic:


Here's the location of the port in relation to the O2 sensor - both on opposite sides of the front of the CAT:


Looking at the inside of the test port from the O2 sensor opening:


One other thing I noticed is that the two arms of the "Y" pipe are movable. In other words, they rotate and move in and out. This works great for installation because they give enough to bolt right up to the exhaust manifold - no problem. I'm assuming they are designed this way?? I'm wondering if this "flexible" connection can also be a source of exhaust leak?? I could only get it to pull out about 1/2" and ran into resistance so didn't force it. Here's a picture of the movable arm - the green arrow shows the fixed part of the "Y" arm and the red arrow shows where it rotates about the rest of the CAT:


There is also an air injection tube coming from the air pump that injects cold air into the CAT - I don't think this is causing any problems with the O2 sensor.


If this leak at the test port is likely causing the O2 sensor to cause surging, what's the best way to seal it and any other leaks in the "Y" pipe and CAT? Seal it and pressurize it and look for escaping air and weld the leaks?? Buy an aftermarket "Y" pipe without the movable parts?? THANKS for reading - opinions and recommendations welcome!

blown 87

Yes, a leak can affect the O2 sensor, might not be your problem, but it can affect them.

Greg Nettles

blown 87

Also, I do not think that the Y should move.

I bet that is where your leak is coming from, and it is just showing up from around the heat guard at the test port.

If you can hear it it is a pretty good leak and I do not see the typical black smudging that you normally see at a leak at your test port.

Greg

Dwayne

THANKS, Greg.

OK...I think I see what you're saying. The leak may be coming from the "sliding" parts of the "Y" arms and travelling just under the outer metal skin (heat guard?) and escaping at the test port (where there are only 3 spot welds on the outer skin. The test port itself may not be leaking at all. If this is the case, I probably don't need to worry about drilling/fixing the broken bolt in the test port.

Any ideas on whether these "Y" arms can be fixed. Seems like one would have to remove (and probably destroy) the outer skin/shield of the arms to get to the pipe - then maybe weld these in place while the system is bolted up?? Or is it better to just buy a new replacement, a used replacement, or an aftermarket solid "Y" with no CAT?

blown 87

I will check mine in the AM, my motor is out so it will not be a problem to see if my "Y" moves, but I am almost positive that it does not.

As far as the heat shied, yes you can take them off and weld them back on, after you fix the pipes, or leave them off for that matter.

A lot of time they will get loose and rattle.

Greg

Dwayne

OK Greg, I'll stand by. THANKS!

shmark

The Y pipe definitely moves and I think it's by design. I just had mine out and back in and if it didn't move it would be really tough to refit. I had a test pipe on my old 928 for a while (its Y pipe also moved) and it was a beeyatch to fit - had it on and off several times chasing leaks, emissions tests and other stuff.

blown 87

Mine does not move at all, if you are talking about the y moving independently of the convertor, then there is a problem.

shmark

I hope not. My experience is limited to two different '83 928s, but they were both identical and moved exactly like Dwayne is showing. Not easily I'll grant, and I actually forgot about it when I started reinstalling this one last week. With one branch loosely bolted to the manifold, the other was nowhere near fitting and I started cussing...then I realized they moved and it was relatively easy to fit. FWIW part of the problem I was fixing was an exhaust leak and I can confirm zero leaks now, having been over every inch of the exhaust, including those sliding joints. I wonder if yours do move, but are stuck from corrosion over the years? Or both of my 83s and Dwayne's 84 have somehow broken a small weld somewhere? I don't see any evidence of that, but anything is possible.

blown 87

That could be the case on mine, but if Dwayne is feeling and hearing a leak then there is a problem.

We have taken many heat shields off (not on 928's) and I have never seen one move. but that does not mean that it should not move.

If three of them move and mine does not move, then I would guess mine is the one with the problem.

Oh well.

a4sfed928

[QUOTE
If three of them move and mine does not move, then I would guess mine is the one with the problem.

[/QUOTE]

These type of odds don't apply to 928s

Jim R.

I don't know about the "y" on the older cars, but the pipes on the '87 and up are solid with no movement..........or at least the ones sitting in my basement. If they move freely, definitely a source for a leak.

Jim

Mrmerlin

one way to tell use some duct tape to seal off the ends of the Y pipe and fill the Y with water if water leaks out then the Y pipe has cracked, i dont think there should be any movment in the Y. the heat shields may have a heat limiting cloth or other type of substance sandwiched between the pipe and the heat shield, It was very common for the exahust manifolds to crack on the 951, and it was pretty easy to tell how bad it was either you heard an exhaust leak or the whole manifold end would move around in the heatshield

blown 87

These type of odds don't apply to 928s [/QUOTE]

Good point.

Imo000

The part that you guys refert to as "moving" is only a heat shield and not part of the actual piping!
The part that moves has fibreglass insulation under it. If you actually remove it, you’ll see the real exhaust pipe. Also I don’t think the broken port plug (the one across from the O2) is leaking. However, as suggested, fill it up with water and see what happens.