Unofficialbmw.com

[Paulchartley]

Without warning my 1990 (e32) 735il automatic has suddenly developed engine problems. This was not gradual it just happened out of the blue one morning.

It is difficult to start but once it is going it appears to idle fine however when anything more than gentle gradual pressure is applied to the accelerator the engine badly missfires.

Without load the engine will rapidly accelerate however it is realy rough but will hold high revs smoothly once it gets there. Try this under load and it missfires and all but cuts out. Hold the peddal down two long and the car backfires once you ease off again so suggesting that fuel is getting into the cylinders.

The Air Meter has been replaced with a known working unit. There is fuel pressure as the thing runs happy on idle and with a run up the car will reach high speed eventually.

Clamping the rubber fuel hose shut on the tank return line has no effect so ruling out a leaky fuel pressure regulator.

The fuel pump has been checked for damage and none is apparent.

the car starts better when the blue temperature sensor is unplugged (obviously the engine is getting more fuel as the motronic system thinks that that the engine is cold)

The car does not appear to have a CAT and there are no O2 sensors in the exhaust system.

The narror rubber vacuum hoses to the fuel regulator and the cam case to inlet boot connector have been replaced as a matter of course.

It feels as though when the engine is reved it is flooding, starving or the spark is not getting there at the right moment.

I have now run out of ideas

HELP!!!!!

[edjack]

Sounds like a sticking open idle control valve. Remove it, and clean it thoroughly with carb cleaner, until the innards are free to move when shaken.

This doesn't always work, so you may end up replacing the valve.

BTW, if there are no O2 sensors, the fuel injection system will run open loop, and run rich, since the mixture feedback is absent. I'm surprised that the Check Engine light is not on. Wonder if the previous owner took the bulb out?

Where do you live to be able to run without a cat? There is no advantage, since a cat offers little, if any back pressure, and it cleans up the exhaust.
_________________


Ed BMW CCA member since 1987 (Nr. 62319).
'97 540i 6 speed Aspensilber over Aubergine interior.

[Paulchartley]

Hi Edjack, thankyou for your prompt reply.

I am in the UK and not all cars in the UK have CATs. I think it is only recently that this is the case. I have checked al the way along the exhaust system, there are no sensors of any type and the whole mixture thing really does appear to be done from the intake side of the cylinders with no feed back loop or expensive sensors on the exhaust side. This is just as well as when this problem first happened a few days ago I suddenly ended up with quite a bit of unburnt fuel in the exhaust which exploded in a god awful back fire splitting the weld seams in in the final silencer box and leaving half of it in the road - strangely the absence of back pressure from the exhaust now that half of it was missing actually made the car run slightly better.

The ICV does buzz and it was my first thought to look at that first (anything that sits there making a buzzing sound when it is not pumping something is suspect in my book!). With the two relevent holes plugged I ran the engine so that I could look inside the ICV and the little shutter opens and closes freely. With the thing disconnected the mechanism moves freely.

Things have moved on in the last 24 hours and I recon I might be down to the last two or three culprits which is an improvement as before it could have been anything. (to be honest I think I have two things going on now and not one, at least one of these the result of taking most things appart to find the first problem)

I went right back to the beginning again. I reset the computer by shorting out the negative and positive cables for the battery for 10 minutes (following the ECU reset as detailed in the E32 section of www.E38,org at http://www.e38.org/e32/rough_idle_and_shimmy_ultimate_guide.htm ) and started testing from the top. By applying a flat faced g clamp (mole grips are not flat faced and so can slip on the rubber hose and let fuel through) I was able to completely restrict the flow of fuel back to the tank so any pressure in the fuel rail feeding the injectors was entirely as a result of the fuel pump.

The change at idle and when the accelerator cable is pulled on the throttle body is only slightly different and it is now surging and falling but not as violent or marked as you would expect from a duff air flow meter or ICV (I suspect this is because the pump is ramping up the pressure then letting it fall again until it drops to a point where the pump ramps it up again) however the change when the car is driven is marked.

This was not apparent at first as the original test where done without driving the car more than a few yards in and out of the garage. today I was able to take it for a scream around the block as the weather was a lot better.

With the return flow completely restricted the car is now drivable and there is only hesitation now when the accelerator pedal is really pressed down hard (I can still not wheel spin it off the mark but at least it no longer coughs and splutters embarasingly to a back firing halt so it crawls off the mark at 5mph ) but it is safe to say there is an improvement.

Examination of the rotor arm and the inside of the distributer cap revieled that it is probably the original one and I am will be replacing them now as soon as I can get a new set - today is Sunday and the shops are shut (This is more of a precaution on my part however the spark is still good on all cylinders at present)

The fuel pump was checked again and by shorting pin 87 and 30 on the pump relay I could check that the fuel pump gives a constant flow of fuel and does not splutter and that it does not leak when at full pressure. On visual inspection all is well, there is pressure there but my gut feeling is that it may not be sufficient and I will not know for sure untill I put a pressure gauge on it.

I have only done around 5 miles in it since resetting the ECU computer so I still have another 10 or so to do before the ECU finishes relearning the engine settings

Now it drives I will do a few more miles in it tommorow to finish training the ECU. My choice is between a fuel pump and the pressure regulator. Hopefully tommorow I can lay my hands on a fuel pressure testing gauge so I will know for sure if the pump is supplying enough pressure.

I am leaving the car to stand at present to see if it is hard to start in an hour or so (test of the fuel pump non return valve - with the return pipe blocked off pressure can only bleed away back through the pump.

As you said, the ICV could be faulty however the problems do appear to be on sudden acceleration rather than idle.

Does anyone have any idea what the fuel line pressure should be?

At least at this point it looks as though the replacement part that is going to be needed is expensive but easy to fit. I really hate spending days trying to fix a fault and it turns out to be a stupid 5p rubber washer but you need to take the whole car to bits to get to it (think along the lines of stepper motor to vent flap linkage on the E32 climate control unit behind the dash board).

[edjack]

My E12 pressure was on the order of 50 psi. Consider doing a fuel rate delivery test. The pump may be capable of pumping up, but still not deliver enough fuel.

As I remember, I ran the E12 220K miles with the original fuel pump.

Yes, the whole business sure sounds like a fuel delivery problem. Looks like you're on the right track.
_________________


Ed BMW CCA member since 1987 (Nr. 62319).
'97 540i 6 speed Aspensilber over Aubergine interior.

[Paulchartley]

Hi All

We have a resolution. Blood sweat and tears as well as quite a bit of oil has been shed over this one.

In my case we have three things going on here not one. You may be more fortunate and just have the one.

I am going to post the resolution for this one as the only other instance of this fault that I found on the net does not. I am also posting all the other posibilities that it could be which in my case it was not but in your case it may be.

It is apparent that there are some differences between the US and European models of this car, principally the US models all have Catalytic converters, lambda sensors and the like and so have a feed back loop to the EMU. The European models do not always have this as in my case. Also the US version alows you to do the stomp test to interogate the CPU for fault codes via the check engine light. The European models do not always have this and in fact the UK E32 that I own does not even have the check engine light.

The car I am refering to is a UK E32 735iL with the M30 engine however most of this is valid for the UK BMW 5 series with the M30 engine

As I said the failure was sudden with no warning. Unfortunately (and fortunately) the true cause showed itself in spectacular style. At the heart of the matter was a blown head gasket. The head gasket had blown past the metal seal on the very back lefthand corner under the oil pressure switch past the copper seal in the gasket into the oil channel and so was over pressurising the oil system causing a significant amount of pressure in the rocker case (this was evident when the oil cap was removed) This as can be seen is connected to the rubber boot between the air flow meter and the throttle flap and so when the engine revs increase suddenly it was putting a pulse of air pressure into the inlet manifold after the air flow meter - enough to slam it shut and cut air to the engine aswell as obviously diluting the air fuel mix. It was also putting pressure into the top of the fuel pressure regulator by way of a vacuum hose which should be a low vacuum and so damaging it. Fuel was coming out of the connection from the vacuum hose and draining directly into the intel mainfold.

In my case both the air flow meter and the fuel pressure regulator were replaced as the BMW computer and the BMW expert operating it said to do so.

The failure of the head gasket was only apparent once oil started spraying out from between the head and the block however examination of the gasget once the head was removed left little doubt that this was most likely eiter the original cause of the problem or was brought about because some other component had failed and the head gasket was on its way out anyway - the material the gasket is made from was crumbling to pieces when examined after removal. Strangely though there was a small amount of coolant loss prior to the symtoms I descrided initially, not enough to suspect the head gasket was failing, this was minimal and there was no water mixed in with the oil.

So that was issue one and two.

after replacing the head gasket and putting everything together again the engine just would not start at all and after much wailing and gnashing of teeth and after loosing most of my hair from pulling it out in frustration virually all of the sensors around the engine were retested without any luck.

Fortunatly this morning a crash damaged e32 with an identical running engine was dropped off at my local breakers yard (due to a back end shunt it was now a 735i compact but the engine was OK) so I took the opportunaty to buy all the engine sensors - impulse sensor off the front of the block, Throttle position sensor (TPS) off the throttle unit, thermo switch (the blue temperature sensor on the top of the thermostat housing) No 6 spark plug lead with the pulse sensor, the inlet control valve that plugs into the back of the rubber boot between the air flow sensor and the throttle mechanism as well as the DME unit (the EMU marked "Motronic" in the black box at the back of the engine bay - the other two are for the cruise control - the one with the blue connector - and the ABS - the one at the front of this compartment)

After swapping various sensors the engine finally fired once the TPS sensor had been changed - it appears that this was operating as a switch rather than a potentiometer. It was either regestering fully open or fully closed but not regestering anything inbetween so it was telling the DME that the engine required either no fuel of lots and so flooding the cylinders

I suspect that this component probably died as a result of the engine overhaul but I also suspect it wasnt that clever to start with and in theory a failed TPS unit could give exactly the same symtoms as a failed air flow meter, fuel pump or fuel pressure regulator. Its just that in my case it was the head gasket.

To anyone trying to figure out this problem like I had to I would recomend you bear the following in mind.

The air flow meter is one of the most expensive parts that can show this symtom if failed and in my case the BMW fault finding computer said that this was the fault - the over pressure in the inlet manifold was closing the baffle flap and the amount of air in the inlet manifold had no relation to the air that was coming past the air flow sensor because some of it was coming from the cylinders via the oil distribution system. The symtoms that I got where classic stuck air sensor flap (well it was being pushed shut when the throttle was opened by the pressure coming in from the damaged head gasket) so the air sensor was replaced. These are like hens teeth to get hold of in a breakers yard and have to be ordered in specially at an auto factors or car parts supplier. Brand new they cost around £200

The same symtoms are classic signs of a failed fuel pressure regulator as there isnt enough fuel pressure in the fuel rail to feed the injectors when the throttle is suddenly opened - the pressure regulator is letting the pressure bleed away - an easy way to check is to clamp the return to fuel tank pipe closed so that the pressure can not bleed away past the regulator. Another clasic sign is that it takes an age to start the engine in the morning because the pressure has to build in the fuel rail. A new one of these costs around £60 and has to be ordered in for most auto factors also.

The fuel pump could be failing. If the fuel pump can not give enough pressure to the fuel rail then the symptoms are the same as the failed fuel pressure regulator only the trick with the clamp on the hose back to the fuel tank does not work. The BMW fuel pump is a common problem for cars that have stood some time and it is possible for the seal between the damper on the top to go so even though it runs most of the fuel runs back into the tank after the fuel pump so a leak is not apparent. Pump units with a plastic top rather than a metal one have a bad habit of splitting. A new one of these costs around £120 and also has to be ordered in. It is a piece of cake to change the fuel pump. There is an inspection hatch in the boot and the whole job takes around 15 - 20 minutes. Remember to also buy a suitably sized jubilee clip to connect the pump to the flexible fuel hose as you will probably find that the existing pump has one of those crimp fittings on it that will need cutting off.

A failed TPS unit has identical symtoms to all of the above and it is a pig to test. The Haynes E34 535 book (The bentley E32 735 book may be different but I couldnt lay my hands on a copy of this) says that to test the TPS unit you need to put a multimeter on to pins 6 & 4 and 5 & 4 and that the unit is OK if there is no resistance on 6 & 4 at fully shut and 5 & 4 at fully open but gives no indication of which other pins to test and what a good result is in between when it is supposed to be reading a half opened throttle hence the reason that I believed mine to be working. Swapping it with another one was another matter and proved that mine was duff. I could not get a new one of these for love nor money as my local BMW specialist car part suppliers do not appear to carry it as a stock item so I guess they dont fail very often and they cost a lot to replace. I got mine second hand with a load of other sensors and stuff for £10 from a breakers yard. Changing the unit is a walk in the park as all you do is unplug it and unscrew two screws. reassembly is the reverse of removal. Now lining it up correctly is a different matter but noting the exact position of the old one before you remove it is a good estimate. If the engine runs a bit lumpy or drinks fuel after replacing this part then a bit of fine adjustment will be needed.

Workshop manuals

It is a point to note that though Haynes do not publish an E32 735iL workshop manual in the UK (or anywhere else for that matter) and the only workshop manual appears to be the £80 Bentley book which you would probably buy mail order or on the internet (and though it is undoubtably better than the Haynes manual it still costs £80 and takes a week to arive). Haynes do publish a workshop manual for the E34 (BMW 5 Series) which is very similar to the E32 and for most parts can be extremely useful in a tight spot and is readily available off the shelf down at Halfords for around £14.99. BMW used the same engine in the 535 and the 735 and although the book is aimed at the home mechanic you can use it to take the engine to pieces and put it back again sucessfully without much fuss. There are some usefull basic faultfinding charts in the back of the book aswell

Most of this Haynes manual is also available on the net as a PDF document to download from a web site somewhere in Slovinia. I will not say exactly where but if you were to google the words E34, BMW, Manual and PDF it would be one of the top 5 results. Some chapters are missing but if you dont have £14.99 because you have spent all of your cash on replacement parts for your BMW then it is a handy alternative.

Many thanks to Edjack for his advice.

Happy motoring

[edjack]

Paul,

You get an A+ for tenacity! Congrats!

And thanks for the post-game analysis. Hope it helps another hapless victim.
_________________


Ed BMW CCA member since 1987 (Nr. 62319).
'97 540i 6 speed Aspensilber over Aubergine interior.