2002 3.0 X5 engine stops while driving
 

X5 E53, 2002
Engine: M54
Trans: GM 5L40E
76K miles

The car has seen little driving in that last couple of years, just occasional runs to stores etc. never more than 30 miles each trip.

Yesterday I took a longer trip. 60 Miles highway driving 75% at ~70 mph, 25% with traffic jam going as slow as 25 mph. Arrived at destination no issue.
Two hours later, start driving back. Driving slowly (<40mph) for about half a mile, the car lost drive. I gave it some more gas, but rpm was not responding. Then the engine just stopped and "service engine" light came on. Coasted to a stop. Attempted to start the engine twice. It started to run, but then immediately dies. No error msg other than the "service engine light". Had the car toweled back home.

Next day I tried to use a Foxwell scanner to read diagnostic data. But for the three key items
• Motor Electronics2 DME
• Transmission Control EGS
• Transmission control unit EGS
I got no information because the scanner says no communication with these units. Tried multiple time with key in different position.

Finally I decided to try start the engine again and it started right away. But it was running rough especially when the engine speed start to drop from the initial higher rpm. Eventually it settled to a smooth idle. I tried to engage both reverse and forward gears, could move the car ok, but didn't want to risk to drive it more than a few feet. During this time, the "service engine" light didn't come on.

Tried to use the scanner again with engine running, still no communication. All this took about 15 mins, then the engine started to run rough, and the "service engine" light came on the engine shut off itself.

One other interesting thing was that when I turn the key to run position, I hear a high pitched whistle, last may be 10 seconds. This was something I never heard before, it is on top of the normal sound a X5 makes when you turn the key to run position.

Now I have no clue what is wrong with the car, hope someone could give me some hint.

Thanks!

I would suspect the fuel pump for starters. The Foxwell fans can maybe help with the scanner.

Thanks 80stech.

That is a good one. I was wondering if the ECU are dead, but then the car should not start at all. I will go investigate how to diag fuel pump problem.

+1 on the fuel pump

I guess I'll step in as the Foxwell fan[emoji1787]
But honestly I recommend you choose the auto option for reading the VIN as sometimes a low battery will let a VIN from another installed car part shine thru and that causes something like a logic loop (guessing the term)

I would also check your battery/alternator voltage, function.

Sent from Embassy network using Tapatalk

How much fuel is in the tank?

This could be the infamous O-ring issue that happens when fuel is low (< 1/4)...

That's another +1 I don't know why I didn't remember that one.

Sent from Embassy network using Tapatalk

The start up whistle may be the auxiliary engine blower fan. Located on the fender right side. It runs a few seconds at start up to add air into the exhaust. Mine failed a while back. Took three units to keep the service light off. A Hella replacement air pump works great. Two other brands didn't work. If the noise is inside, it's probably the blower motor fan. Mine does it too. Mostly while driving though and not often.

Fuel and bad battery are my primary guesses. Check the battery voltage. If the battery is over three years old, I would definitely keep it on a trickle charger.

Wow, low mileage survivor. Post a small pic or two of it.

That or alternator. When that fails you'll get DSC/transmission errors.

The fuel pump runs approximately 6-8 seconds at key on.

Is the whistle coming from under the back seat right?

Fuel pumps act intermittent at end of life and last about 5000 hours on average.

Divide your odometer by your average speed to see how close to 5000 hours.

Example my fuel pump failed at about 134,000 and 27 mph average

134000/27=4962.963 lasted exactly the expected amount. Wife's also was within a couple percent of 5000 hours.

With pure gas 4000 hours. E20, 6000 hours. So they have discovered the only good thing about putting an entire year of food for a person into every suv fill up: longer fuel pump life.

My fuel pump has ~6,538 hours! :wow:
Yeah, it's on my to do maintenance list. :p:

Borrowed time for sure!


Sent from my iPhone using Tapatalk Pro

I replied to the OP about his problem on another forum earlier, suggesting that his problem might be electrical in nature. And that the whine he heard was possibly the alternator or Secondary Air Pump.

I also said that when my battery went bad (failure to retain above 12 volts for long after charging), I also had the codes he posted. I think that he needs to check alternator output, and localize exactly where the whine originated. I would think that if the fuel pump was whining, he should be able to instantly recognize that it was inside the cabin.



Sent from my moto g power 5G - 2024 using Tapatalk

On this forum he's saying this happened yesterday ??

as previously posted on Bimmerfest
 

see post #1, posted at 1:26am August 10, 2025 https://www.bimmerfest.com/threads/2...1481330/unread

Yeah, I'm not a fan of this confusing stealth cross posting!

Thanks everyone for all the replies. The problem is definitely pointing to fuel pump --

I dug out my old generic scanner and found:

Live: DataStream
Fuel SYS1, OL, not ready CL
Fuel SYS2, OL, not ready CL
Error Codes:
P0171 Pd Generic System too lean (bank 1)
P0174 Pd Generic System too lean (bank 2)
P1083 BMW Fuel control limit mixture too lean (bank1 sensor)
P1084 BMW Fuel control limit mixture too lean (bank2 sensor)

When I turn key to run position, I don't hear any sound coming from under the rear seat, so seems like the pump is not running. The noise I mentioned was coming from under the plastic engine cover - I was able to track this down after realizing had the HVAC fun running, turning that off helped. I guess I will worry that later. May be the fuel system is trying to do something without getting enough pressure from the fuel pump. I watch a bunch of fuel pump replacement videos last night till very late, so task today is open up the access to the fuel pump and see if I get any voltage to it and if it is pumping fuel out.

It did the VIN reading fine. I first did the quick scan option and it scanned a bound of stuff and came back with no error code reported. That was when I went to the specific module scanning mode and got no communication error.

Very disappointed with the Foxwell scanner. Supposed to be a way better scanner compared to the generic ones. But it hasn't given me more info then the generic every time I used both, now it does not even work. I was using it simply based on the believe it is supposed to be better because it costed more, with specific software module for each car make, which of course one has pay for each. The generic scanner works for any car.

One other thing with the Foxwell is that it takes forver to read anything...

What is this O-ring issue? Indeed the fuel was just less than 1/4 tank.

I am pretty sure my average speed was more than 27, with only ~70K miles, it would be way less than 5000 hrs. But it is 24 years old.

If you do a search on "E53 jet pump (o-ring)" or something like that. you should get a lot of info on that.

Get a gas can and put in a few more gallons.
If your problem goes away, then it is the infamous O-ring.

I used the modified O-ring bought on eBay yrs ago.
I don't think it is on eBay now

Sorry guys, I thought these two forums have different people on them, so posted in both places to get more ideas. Didn't try to do anything stealthy.

I will now only carry on the discussion here. Once I have this all figured out, will post the same findings there.

Thanks CN90 and 80stech, I read up on the o-ring thing it all fit problem. Put some gas in and it started up right away. Now onto open up the sender side and checking out the o-ring.

The key thing is finding the O-ring a tiny tiny bit bigger than the factory O-ring.

All info is in this DIY I wrote back in 2018...

https://xoutpost.com/bmw-sav-forums/...flon-wrap.html

Also,

If you don't have time/desire to fix this issue, then keep fuel level > 3/8 at all time.

It took some doing to open up the sender side. The plastic part is semi-transparent especially if I shine light onto it, so I can clearly see the o-ring through the plastic. It looked perfectly seated in its grove, nothing like what was shown in the pictures posted on-line. I was very baffled and decided to take the thing apart anyways. When I pulled the pieces apart, the o-ring just popped out of the grove! Turned out, it was broken. So even still seated nicely, it was no longer air tight.

Onto to ordering replacement o-ring and putting everything back.

Thanks to all the replies to my posted and guided me to this fix!

Yes, I read through your post, that was what gave guts to try pull the sender side apart! Thanks for that!

Where did you find the larger o-ring? Do you remember the size? I can't measure from the broken one...

Did you check with the pump running if the O-ring was leaking ? You should be able to get away with the same size and a few wraps of teflon tape as well. I think the o-ring is usually pushed out if it has failed. Sometimes the lower part which is a plastic to plastic fitting can leak too which you could see with the pump running.

I found this on Amazon, figured one of them should fit :-)
I am thinking either the 22 or 23.5mm ones.
The o-ring grove measured 23.38mm, a very strange size.

Ah, didn't check with pump running, but now I already took it apart :-(
But does the pump send fuel back to the sender side?

Get O-ring size 22 x 2.5 mm (I.D. x thickness).
You can find it at "O-ring and more" website, Henrico, Virginia.

PS: iirc, factory is 22 x 2 mm. The difference of 0.5 mm in thickness makes a difference.

A note for everyone...

After you fix this O-ring issue, to test the siphon pump system, once the fuel level is approaching 1/4, carry a gas can (the stuff you use to fill the lawnmower) in the trunk with 2-3 gallons, then keep driving so it drops to 1/8. This is how you safely test the system w/o being stranded.

On long road trips with gas stations far apart, keep fuel > 1/4 at all time if you can. It is really not fun being stranded in the middle of nowhere. In the city, you can easily get help. In the middle of the corn field, only the cows have sympathy for you.

You do not need to fill up early just confirm your siphon jet is working.

You do this by monitoring test six and when the total fuel level gets below twenty seven liters, the right side should stay at one point four liters.

If the right side drops below one liter, stop and get gas. You only really need to do that test one or two times a year.

The o-ring failure is from a design defect. The o-ring seal is held together from the center of the assembly so it just torques sideways. You can put a shim in the opposite side and it's forever fixed.

Every first gen x will have the problem eventually.

Great, that is one of the sizes in the set I ordered from Amazon, which I can get tomorrow morning. Thanks!

Yeap, this is what I will do to test！

You should be looking for leaks (gas gushing out NOT air in) that way you can be more sure that you found the problem(s) (maybe the lower fitting is leaking as well) and can be sure when/if you have them fixed.

Thanks Andrew. What is test 6? Is this something available in some scanners?

In my case, I don't see the misalignment yet. I will look more into the zip tie fix.

Got it, I will do that after the fix to make sure the fix is good.

If the pump side of the tank is empty, would I still see any gas coming out of the sender? I though the sender sends gas by syphoning, which would require some fuel from the pump to prim it.

The pump side isn't empty.

When I had my problem it was empty, right?

On this forum you said the car was running again but you didn't want to risk driving it right? So when CN90 told you to put gas in the tank and the car started it really meant nothing because the car was already starting and running. Unless there is something I'm missing from what you are posting on other sites.

When leaking siphon jet causes fuel starvation it's not uncommon for after a rest, environmental changes will allow a little bit of fuel to siphon over to the pump and get the car to start but usually run terribly and maybe just fifty feet.

If the car stalls and putting in two gallons of fuel puts you back in business you are almost assured of the siphon jet o-ring failure.

Test six is in the hidden menu. YouTube or google is your friend.

Bring up test six on the dashboard and it will report real-time how many liters of gas are in the left side and right side tank.

To confirm o-ring failure open up the left side and lift it just a little bit. Wrap with shop rags you can burn if gas sprays and just turn the key to on.

The pump will prime and send fuel back to the siphon. If the o-ring has failed it will spray sideways.

(And directly into your iphone if you happen to be recording to shown to others).

If you pull out the sender unit it's usually quite obvious when the o-ring has failed it will be pushed out of its groove.

Some times the o-ring itself will fail and leak without visible clue though so the only way to be sure is key on while observing.

You can be 99% sure just from reading test six.

When total fuel is below twenty eight liters, the right side should stay at ab9ut one point four liters. If it drops to zero the siphon pump is not working. Almost always from o-ring failure but can also be from the electric pump end of life so it's not one hundred percent.

[mention]kfm [/mention] which foxwell are you using? Only the 500 and 700 series are useful the others are just like generic.

510 and 710 are the ones I have and they are very good. Mostly because they are bi-directional and have solid BMW modules.


Sent from my iPhone using Tapatalk Pro

OP's other thread is on BimmerFest
 

In his parallel thread on BimmerFest, the OP stated he was using a Foxwell NT530, and didn't trust it. I have no experience with the 530, so I can't comment on it, but my NT510 Elite has done all I've wanted from it, and it's paid for itself many times over.

530 should work very much the same but it's non intuitive and there is a learning curve.

In my case, the o-ring didn't blow out. If fact, from outside, it looked perfectly seated in its grove and the two parts that holds the o-ring has no misalignment in angle. Except the o-ring was broken, which became obvious when I separated the two parts. I didn't realize I could test it to confirm it was bad. But I am pretty sure no way that o-ring could have kept the fuel inside. After repairing, - just got my o-ring delivered - I will test.

I have indeed the NT530. My problem is that it give "No Communication" when trying to test the engine and transmission ECUs. And quick scan tells me no error code found.

Also, it took forever, like > 5 minutes, to get VIN, establish communion, downloading info, before I could do any useful diag action when it worked.

Ah, sorry, it was a convoluted story, I dropped some details in between the events. What happened was that after it was towed back home and parked on the driveway which has a decent slop, I could start the car again. I didn't want to risk driving it but also didn't want to leave it out of the garage because rats will at first opportunity get into the engine compartment! I saw rat dropping in there from just that one night outside. So I drove the car back into the garage. But then it stopped runing again. Altogether it run may be 5 mins.

I speculate what happened was that when we loaded the car onto the flatbed tow truck, the car was serious titled front up. When it was parked on the drive way, it was front down. These inclination changes must have sloshed fuels from the sender side to the pump side, allowed me to run it again briefly.

But now the pump side was empty again. Until I put in more gas following CN90 post.

So my question was meant to confirm, when I couldn't start, i.e., no fuel in the pump side, then trying to start the car won't see fuel coming out of bad o-ring. I believe based on all the additional postings here, this is indeed the case.

Typically from low voltage will have a problem with communication. It's recommended to be on a battery tender but they should usually work.

When it won't scan the VIN try manually enter or use a saved scan to start.

Also some contact cleaner can help or the problem can be a damaged OBD port.

higher voltage, better scans?
 

Yep, my NT510 Elite is also definitely slow, but actually speeds up a bit when fed with more voltage, when I use the "stable power supply" feature on my "Clore Automotive PL2320 20-Amp Fully-Automatic Smart Charger" on power supply mode. It even seems to find more modules on quick scan than it does when on battery alone, if at 12.5v or less.

Someday, I'm gonna get the NT710, which is surely faster.

Lifting onto a tow truck almost certainly sloshed some gas to the right lobe of the tank.

It won't last long as more fuel is sent to the left side than is consumed by the car.

When i tested mine to confirm leaking, I was expecting fast dribble but i got garden hose right into the lens of my iphone. Definitely put shop rags around the opening when you do this test.

It's interesting to see age killing the o-ring itself vs. the joint bending.

I was actual on battery tender.

It is not that it wont' scan the VIN, just that it takes at least 2 mins (although I didn't precisely time it) for that part of the start up. It does reads the VIN correctly.

The generic scannere works no problem, so it shouldn't be OBD port problem.

I will need to take it up with Foxwell on this problem. The last time I used the scanner about two years ago, it gave a bunch of error codes related heater, AC, and something else. Although the heater and AC were all working fine. I wonder if the scanner may have already been going bad.

It did seem like there was regression from 510 to 520 to 530.

Make sure you have the latest firmware as much of a pain that is.

I have 510 and 710.

The 710 has WiFi and updates ota so that's a major perk.

Still pretty slow on the e53

Yeah, FW upgrade is painful as I remember, I had to mess with a SD card. But will give that a try later.

Thanks to everyone's help, I have completed the fix. Here are a few things may help others new to this problem.

I didn't removed the center safety belt anchor bolt, but was able to put the seat cushion on top of the seat back over the cargo area to fully open up the space to work on the problem.https://drive.google.com/file/d/16yq...ew?usp=sharing

I installed the o-ring, but didn't do the zip tie fix. This was because in my case, there was no misalignment and I couldn't find a zip tie rated for submerging in gasoline. Also, I noticed that the system is designed to have its plastic rod end pushing on the bottom of the tank and through the spring to apply a continuous pressure to keep the lower portion of the assembly against the top portion. I wonder some of the failures where the o-ring gotten blown out were because the rod end slipped in the tank and started to push side ways causing misalignment between the two parts making possible for the o-ring to be blown out.

For the "taflon tap fix", I did it differently because I don't think I could properly wrap the tap securely to guaranty it does not work loose some how and end up flowing in the gas, eventually clog up something. I used a stainless steel safety wire to tighten the parts together by applying just enough pressure to close the gap while be careful to not crush anything.

https://drive.google.com/file/d/1sxp...ew?usp=sharing
https://drive.google.com/file/d/1yS_...ew?usp=sharing

Finally, I couldn't do the test to actually see if any fuel comes out from those repaired areas. I just couldn't keep the assembly internal to the tank sufficiently visible while still have the external pipe attached :-(

I will just have to drive the gas level down to make sure all is good.

Thanks for all the help again!

Hose clamp around is great fix.

Learn test six I'll let you confirm the fix.

Hmm, not sure why the pics don't show. But here they are if you like to check them out:

https://drive.google.com/file/d/16yq...ew?usp=sharing
https://drive.google.com/file/d/1sxp...ew?usp=sharing
https://drive.google.com/file/d/1yS_...ew?usp=sharing

Ah, that was something I was going to ask. Still don't know what is this test six. I messed around with the NT530, looked at all the different thing it can do, but don't see any test for fuel tank stuff :-(

I also sent a msg to Foxwell about my "no communication" problem, hope they can shine some light on it too.

Test #6 in the hidden menu shows the tanks levels, as Andrew has stated.

If you have a high cluster (variable text message along the bottom of the cluster), watch this to learn how to access the hidden menu: -

https://youtu.be/tdVrXp_RQGk?si=h1AcUl2mIruPggBN

If you have the low cluster, there is a slightly different process: -

https://www.bimmerfest.com/threads/h...luster.432740/

Search YouTube for bmw hidden menu. Very important tool you need to know how to use

Wow, very cool, thanks wpoll and andrewwynn.
That QSilver7 document explains everything.

Interesting what I found on my (actually my wife's) X5.
Running the test with the car not running: 17 13.1 20.1 -> TYPO should be 17 3.1 20.1
After idling for about 3 minutes: 15.9 4.2 20.1
The the right tank value keeps dropping until it settles at: 16.6 3.5 20.1

I wonder why the gas was going back to the left tank after engine stops? Since gas is being pumped out of the right tank to be consumed with any over flow going back to the left tank, I would have thought after the engine stops, the right tank should rise from siphoning. I must be missing something here?

The good think for me is that the fix is working.

2002 3.0 X5 engine stops while driving
 

It's not really "siphon" it's really a venturi pump. It pumps left to right when the electric pump is running.

The two lobes are connected at about 28L and the right side only holds 5L maximum when below 27L total.

Usually the right side will land at 1.4L so yours showing more suggests an aftermarket sender/pump or a bent lever arm. (Bent differently than from new).

I tired to bend mine to calibrate to get to zero that's a very bad idea it read wrong all the rest of the time.

All the BMWs i worked on showed 1.4-1.5L right side when right side was at overflow.

One possibility that I've seen is that during assembly the rectangular float was vertical and held the float off the bottom.

That said you should watch test six and drive down to single digit DTE to confirm fuel senders are accurate.

If you see right side drop before it's baseline before left goes to zero you want to investigate.

When siphon jet isn't at fault the pump wearing out will cause similar symptoms.

I'm curious about the pre-running number that aren't actually possible when everything is working normally. I'm hoping was just a mirage from the car doing averaging. (Or possibly the float sideways)

You should run test 21 which resets the computer including gas averaging but you now know how to use test six which is invaluable for gas supply diagnostic obviously. (Warning it will reset the clock and dashboard mph avg etc)

The right side shouldn't be dropping so I would not be confident about having it fixed.

When left is less than 27, right should be about 1.4. Every car can a different baseline you'll only know when driving steady speed on level road.

It's weird that it shows an implausible number at first but i suspect the key was on at some point with the right side disassembled to put that starting number into memory.

The right side holds exactly 5L of fuel when the left side is lower than the driveshaft hump.

It only takes a few minutes to pull the pump on the right side and then you can have a look at how much fuel is there, if the syphon jet lines and the relief valve are still in place (things can get out of place when pulling on them while working on the other side) and check the connector if there is any problems.

You can tip the seat ahead instead of putting it over the back, might be a bit easier ;) Also like you found out on the other side, be careful when putting the pump back in that the foot goes into the right place, the pump can have a tendency to swing outward and miss. ;)

I got a bunch of stuff going, so didn't get to work on the car today.

I am pretty sure everything is original, since we are the first owner :-)

Yesterday I dropped by a store as part of my testing, coming ut seeing the car in setting sun light, so very pretty it still is. Remembering someone asked to post a pic, so put it on my avatar since I can't seem to get the embedding pic working...

My right side consistently settled down to 3.0-3.1L right now. This seems consistent with "The right side holds exactly 5L of fuel when the left side is lower than the driveshaft hump." Why do you think it should be 1.4L?

I won't get much time to work on it again, so will drive it when I can to get the total close to empty and see if it is still Ok.

Every car can be different, for me that was 1.4L both our E53.

There was 3.6L below the lowest level of the float. (Weighed the remaining gas with a gram scale and did the math to measure). That's how i arrived at the 5.0 number.

I'm not sure the range of normal but most who have given input into the gas tank threads have also said they get similar baseline number.

It would not take much angle difference to change that baseline and i believe you had values lower on the right side before the fix. (Which means the method still applies; you just have a different baseline)

Just monitor test six while watching the left tank drop it should be able to go to zero before the right side drops below the baseline.

If right drops to zero you have to get gas asap. Even with the siphon jet fixed, an aged fuel pump can make for a weak siphon and you can't get all the fuel out of the left tank.

Odds are you're good for now.

Just need to confirm you can get left tank to zero (you will still have 5L of fuel). It's safest to have a Jerry can of fuel with you on the down to zero test but if you are near gas stations it's not necessary you should have 10-15 mile range when the left hits zero.

If right starts dropping before left hits zero then get gas right away. Sign of issue still remains on siphon or the pump itself is weak and doesn't produce enough volume to run the siphon all the way down.

I'm curious about your 3.05 ± .05 reading and suspect the float arm got slightly bent from factory when doing right tank exam. I think i discovered you can pull the float all the way up to remove the whole unit but the first time or three i did it, i pulled the pump out and let the float follow behind, it wasn't smooth.

Finally, I got to drive the car to drain it all the way down to near empty with left side down to 1.3 liter while right side still at 3.1. I supposed it does not take much deviation in the float lever to get a 1.5l difference. The car has never had its right tank opened up, so it must came from factory that way. Any ways at least I can say now the problem is fixed.

Thanks to everyone's help!

Your right side baseline is 3.1. Everybody can have a different value. That's great you got it fixed that's what's xo for


–awr–

Using Tapatalk VIP on iPhone

Yep! Thanks again!

Glad you got it fixed.

However, I still follow my own trick...

1. In town, fuel tank can be down to 1/4 to 1/8.

2. Long distance driving: I keep fuel above 1/4 at all time.
Not worth the headache...

This is a good rule for this old car.