Anyone solved an EML (Drive by wire throttle) crash?

[Manolito]

I know my problem is not unique as I find multiple references to it on different forums. There are even a couple of YouTube videos of it, though it was apparently more common on earlier e46s.

To my frustration, I found only one person who has ever posted anywhere how they ultimately resolved the problem. One guy claimed to have solved it by buying a junkyard wiring harness and cutting out the parts related to EML and taping the replacement harness to his existing harness as a bypass.

2004 325iT Manuf 9/2003 M54 Engine MS45 Engine Control

E46 BMW EML p1634 Throttle valve adaptation sping test failed - YouTube
This is what it looks and sounds like from the driver seat when the EML crashes completely (as opposed to Limp Home Mode)

Sometimes I get EML, Brake (traction control) and Check Engine warning lights when it crashes. Other times I get no warning lights but it crashes just the same. It used to happen only when the car was warm, but now it will do it when the car is cold as well.

The crash is triggered when I give it just a little too much throttle, such as on a freeway on-ramp. It crashes the EML system and looses all power. Pressing the throttle pedal has no effect whatsoever. If I press in the clutch. it drops back to this 1000 RPM surge that you see in the video. Rather than pressing the clutch, however, I turn off the ignition (not so far as to lock the steering) and immediately turn it back on to reboot the EML and the car instantly restarts and drives normally until the next time I give it a bit more throttle than it wants. Turning off my ignition for a half-second probably startles the drivers around me at night because it flashes my headlights off for a half-second.

My Peake tool reports the code for "Air Path" the BMW diagnostic scanner says the code is related to the MAF, but has no specific failure criteria as most codes do in the BMW book. Even AllData has no specifics on failure criteria or diagnostics for this code. Sorry I don't have this code number in my notes right now, but trust me, knowing it is not helpful. No, there are no leaks in the air path.

Done so far:
Replaced MAF with new OEM unit - no change.
Checked wiring to MAF - no opens or shorts
Tried throttle pedal from wife's 330i - no change
Replaced throttle body and calibrated - no change
Checked plugs at throttle body, MAF, and DME for corrosion - none apparent

My next steps:
1) Connect an Android phone with Torque Pro app and monitor everything it will tell me about the air inlet system while the system runs and crashes.
2) Go to BMWTechInfo.com and pay $30 for decent wiring diagrams because the wiring diagrams for e46 in the Bentley manual are worthless. Check all wiring in the EML system.
3) See if I can cut open the plugs at the MAF and Throttle body and solder over the crimped connectors without completely destroying the plugs.
4) Try the junkyard wiring harness approach, though it may be tough to find the correct one since there were running changes in this electrical system.

Has anyone else had this problem and solved it?

Thanks.

[bazar01]

You have a big challenge ahead of you. Is it worth it now to bring it to an indy shop for diagnostics?
It looks like it is already beyond a home diy.
Good luck to you.

[Manolito]

I had it to what is arguably the best indy Mecerdes/BMW shop in town. I paid him to put on the new throttle body because I understand that a new throttle body has to be calibrated using some special BMW tool. He kept the car for two extra days running some tests and researching the problem (for which he charged nothing) and came up empty handed. His next best suggestion is a new wiring harness, though he found no faults in the wiring. He seriously doubts that it is the engine control computer because he has never heard of one of these computers failing, though that is a possibility.

[Manolito]

I think I know what the problem is. I think that my front catalyst is plugged up.

The code is P16A7, which means "MAF signal implausible". I missed this because my test for obstructed exhaust is to put my hand at the exhaust opening and feel for individual bursts of exhaust. That is not a good test when you have two individual cats feeding the same exhaust opening.

The reason the 'MAF signal is implausible' is because there is so much back pressure at high RPM that when I give it more throttle, the MAF is reporting very little air throughput relative to the RPM, the throttle opening, and ambient pressure.

The reason I suspect the front catalyst is because this car consumes oil and the #1 cylinder is about 15% below the others in compression so I suspect that is where the oil is going.

Yesterday I pulled both the upstream oxygen sensors and took it for a ride. I was afraid I might warp an exhaust valve or set fire to something under the hood, but nothing bad happened. The important thing was that I was unable to get the EML system to crash while driving that way.

[bazar01]

That makes sense.

I would drill a small hole before and after the cat and measure the pressure drop across to check the CAT condition.

[Manolito]

Tough to drill holes, but if I had not tossed out my air/fuel sensors when I replaced them, I would turn one into a tap to check pressure at the upstream oxygen sensor hole and compare the pressure at both manifolds and perhaps one of the manifolds on my wife's 330. I will ask a mechanic shop to hold a junk oxygen sensor for me.

The plugged catalyst theory had a couple of setbacks. A borescope inserted in the manifold indicated that both catalysts looked just fine on top, and the vacuum signal from the manifold air pressure sensor, as read using the Torque ap on an Android phone, indicates a healthy 22" of vacuum at no load or light load at any RPM. The vacuum is the same on this car as it is on the healthy 330, except that flooring it at high RPM pushes the 330 vacuum all the way down to 3" Hg, but on the sick car, when the vacuum gets down to 8" Hg, the system crashes and I loose throttle control. I still suspect a plugged catalyst, but at $850 ea and a PITA to replace (must lift engine), I want to be sure before proceeding.

[Manolito]

I was correct about the plugged catalyst, but incorrect in my guess as to which one. I dropped the exhaust system today. That was surprisingly easy, though the exhaust system is surprisingly heavy.

Started car and put my hand at openings of downpipes. Clean bursts of exhaust coming from front catalyst, almost nothing coming out of the rear catalyst. This is surprising because when the computer sets misfire codes, which it occasionally does when the EML crashes, the misfire codes are for cylinders 1-3.

Removing the catalysts is a PITA, and you have to remove the front one to get the rear one off. You also have to hoist the engine to remove the passenger side engine mount to get them out. I elected to go with a rented engine hoist rather than jacking up the engine with a floor jack.

The cats actually look good from both sides, I am going to try back flushing the cats with a pressure sprayer. Probably won't work, but I have nothing to loose.

Next day..........

Lifted engine and dropped the manifolds which are one piece with the cats. Yup, PITA

If I plug the lower opening with a plastic bag and blow in the lower oxygen sensor hole, the front cat flows freely. I do the same thing with the rear cat pipe and it is almost air tight. The only reason my rear three cylinders were working at all is the exhaust from Cyl 4,5,6 was going through the secondary air system to the front manifold.

Tapped the plugged catalyst with a light hammer and got a spoon full of what looked like white abrasive sand from a sandblaster. Sprayed a half a can of WD-40 in the cat and let it set soaking overnight. Then took it to the car wash this morning and sprayed $2.00 worth of hot, high pressure water backwards through the cat. I say "through" but in fact, hardly any water made it through. It is still completely plugged up.

The only source for a California-legal cat for this car is the dealer. If I buy any other brand, I have to have it shipped out of state and smuggle it in. I have no problem doing that, but I worry that an aftermarket unit may not fit perfectly, so I ordered a new cat from the dealer today.

More later...........

[Manolito]

CONCLUSION (for now)

The new catalyst has resolved the problem for now. The car has more power and accelerates smoother, and the throttle circuit does not crash any more.

I bought a cat from the dealer ($850 +tax) because I wanted to make darn sure it fit perfectly, and it was less hassle than having a 49-state cat shipped to an out-of-state address and having it forwarded to me in California.

I cobbled up a pressure test tap made from an old oxygen sensor, but it failed soon into the first test. I know that my front catalyst pegs a 10 psi guage, so it sounds like that one is not in very good shape and will need to be replaced soon. My pressure test tap failed early in the test drive because JB Weld won't take manifold heat, so I have to make a better one before I can measure the back pressure in the new cat to have a 'known good' pressure reference point.

In conclusion, what I have learned is - things that can cause an EML system crash include:
Failed throttle body.
Poor electrical connections - throttle pedal-MAF-throttle body-computer.
But if you get a P16A7 code like I did, suspect a plugged catalytic converter (or two).

[Manolito]

Upgraded my pressure tap by pressing a 1/4" brass plumbing fitting into the hollowed out oxygen sensor body, and attaching an 1/8" copper tube 18" long to the fittings. To that I plug my hose, which runs to the front seat of the car where I have a 0-10 psig pressure gauge. The copper tube was necessary because it gets so hot down next to the hollowed out oxygen sensor that the bottom of the copper tube discolors.

Pressure readings above old catalyst. (140k mi)
Idle – needle rattles at around 1 psig
Light load stoplight cruising up to 2500 RPM - 1-2 psig
Heavy foot between stoplights – up to 7 psig
WOT at 3000 RPM – easily pegs the needle at 10 psig

New catalyst
Idle – needle rattles at around 0 psig
Light load stoplight cruising up to 2500 RPM – 0.5 – 1 psig
Heavy foot between stoplights – up to 3 psig
WOT at 5000 RPM – 4 psig

Conclusion:
My ‘good’ old catalyst has two to three times the back pressure of a new one