Smokes Liek a steam Engine

[GSB]

Running very rich then. See above, re: coolant temp. Missing spark will post a misfire code to the ECU.

Although, if oil levels are down too, that could be a slightly more... ...terminal... kind of problem.

[shnazzle]

Running very rich then. See above, re: coolant temp. Missing spark will post a misfire code to the ECU.

Although, if oil levels are down too, that could be a slightly more... ...terminal... kind of problem.

See what I'm thinking about the coolant sensor is:
A) Temp readings in dash seem to work fine (start at 0, go to 90)
B) I get lambda readings and corresponding ECU responses to short term fuel trim. So, that means the loop must be closed right? Except for when you let off throttle obviously.

So...doesn't that mean the temp sensor is fine?

I wish someone would just give this engine a final thumbs up or thumbs down. Don't even want to drive it now with fuel in the oil.

[Wabbitkilla]

It only runs closed loop off throttle.  s:wink: :wink: s:wink:  

Interesting problem, sounds for all the world like massively over fuelling, but when I followed you it didn't smell like it was massively over fuelling, believe me it would give me a headache if it was. Sounds more like ring failure to me.

[shnazzle]

Always get the open/closed mixed up....

That's a 700 quid MattPerformance job isn't it?  s:( s:(

[Wabbitkilla]

Not 100% convinced myself but it certainly needs in depth investigation of the engine and its bores.
Anything involving engine work is going to cost £100's, you have to remember our engines get a hard life in a sports car and will need more care than your average car. Mine has started to consume oil now so I'm planning a rebuild, it's done 113000 miles so I reckon I've done quite well really. I remember a day when if you got 40000 miles on Ford engines it was a miracle.

[shnazzle]

Not 100% convinced myself but it certainly needs in depth investigation of the engine and its bores.
Anything involving engine work is going to cost £100's, you have to remember our engines get a hard life in a sports car and will need more care than your average car. Mine has started to consume oil now so I'm planning a rebuild, it's done 113000 miles so I reckon I've done quite well really. I remember a day when if you got 40000 miles on Ford engines it was a miracle.

Makes sense then as mine's also at 113k. So it hasn't done badly for a seemingly poorly maintained 2

[Wabbitkilla]

Does sound similar to mine, I worked out that I go from top to bottom on the dipstick in just over 1000 miles.
Mine smells rich once it's stood a while and cooled off, I just figure it was running cold start cycle (for at least a short time) so that explained the rich smell when starting up. But then I am running an Apexi PFC ECU.

Remember your cam chain will have a certain amount of stretch now so combustion timing will have a habit of wandering, the vvti can cope with a bit of it ... but it has its limits. My running seems to get seriously upset by fuel quality so I aim to keep it consistent by sticking with a fuel i know gives me the least problems ... Shell V-Power Nitro+ at the moment. I was originally just going to service the timing end of the engine, but then the oil use started so now, with the help of a friend with a garage, I'm going to pull the engine out ... list of tasks;
Hone cylinders
New design pistons and rings
Replace valve stem oil seals
Replace valve springs
Check and re-shim as necessary
New chain
New tensioner
New slippers and guides
New oil pump
New water pump
New thermostat
New head bolts
New accessory belt
New oil pan (it's got battered and rusty over the years)
Possibly adding an oil cooler
Probably fitting oil pressure and temperature gauges.

We have to budget for this kind of thing as our cars get older and the mileage piles on, people are now buying these cars really cheaply and expecting them to fresh and perfect. They aren't, they're seeing high mileage, they've been getting used enthusiastically, and they will cost money to keep them running.

It sounds like yours is worn like mine, if you choose to act soon then you can save your engine by having it refreshed and having your original engine kind of makes you feel better with a known history and there's a bit more trust there. Keep an eye on your oil and try and measure how much it uses over distance, it just sounds like your car needs similar treatment to what I'm planning for mine. Of course once I get my engine opened up then I might find something more worrying ... but the only way to find out is to take the plunge and do it.

[shnazzle]

I factored in a bit more service than usual. I didn't factor in an engine rebuild unfortunately. 113k isn't that much in a "normal" car but I guess on these it translates to about 200k.

Looking like a SORN until I can get some cash together to take it to matt for the full works.

Any immediate danger in running it with the oil smelling as strongly of petrol? Or might as well SORN it today?

[Wabbitkilla]

Any immediate danger in running it with the oil smelling as strongly of petrol? Or might as well SORN it today?

I would be tempted to do a oil and filter change using a flush, then see what it's like afterwards.
Te be honest the only time I've come across oil smelling of petrol was in a 2-stroke.
Or was that petrol smelling of oil?

Anyway, it's weird and worth trying a flush.

[GSB]

Running very rich then. See above, re: coolant temp. Missing spark will post a misfire code to the ECU.

Although, if oil levels are down too, that could be a slightly more... ...terminal... kind of problem.

See what I'm thinking about the coolant sensor is:
A) Temp readings in dash seem to work fine (start at 0, go to 90)
B) I get lambda readings and corresponding ECU responses to short term fuel trim. So, that means the loop must be closed right? Except for when you let off throttle obviously.

So...doesn't that mean the temp sensor is fine?

Certainly the sensor that runs the gauge is fine, but that may not be the sensor that feeds the ECU. It's not uncommon to have 3, one for gauge, one for fans, and one for ECU.. (Dont know if Toyota do it that way, but Peugeot certainly do)

As I said, its been a long time since I've delved into the intricacies of engine management, but fuel in oil sounds like over fuelling to me, and overfuelling a modern computer controlled engine without it objecting and thowing all sorts of ODB fault codes is remarkably hard to do. As you've seen, the engine will trim its fuelling within a set band to acquire the right fuel air ratio, and show fault codes if cant do it. For it to add to much fuel is almost impossible, unless its running in a state where excess fuel is actually required, I.e. stone cold. Fool an engine into thinking its cold, and it'll input more fuel.  That's why I keep coming back to the coolant temperature issue, because that's the single point of common failure most likely to produce the symptoms you are experiencing.

There is a separate issue at work her that is worrying though. Fuel in oil. Barring a head gasket and head casting failure, there's only one way for the fuel to get in the oil, and that is past the piston rings. Running lots of fuel into a cylinder is crap for efficiency and power, but it also increases the propensity for 'bore washing'. Assuming you can use it with turning yourself into a walking inferno, petrol is a fantastic de-greaser. Nothing removes oil quite like it, so if you get to much into your cylinders you end up washing the lubricating film of oil off of the cylinder walls. Cue highly accelerated wear, piston ring blow by, and eventual failure.

My first step now would be to do a compression test.  

Disable the fuel injection, take out all the spark plugs, and do a dry and wet compression test. See how long it takes each cylinder to come to pressure, and what the variance between cylinders look like. Do it dry first, and then wet, with a teaspoon full of oil poured into the spark plug hole. Any huge differences between wet and dry will indicate bore/ring issues.

Then, if that's all ok, fix your rich running issue before you really toast the engine.

[shnazzle]

Running very rich then. See above, re: coolant temp. Missing spark will post a misfire code to the ECU.

Although, if oil levels are down too, that could be a slightly more... ...terminal... kind of problem.

See what I'm thinking about the coolant sensor is:
A) Temp readings in dash seem to work fine (start at 0, go to 90)
B) I get lambda readings and corresponding ECU responses to short term fuel trim. So, that means the loop must be closed right? Except for when you let off throttle obviously.

So...doesn't that mean the temp sensor is fine?

Certainly the sensor that runs the gauge is fine, but that may not be the sensor that feeds the ECU. It's not uncommon to have 3, one for gauge, one for fans, and one for ECU.. (Dont know if Toyota do it that way, but Peugeot certainly do)

As I said, its been a long time since I've delved into the intricacies of engine management, but fuel in oil sounds like over fuelling to me, and overfuelling a modern computer controlled engine without it objecting and thowing all sorts of ODB fault codes is remarkably hard to do. As you've seen, the engine will trim its fuelling within a set band to acquire the right fuel air ratio, and show fault codes if cant do it. For it to add to much fuel is almost impossible, unless its running in a state where excess fuel is actually required, I.e. stone cold. Fool an engine into thinking its cold, and it'll input more fuel.  That's why I keep coming back to the coolant temperature issue, because that's the single point of common failure most likely to produce the symptoms you are experiencing.

There is a separate issue at work her that is worrying though. Fuel in oil. Barring a head gasket and head casting failure, there's only one way for the fuel to get in the oil, and that is past the piston rings. Running lots of fuel into a cylinder is crap for efficiency and power, but it also increases the propensity for 'bore washing'. Assuming you can use it with turning yourself into a walking inferno, petrol is a fantastic de-greaser. Nothing removes oil quite like it, so if you get to much into your cylinders you end up washing the lubricating film of oil off of the cylinder walls. Cue highly accelerated wear, piston ring blow by, and eventual failure.

My first step now would be to do a compression test.  

Disable the fuel injection, take out all the spark plugs, and do a dry and wet compression test. See how long it takes each cylinder to come to pressure, and what the variance between cylinders look like. Do it dry first, and then wet, with a teaspoon full of oil poured into the spark plug hole. Any huge differences between wet and dry will indicate bore/ring issues.

Then, if that's all ok, fix your rich running issue before you really toast the engine.

Now that I can work with. I'll fetch a compression gauge. Really appreciate the time guys. Great stuff.

How does one disable fuel injection in the MR2?
Done it once before by pulling a fuse out, is it as simple on the 2?

I'll see what I can do to determine whether the coolant temp sensor is gone or faulty.
But I agree, only having a P0135 for the level of over fuelling I have seems odd.

Next is a flush/refill of oil. However...with such an old engine, do I not run the risk that the flush will dislodge all the nice gunk that's helping seal my engine?

[Wabbitkilla]

I wonder is bore wash would result in petrol smell from the oil?
I still say you would pop up an eml alerting it was too rich.

[shnazzle]

I wonder is bore wash would result in petrol smell from the oil?
I still say you would pop up an eml alerting it was too rich.

In any case I want to drain that crap out to prevent further washing.

Agreed on the fault code though. You'd think after having long term fuel trims of max (25) for ages, the car would get the hint that owt is amiss and throw an error

[mrzwei]

Just pulling the plugs after a run will give you some useful pointers (you will have to do that for the compression test anyway).

 m http://www.triumphspitfire.nl/plugs.html m  may be of use. Certainly black, sooty and wet electrodes won't bode well.
I've had oiled plugs on an 'A' series and on an old Rootes motor, starting and tickover were a problem until it burned off. Modern engines run a bit leaner so the 'Normal' picture may look a bit lighter.

If they are contaminated then try just cleaning them (wire brush / electric drill) and see if it runs better for a bit, but it will come back.

I agree with the others that running that rich would cause a CEL because that was the main point of OBD systems.

[shnazzle]

Methinks I shall just get the 50 quid service special from CTP and just do the lot while I'm at it. And get a new O2 sensor to have a EML-less MR2 for the first time since ownership.

I'll post pics of the sparks up here. That site will help heaps. Thanks!

[GSB]

I wonder is bore wash would result in petrol smell from the oil?
I still say you would pop up an eml alerting it was too rich.

In any case I want to drain that crap out to prevent further washing.

Agreed on the fault code though. You'd think after having long term fuel trims of max (25) for ages, the car would get the hint that owt is amiss and throw an error

You know, up until about 2 weeks ago, I'd have agreed with you. In fact I'd have put money on it, but recent experience has proven a few oddities to me.

The Peugeot I keep mentioning has left me stuck at the side of the road twice in the last couple of weeks. It's been an absolute sod to find the problem. It starts fine, idles like a bit of a dog, but no worse than its ever done, and no worse than you expect for a 100,000 mile old engine.  After a few minutes though, a random lottery of symptoms can make themselves very very apparent. Take your choice from any or all of the following; Idle control goes out of the window,
it runs like a total dog,
it won't rev,
it develops almost, but not quite exactly, NO power whatsoever
and it happily leaves you stranded in the middle of a road whilst everyone swears at you and calls you a  swanker wanker swanker .

At this point, it will not start. It cranks over, and almost chokes you with the smell of unburnt fuel, but wont fire. After 2 minutes or so, it will begrudgingly come back to life, clear its throat (plugs fouled), and carry on as though nothing has happened, other than the fact that it drinks fuel like George Best drank scotch.

Plugs are good, fuel is good, ht leads are good, coil is good, position sensor is good, temp gauge works just fine. Much digging and testing revealed that the problem occurs at the exact same time the thermostat opens to admit water to the radiator. (Pure luck, happened to holding onto the top hose when it suddenly went warm as the stat opened, and the engine immediately started misbehaving). The ridiculous smell of excess fuel led me down the path of dodgy mixture control, idle air control valves, and many other dead ends that had me pulling my hair out.

Then I discovered the seperate coolant temp sensor feeding only the ECU. There are 3 of them, one in the radiator water inlet controls the cooling fan, one in the side of the cylinder head feeds the gauge on the dash, and another, right next to it and mounted just downstream of the 'stat, feeds the ECU. According to the bible (Haynes) a defective ECU coolant temp probe WILL trigger a CEL. So, naturally, I took at at its word and carried on looking. It was only when I disconnected the sensor that I discovered this to be complete balls.

The sensor has been diconnected since Thursday and the CEL steadfastly refuses to light. Cold starts are obviously a bit crap, but the engine no longer cuts out, and once warm is behaving itself quite nicely. It has also stopped gassing me out with petrol vapour, its using less fuel and making more power too. I suspect that the new sensor I've ordered for it will restore the cold start performance and allow the full range of timing and fuelling control, restoring the full quota of 90bhp (Woohoo!).

That's the story with my little French shopping trolley, I suspect that its worth investigating on your car, just in case.

[shnazzle]

I wonder is bore wash would result in petrol smell from the oil?
I still say you would pop up an eml alerting it was too rich.

In any case I want to drain that crap out to prevent further washing.

Agreed on the fault code though. You'd think after having long term fuel trims of max (25) for ages, the car would get the hint that owt is amiss and throw an error

You know, up until about 2 weeks ago, I'd have agreed with you. In fact I'd have put money on it, but recent experience has proven a few oddities to me.

The Peugeot I keep mentioning has left me stuck at the side of the road twice in the last couple of weeks. It's been an absolute sod to find the problem. It starts fine, idles like a bit of a dog, but no worse than its ever done, and no worse than you expect for a 100,000 mile old engine.  After a few minutes though, a random lottery of symptoms can make themselves very very apparent. Take your choice from any or all of the following; Idle control goes out of the window,
it runs like a total dog,
it won't rev,
it develops almost, but not quite exactly, NO power whatsoever
and it happily leaves you stranded in the middle of a road whilst everyone swears at you and calls you a *******.

At this point, it will not start. It cranks over, and almost chokes you with the smell of unburnt fuel, but wont fire. After 2 minutes or so, it will begrudgingly come back to life, clear its throat (plugs fouled), and carry on as though nothing has happened, other than the fact that it drinks fuel like George Best drank scotch.

Plugs are good, fuel is good, ht leads are good, coil is good, position sensor is good, temp gauge works just fine. Much digging and testing revealed that the problem occurs at the exact same time the thermostat opens to admit water to the radiator. (Pure luck, happened to holding onto the top hose when it suddenly went warm as the stat opened, and the engine immediately started misbehaving). The ridiculous smell of excess fuel led me down the path of dodgy mixture control, idle air control valves, and many other dead ends that had me pulling my hair out.

Then I discovered the seperate coolant temp sensor feeding only the ECU. There are 3 of them, one in the radiator water inlet controls the cooling fan, one in the side of the cylinder head feeds the gauge on the dash, and another, right next to it and mounted just downstream of the 'stat, feeds the ECU. According to the bible (Haynes) a defective ECU coolant temp probe WILL trigger a CEL. So, naturally, I took at at its word and carried on looking. It was only when I disconnected the sensor that I discovered this to be complete balls.

The sensor has been diconnected since Thursday and the CEL steadfastly refuses to light. Cold starts are obviously a bit crap, but the engine no longer cuts out, and once warm is behaving itself quite nicely. It has also stopped gassing me out with petrol vapour, its using less fuel and making more power too. I suspect that the new sensor I've ordered for it will restore the cold start performance and allow the full range of timing and fuelling control, restoring the full quota of 90bhp (Woohoo!).

That's the story with my little French shopping trolley, I suspect that its worth investigating on your car, just in case.

Hmm...I could have written that exact story. I see why you think I may have the same problem.
As far as I can tell (after 1.5 minutes of "research") there is but one coolant temp sensor on the MR2. Tempted to unplug it and see what happens. However, how does this ECU interpret np signal from the sensor? Does it go to a base of some sort, or does it always report 0?

I'll have a peek to see how many sensors I'm dealing with. in the meantime I've ordered a full service kit and an O2 sensor, and will get a compression tester tomorrow.

[Wabbitkilla]

It does sound plausible, but with the 2 having only one sensor then I doubt it.
If you have the right obdii reader it will tell you the actually temperature the sensor is seeing.
That's one of the handy things with the PFC, you can actually read off the temperature the ECU is seeing.

[shnazzle]

Just had another peek at a semi-recent log I took, in ScanXL. Off the back of this a question; is it possible for an o2 sensor to not report anything back but not throw up an error code?

I ask this because I'm seeing SAE.O2S11.O2SV (Bank 1 Sens 1) reporting voltages nicely (although erratically, not quite the nice sine-wave I want), but SAEO2S21.O2SV (Bank 2 Sens 1) is returning 0.000v. But the only error code I have is P0135...which relates to the working sensor.

Does it even need both sensors working to get the fuelling right?

[Wabbitkilla]

Odd ... it should report a code, even if they get a bit tired they usually kick off a code.
Fuelling will not be right without any reading from the sensor and this could explain your running rich.
I suggest you swap them over (obviously maintaining the wiring correct for left & right) and see if the fault follows the sensor.
If it does then replace the sensor that's reading 0v, if it doesn't then you could have a faulty wire.

[shnazzle]

Swap sounds like a good idea eh? My car seems to be skimping on codes doesn't it?  s:) s:)  You'd think it pays toll for every code.

I identified this a few weeks ago with the multimeter as well. I just figured I was doing something wrong and ignored it, as it didn't throw an error.
The bank2 (left) sensor was returning values for the heater element circuit, but I couldn't get it to show anything on the actual o2 circuit.
Bank 1 (right) was obviously dead on the heater circuit...hence P0135. Confirmed  s:) s:)  hehe.

[Wabbitkilla]

So both sensors are faulty ... just different faults!  s:oops: :oops: s:oops:

[shnazzle]

Good thing I ordered that servicing set.


For removing the sparks, do you just use one of those 2-sides bars to get into the holes?
Never seen them so far down (shows how long it's been since I've done this)

Had my compression kit at the ready and then saw the deep abyss under each coil pack

[shnazzle]

Sparks look good to me. Good sign. The first one is a bit more corroded.
Haven't taken cyl 3 out yet as the bars were in the way and couldn't be bothered just yet.


Also, no compression test yet as I lack a 19mm socket to turn the engine after replacing the timing chain tensioner.

[Wabbitkilla]

Jeez mate, those are worn out!
Replace them and the car will run far better  s;-) ;-) s;-)