'99 2.8 Roadster not running right (details inside)

[JKuhn]

As many people here will undoubtedly know by now, I'm struggling to get my Z3 to run right, and considering how many things I've tried so far, decided to rather create a proper thread to figure out what needs to be done.

Symptoms:
Poor low-end performance when fully warmed up - It runs better until it hits the regulated operating temp (I can actually feel it entering closed loop)
The power comes in very noticeable "steps", with the last being a massive bump around 4k RPM - Not as much of a problem right now
Very slow to rev, this is both warm and cold
Quite a bit of rev-hang
Lots of gurgling while coasting in gear (keep in mind this is a decatted exhaust with a straight-through box, it does still have the normal resonator)
No power at WOT
Misfire around 3.7k RPM at wide throttle
P0340 intake cam code
Strange exhaust smell - I'm not experienced enough to say if it's rich or lean
(More to be added as I think of the problems)

What's normal:
Top-end power feels decent, unless going WOT
No heat issues - I have driven 3 hours+ without any issues
No trouble starting
Smooth idle

Already done/checked:
VANOS
Intake leaks
Two cooling system pressure tests have been done, one by a radiator shop. The radiator shop also checked the hoses for abnormal pressure, and smelled the steam for signs of gasses when they opened the expension tank. It might not have been a chemical test, but it was free. So no signs of combustion gasses in the coolant, and no signs of coolant in the cylinders.

Current list of suspects:
Bad cam sensor - To be replaced later
Bad MAF - Replaced with a knockoff, to be redone later if needed
Bad downstream O2 sensors (no signal)
Unknown condition front O2 sensors
Bad fuel pump???
Bad EVAP purge valve - The tank side hose fitting is broken off - stealership only, temporarily disconnected the power

For those who don't know those threads yet, more details on what's been done and previous discussions can be found here:
https://www.bimmerforums.com/forum/s...-leak-symptoms
https://www.bimmerforums.com/forum/s...-anything-else

[deni2s]

Sounds like a small crack in a head to me. Had similar symptoms until it developed bigger.

[BimmerBreaker]

Coolant consumption but no visible traces in oil. At that consumption rate I would also think head gasket

[JKuhn]

It took a while to remember again, but there was someting I wanted to add (I wsn't at my computer at the time). After replacing the thermostat we also felt the hoses, and they didn't seem to become as hard as one would expect from a compression leak. I'll check again to be sure, but as far as I know this is still the case. I'll also add it to the OP, for future reference.

Anyway, I'll have another test done after the weekend. I'm told that the system should be pressurised for half an hour, an then if the engine misfires directly after that it means water got into a cylinder.

[JKuhn]

After speaking to some club members, I'm suspecting that it's just an external coolant leak. I'll look into it when the car is cooled down.

They said (and it makes sense to me) that a blown headgasket should cause misfiring on startup, due to pressurised water pushing into the standing cylinder(s).

[Mr Bingley]

Analyze your coolant for combustion gasses if you still have any doubt. The tell is that it consumes coolant--if it's not leaking, it's a blown head, it's as simple as that--there's nowhere else for the coolant to go. The blown head may or may not be the source of all of your symptoms, that's a different question. For sh**s and giggles, check your carpeting under the dash to make sure its not leaking out at the heater core. As I suggested in an earlier thread, you only get lucky once.

[JKuhn]

I'll have the second test done. I do however suspect that it's not the case. I found a coolant "stain" in the engine bay that wasn't there when I did the thermostat, so I tightened the clamp a bit more. Hopefully the coolant escaped there, and only there. It is however odd, because the car passed a pressure test.

[deni2s]

Still sounds like a developing crack in a head (or blown gasket) to me. From my personal experience with cracked head I can tell that you had been told a lot of bullshit. Crack in a head might take a lot of time to develop so big that it will cause some cylinder to misfire or white smoke is noticeable coming out of the exhaust. Leaking cooling system might be one of the first consequences/symptoms of crack in the head. Crack changes it's size depending on temperature and it might suck or push the coolant depending on pressure.

Check your sparks - the color of them might give some hint.

[JKuhn]

By sparks, I assume you mean the plugs?

[magnetchief]

Has a similar situation with my 2.8 5 series.
Using a borescope we found coolant in all the cylinders.
Ended up removing the head. Every coolant port was cracked to the cylinder bores. Coolant was leaking out of the exhaust. When I could get the thing running.

Motto, don’t buy cars from Goodwill.

Swapped out he head and cooling components and flipped the car. Made a few hundred $$ but bad experience


Sent from my iPhone using Tapatalk

[deni2s]

By sparks, I assume you mean the plugs?

Yes, plugs, sorry, my mistake.

[JKuhn]

First off, just know that I do take what's said here into consideration. I'll see if the coolant level keeps dropping, and I'm also planning to take the car for some tests Monday morning.

But still, my gut tells me it must be something else. Today (after tightening the clamp, I still need to top it up again and see if the coolant level is still dropping) the car ran perfect again for a while (normally it runs better when not fully warmed up, but by no means perfect). It wasn't cold, but not fully warmed up either (slightly above the cold mark). Earlier today the car ran decent during a 40km trip, and on the way back it was worse, but not that bad. It's definitely better than when I bought the car though, considering that I can accelerate at highway speed in 5th gear without flooring, when earlier I had to floor it in 4th. The thing is, the car's performance is very unpredictable. It usually runs decent (meaning smooth, but fairly weak) cold and driveable when at normal temp (about 96-97C, according to the cluster test mode), but it also has short bursts when it will run like a dream. And those can happen at any time, for any length of time. Once it happened after stopping for some petrol, and another time it happened after I cleaned the intake (and then the thermostat broke).

Regarding how the car performed when I bought it, I wasn't able to do a full test including highway. I just did some normal streets, plus a major road (80km/h limit). The car also showed signs of having stood some time, so I wasn't expecting perfect performance, plus I wasn't comforable pushing such a car too much since I didn't know it.

[deni2s]

The thing is, the car's performance is very unpredictable.

As I said before - crack increases or decreases unpredictably because of temp/pressure changes. My gut tells engine was overheated and head warped or developed a crack regardless of how much you don't want to believe that the worst had happened. Make sure you fix it before you get it hydrolocked.

[JKuhn]

As I said, I'm going to find out if that's the case.

[LannVouivre]

Look inside the combustion chambers via the spark plug holes; sometimes a head gasket leak into the cylinders will steam clean the tops of the pistons.

Are you topping off with coolant? Coolant will leave whitish residue, water won't leave obvious traces.

Have you checked the pipes under the intake manifold for leaks? They're plastic. On my M52tu, they were completely rotten.

[Randy Forbes]

Just spitballin' here, but are you refilling the reservoir ONLY to the KALT (sp?) level, and it's dropping from there, or are you topping it off to the neck, and each time recheck it, it's lower than where you last left it?

I'm not ready to condemn anything yet...

[JKuhn]

Look inside the combustion chambers via the spark plug holes; sometimes a head gasket leak into the cylinders will steam clean the tops of the pistons.

Are you topping off with coolant? Coolant will leave whitish residue, water won't leave obvious traces.

Have you checked the pipes under the intake manifold for leaks? They're plastic. On my M52tu, they were completely rotten.

Currently straight water, so the coolant is getting diluted. But I'm planning to find the problem soon enough.

Just spitballin' here, but are you refilling the reservoir ONLY to the KALT (sp?) level, and it's dropping from there, or are you topping it off to the neck, and each time recheck it, it's lower than where you last left it?

I'm not ready to condemn anything yet...

I fill it until the float reaches the top of the filler, as I understand that's how it should be.


Anyway, I topped it up this morning, and drove a bit (to a local Z meet that I only heard of today). I'll use it a bit more today, and then I'll see what the level is tomorrow morning.

EDIT: By the way Randy. This is a bit off-topic, but what would you suggest I do about the rear-end, as I can't exactly import a large kit? I'm thinking of just getting it rewelded (it's not bad yet, there's a couple broken welds).

[BimmerBreaker]

The kit ships in a small box. He has shipped everywhere (except Antarctica... Same for me then... First to sell to the frozen tundra wins?) so no reason he cant ship to you

I wonder if the car is leaking coolant into combustion chambers, possibly after long drives it reaches a sort of equillibrium and runs ok but once you park it the coolant slowly continues to leak in for a bit, pooling in the cylinders, soaking some spark plugs once started, causing poor running until it burns out and plugs dry up. Checking plugs is a good idea. Have you done a leakdown test yet?

[JKuhn]

But then wouldn't it misfire on startup? It always starts clean, and with a perfect idle.

I drove it a bit more today. On the way out it performed like almost always. When I started the car again (probably about 2 minutes later) it ran well (meaning smooth and fairly powerful, but not as rev-happy as it should be) for a couple minutes. So there wasn't any significant time for it to cool down, and it didn't get better by warming up. Just like that time when I stopped for a refuel.

EDIT: Oh, and regarding the reinforcement kit, there's also import cost to consider. The Rand isn't exactly doing well, and I don't know if I'll be able to import something like that soon.

[deni2s]

But then wouldn't it misfire on startup? It always starts clean, and with a perfect idle.

Quoting myself in case you haven't read it the first time:

Crack in a head might take a lot of time to develop so big that it will cause some cylinder to misfire or white smoke is noticeable coming out of the exhaust.

Not a very good idea to drive around with cracked head. You can catch hydrolock at any moment. Will be much more expensive than just to replace the head.

[JKuhn]

I'm sure "coolant slowly continues to leak in for a bit, pooling in the cylinders, soaking some spark plugs once started" would cause a misfire.

And to be honest, the way you're saying it sounds like you want me to have the car taken away on a flatbed to have the head done without any further testing. In order to hydrolock, the cylinder must get a significant amount of water. And I don't think it's very likely the supposed crack (or blown gasket) will progress quickly enough for that to happen with no warning, unless something seriously goes wrong. Remember, the coolant will kill that cylinder long before it gets enough water to hydrolock, so will it really lock with no warning?

Anyway, I don't know how reliable this is to determine if there's an internal leak, but a local club member recently did the gaskets on his old Ford Granada, and he said that he long suspected a blown gasket on cyl 5, because the plugs in that cylinder kept dying, and also due to loss of coolant. He was also the first person to tell me that a coolant to cylinder leak should cause a misfire on startup, and then a couple other members also said the same.

EDIT: Just an update, it seems I won't be able to take the car in tomorrow, unless I'm willing to remain in that area for some time. I'll just leave the car at home, drive the Mazda (cheaper anyway, I just need a thermostat (it doesn't have one, and I decided to stop being lazy) and a new air filter), and hopefully take the car for the test Tuesday.

[JKuhn]

UPDATE: I checked the coolant this morning and while it might have lost a bit, it was pretty hard to tell. I'll see tomorrow morning when I (hopefully) take the car in to be sure, but it looks like I caught the culprit. It used to have a very obvious drop in the level after one day.

[deni2s]

And I don't think it's very likely the supposed crack (or blown gasket) will progress quickly enough for that to happen with no warning, unless something seriously goes wrong.

I wonder what might went seriously more wrong than a crack in a head

I'm talking from my personal experience with m52b28. Of course coolant leak will cause misfire at some point, but not in an early crack stage. In early crack stage it's just a loss of coolant and even that is not constant. And some parts of cooling system are failing one after each other. Not even lots of white smoke. No dying plugs. After a few weeks on trying to find out the cause, and hope that it's just cooling system not properly bleeded, engine started to misfire (especially on startup) and there was noticeable white smoke and liquid coolant out of the exhaust. I believe due to temp changes crack can develop to bigger one at one moment causing hydrolock (but haven't experienced that myself), or when engine cools down it could suck more in coolant than while driving.

I hope that's not the case with your engine. Though I still don't see how coolant leak might interfere with random loss of power. It's very interesting to follow how it will develop .

[JKuhn]

The thing is, it might just be two seperate problems, that happen to be at the same time.

But I suppose I'll find out tomorrow.

EDIT: Sure, a hydrolock is possible, but by my logic it's very unlikely that the supposed internal leak will instantly enlarge to the point where the engine can draw in that much water in a single intake stroke (during the compression stroke the pressure will be too high to draw more coolant in).

Oh, and by something seriously going wrong, I mean a massive overheat causing the leak to enlarge enough for a hydrolock.

[Mr Bingley]

Why guess or speculate? Link.