Today, the car is running great.  But let us update you regarding a little issue we've been battling over the past year-and-a-half, and thankfully got resolved.

Project M3, Part 8: The Dreaded EML Light

(and what to do about it)

by Pablo Mazlumian

We apologize for the eight-month hiatus with Project E46 M3, but we’ve been hard at work and—even though we had an issue we didn’t tell you about over the past year—the car is now running like a champ.  We’ve even dyno tested a new performance part and added a bunch of tasteful exterior and interior upgrades, but we’ll have more on that soon.  Today we’ll talk about the aforementioned issue that we took care of.

If you haven’t seen the yellow “EML” light on your BMW E46, it’s probably only a matter of time.  It stands for Elektronische Motorleistungsregelung, and it can mean a myriad of “electronic engine power control” issues, mostly having to do with the drive-by-wire system.

Ahh, yes.  The fun EML light.  In Project M3’s case, when this light would come on, the car would shut itself off in a matter of roughly 3 seconds and not start back up.  The engine would turn over but there would be no fuel or ignition, as if we had a dead ECU.  Not fun if you're not sure what to do.

Yes, that’s the windshield fluid low-level indicator, which is empty (because I accidentally drilled a hole through it while putting on a splitter I'll be sharing with you soon).  And yes, that’s also the airbag light, which seems to rear its ugly head on each of the five M3s I’ve owned.  I’m just so used to seeing it that I don’t worry about it any more.


The most common part to go wrong in the EML circuit is for one of the Throttle Position Sensors (TPS) to go bad, and the one you hope to be malfunctioning is this one—the #1 TPS—which is right above the air filter.  It’s just the easiest one to swap out.

The first time I had the EML light come on, it was right after our Corsa exhaust and VAC Motorsport pulleys swap, which was prior to the AEM Infinity EMS installation.  At first, it was puzzling because I’d swapped out TPS #1 and the car ran well, only to have the light come back within a week or two.


When you swap out the TPS, be careful not to strip the Phillips screws, which are weak-sauce.  I later replaced them with these socket head cap screws, which have an Allen head, size M4 0.7 x 25mm.

Unfortunately, after a couple of weeks the EML light would come on at times, and the car would stall again.  I had hoped that the new AEM Infinity install in Part 6 would clear this, things unfortunately didn’t change.  But then it drove fine for months again.  So far, so weird.

It wasn’t until my wife and I took a road trip last November, 2013, from Kansas City to St. Louis (to watch Argentina take on Bosnia/Herzegovina in a FIFA soccer friendly—holy cow there are a lot of Bosnians in St. Louis!) that I figured out what was going on, or so I thought.  First I’d thought we were stuck on the side of the road for good, and our weekend was ruined.  Luckily, we both figured out—OK she spotted it, and wants my readers to know it—that pulling fuse #29, and reinserting it with the key off cleared the EML light (I recently also realized that pulling TPS #1 and plugging it back in would do the same trick).  This trick would work, but only for 20minutes or so.


Page 1 of 7 Next Page
Bookmark and Share
Tuesday, October 28, 2014 10:38 PM
Personal Opinion: Under-drive pulleys are a bad idea on street cars.

They are fantastic for race cars. Racing engines tend to have high idle speeds due to aggressive cam profiles and spend most of their run time at high RPMs anyway. With the engine screaming at 5000+ RPM all the time the reduced drive ratios of an under-drive pulley are fine, and even desirable so as not to over-run OEM alternators and the like.

But on a street car? A vehicle that spends much of its time cruising on the highway and idling in traffic? I think that's asking for trouble. Lower voltage off the alternator puts added strain on the electrical system. Lower drive speed for the power steering pump makes it work extra-hard during low-speed maneuvers with lots of steering input, like when parking. Lower waterpump speeds from under-driving can be especially bad.

Think about the car idling. While running the A/C. While sitting in traffic. On a really hot day. Ugh. Gives me the willies just thinking about it. Why put all of this added stress on the engine in exchange for a couple of measly horsepower?

I think under-drive pulleys represent a wrong-headed approach to tuning a performance street car. You should be looking at upgrades that do not negatively effect, or even enhance, OEM reliability.

Instead of an under-drive pulley you should run an upgraded fluid harmonic damper such as those made by Fluidampr. Such an upgrade smooths out engine response, frees up power that is normally lost to harmonic vibrations, and drastically increases reliability. VAC Motorsports produces an *excellent* fluid harmonic damper for the E46 M3. It works extremely well, and should be considered a requirement when switching to a lighweight flywheel or when otherwise making changes to the reciprocating mass of the engine.

The M3 may be considered a "race car for the road", but that doesn't make race-oriented upgrades such as under-drive pulleys a good idea for a car that lives on the street.
Wednesday, October 29, 2014 4:18 AM
That new intake manifold looks awesome!

And your drive by wire woes remind me of how lucky I am to have found my project car without the optional dbw traction control.
Wednesday, October 29, 2014 5:19 AM
I'm curious, is there not some type of code reader to allow you to home in on the problem more easily than using the trial and error method?
Wednesday, October 29, 2014 7:55 AM
I have been wondering for quite some time if the engine flush fluids and injector cleaners actually do anything for an engine? I am curious to know what the experts at MotoIQ think of them.

Pablo, did you see a difference in the way the car ran with the additives? I have an E90 330Xi that has about 80k on the clock and would like to drive the car for another 10 to 15 years and 200k miles. So if "cleaning out" the engine periodically would help with that, along with the normal maintenance.

I know its gotten a lot less common with the advent of the internet, but automotive snake oils have been around for years, and these just seem to be one of them to me.
Wednesday, October 29, 2014 9:07 AM
I've done underdrives on 2 cars, and on both they included an alternator overdrive pulley so that the alternator was the only item not underdriven, for reasons you've already seen first hand. None of my other accessories suffer from it and I get peace of mind knowing PS is spinning slower and hopefully running cooler. To me, it was more a matter of reduced wear and tear and increased reliability. Some cars' water pumps will cavitate because the intake side is too restrictive. I don't know that to be the case on mine, but the underdrives should make that less bad if it is the case. Out here in CO, it's not uncommon for my foot to be in the boards for a few minutes at a time while my tow pig screams up a hill expressing its displeasure at expecting 5.7 liters to pull 8000# up hills after taking away 20-25% of its oxygen. Be it there or idling in traffic, everything works just fine. Now the Camaro on the other hand, it has a ridiculous underdrive pulley with no alternator overdrive but it's a dumb OBDI car so it doesn't care and neither do I if the battery doesn't die. Since it's a race car though and it has zero electronic luxuries to power, it'll still put out over 13V at an idle. Pablo you should rename this "Project why my Supra will never be finished"!
Pablo Mazlumian
Pablo Mazlumianlink
Wednesday, October 29, 2014 9:28 AM
thanks for reading, guys!

@Strangelifeform/Supercharged111: Yes I agree not to underdrive pulleys for street cars. I,too, have had pulleys on other cars, including on the Supra and two other M3s and had no issues, but the alt pulley was either not touched or it was a stock-sized lightweight unit (preferred!).
And if I end up doing the flywheel I will also be looking to Fluidampr for its crank pulley as well! so far so good on the Supra with that thing.

@cbgoding: thanks! sounds great and released a few ponies as well!

@wrecked: I do have a couple of OBD2 scanners, and those--along with the one at the shop--won't read the car! The only one that will read it is with the Epic Motorsports software plugger they sent. However, now with AEM Infinity stand-alone EMS I don't think anything would read it anymore anyhow.

@blackdbl0si: I do these more as a preventative measure, and I only do them with companies I've trusted, like Lubro Moly and the likes. When I've tested these on the dyno, I do it after the bunch (plugs, oil, new filter, etc) so it's hard to put a finger on what gives me power as an individual part (I wouldn't dyno test with that fluid in there). But I do it for piece-of-mind, similar to why I take my vitamins (not because I like to pee really yellow!).

@supercharged: your last statement my wife canNOT read, please! I've been telling her the opposite so she won't leave. Actually, I honestly did tell her just this past week that I'm pretty much done building it. But then I realized it's really with anything engine-related, because there are a few other things that I'd also like to...Oh man there she goes...

Wednesday, October 29, 2014 10:14 AM
I can definitely say that the Lubro Moly MO2 (Molybdenum Disulfide) engine additive has a very noticeable effect when used. The engine revs noticeably smoother throughout the rev range. I can usually find it at NAPA for about $7.50 a can, but that's the only place that I've seen it for sale. I would be interested in trying their ceramic product, but it costs a lot more, although it is supposed to last a bit longer. I would say the MO2 only lasts every oil change, so you do have to change it somewhat frequently.
Dusty Duster
Dusty Dusterlink
Wednesday, October 29, 2014 1:40 PM
Why are you running an alpha-n strategy?
Wednesday, October 29, 2014 9:33 PM
arnt there resistance specs for the tps sensors and what not so you can identify the one broken part instead of doing all of them? although ive seen parts that pass the ohm test still not work properly
Wednesday, October 29, 2014 9:40 PM
like speed sensors and the like have you check the voltage as you cycle it by hand to make sure the component is working fine
Pablo Mazlumian
Pablo Mazlumianlink
Wednesday, October 29, 2014 9:45 PM
@ginsu: thanks for input!

@Dusty: I've always had good luck with MAP cars, and in this case you need to with AEM Infinity (which is also great with volumetric efficiency), plus I need it to go this route to run the intake we just put on...unless you can find me a 6-in diameter HFM!

@theneil: yeah as mentioned in the article, I spared you guys (and myself) a lot of details in dealing with this. I checked resistance of the different sensors and the pedal and they were all reading the same, or near it. I'm just glad this little crusade is over!
Thursday, October 30, 2014 4:45 AM
I'm rather confused how you're getting an EML light if you're using an AEM Infinity ECU.

the Infinity can control the DBW, did you leave the stock ECU in there to control the throttle bodies instead?

if not, the AEM interface could have happily told you what the issue was (and it would have been very unlikely to light the EML light if there was an issue)
Dusty Duster
Dusty Dusterlink
Thursday, October 30, 2014 5:18 AM
Pablo: Using a MAP sensor is called speed-density.

Alpha-N is something different.

Pablo Mazlumian
Pablo Mazlumianlink
Thursday, October 30, 2014 7:28 AM
@Dusty: thanks for pointing that out. We are indeed using speed-density. I'll have to change that in article because I know I mentioned Alpha-N a couple of times. My bad!

@Marc: The AEM Infinity does control the DBW but is also programmed to illuminate the EML light if it detects "DBW Fatal Error", and shut the car off to prevent unintended acceleration. The stock ECU is totally out.
Pablo Mazlumian
Pablo Mazlumianlink
Thursday, October 30, 2014 7:38 AM
@Dusty, read article. Great read! Just one small thing: I'm pretty sure the stock MKIV Supra Twin Turbo did come with MAF. I remember converting to speed-density MAP when I had the AEM Series 1 installed back in '03.
EDIT: The US-spec cars used MAF, JDM used MAP
Dusty Duster
Dusty Dusterlink
Thursday, October 30, 2014 11:21 AM
Pablo, I believe the Supra's tuning strategy varied by model year.
Mike Kojima
Mike Kojimalink
Friday, October 31, 2014 12:58 AM
I have run tons of underdrive pulleys on various cars on the street and track and have NEVER had a problem. This car seems to be the odd exception. Unless you are into car audio, you won't either. Debating their safety, etc is something that inexperienced people seem to love to do on the internet. I usually find that pulleys are one of the best bangs for the buck.
Pablo Mazlumian
Pablo Mazlumianlink
Friday, October 31, 2014 1:38 PM
Mike, I assume you're referring to underdrive *alternator* pulleys in those experiences? If so I may give this another whirl and see since it's not a huge job and the alternator did seem to give most of the 7whp or so I gained from the pulleys. I've run alt pulleys before, but usually lighter ones (not bigger/underdrive ones). This one is huge, but VAC seemed to recall not having any issues so I may give them the benefit of the doubt and try again--worse case scenario is I have to swap it back after I check voltage and it's not high enough at idle.
FYI I have stock audio and do still have the VAC water pump/power steering pulleys on there--car runs very cool and turns fine, although I don't think they're underdriven very much either.
Friday, October 31, 2014 4:27 PM
@pablo this is something i repeat to everyone saying "accessories rob up to xx hp" (mostly GM V8 users claiming 35HP losses; seriously guys stop believeing everything written on the internet and think about it)

It all boils down to that damn physics telling the energy output of a system is equal to the input + losses. And you can't beat physics, not in that field.

So, if there is more energy to the wheel now, it means it was consumed before; and there is not much energy used up at WOT by the pumps or the alternator, not in the 7WHP (or more) order, meaning it would mostly be losses, i.e heat. Now, a small glowing red turbo radiates around 4-5HP, in heat. Don't you think your accessories and belts would melt, if they had that order of heat to dissipate ?

An inertia based dyno may give false reading when engine inertia changes though... and that usually is what happens when changing pulleys and getting "that more horsepower".

Bot convinced ? Let's do a little maths and research.

Belt driving has better efficiency (approx 98%) than alternator(50%) + electric motor( 95%). I'll use electric pumps numbers i know to approximate, it still is energy, no matter how it is transmitted, a belt driven pump will consume the same.

First, the alternator : 80amps, 15V alternator, 50% efficiency: that is 3.3HP consumed from the crank. 25% reduction on that: 0.8HP gain.

Electric PS pumps are around 12 amps continuous, that is approx a quarter of HP. No saving there, the belt driven pump regulates itself with a pressure relief valve ... unless it can't do its job properly and you paint your engine room with PS fluid boiling out of its tank (i did that last saturday). Anyway for the sake of it...
25% underdriven means approx 45W saved.That is 0.06HP.

Electric water pump: an EWP150 is rated for 10amps max. Approx the same as the PS pump.

So that is approx 3.8HP of accessories power usage, including the piss poor efficiency of the alternator.
We can ignore the approx 70W of belt transmission losses, and inertial losses too, as it does not weight enough nor spin fast enough. You can't do anything about the oil pump, that is why i did not include it (but i suspect it could eat 1 to 2HP). Now 25% of 4HP is 1HP... and that is at the crank. Nowhere near 7WHP (or the 35HP figure from LS guys). Even removing all your accessories, going electric and running on a BIG battery would only net 3.8HP, at the crank.

Feel free to correct me if i am wrong.

Now, although, there is not much power to be gained on underdriving accessories, it does help engine with response, a lot. Maybe not on pulley weight alone, but as far as i know, i use a 12lbs (instead of 24) flywheel, and i also removed the clutch fan on my ca18det. The clutch fan made a world of difference ( a big, free intercooler also freed the 5000+rpm range). The flywheel also made some changes, but not as much.
I think i'll go all electric pumps on my next build (water, steering, vacuum maybe too, oil ... i'd like but it looks way more complicated), and keep only the alternator on the accessories belt. Maybe it will be worth it...
Friday, October 31, 2014 4:29 PM
Damn ... Output = Input - losses, not +. Need sleep, now.
Pablo Mazlumian
Pablo Mazlumianlink
Friday, October 31, 2014 6:26 PM
@Crousti: thanks for the technical insight! Really appreciate the effort. I would be very curious to see how these would result on the dyno with all electric accessories, as well as some real-world stuff like xx-xxmph acceleration or trap speed improvement if you 1/4. Wonder how much all of that could equate to.
I've done I think three dyno tests with ltw flywheels and I've always seen gains in 1st,2nd and 3rd gears (with big drops in inertial power between shifts), but by fourth it starts to even out. I wish I had a dyno at home (I'd be testing everything).
I hear you on what inertial dyno's can truly show or not, given their nature. I suppose the best way to back it up is with equal back-to-back testing at the quarter, where I find that usually each 10whp gain around the 100mph mark in trap speeds equals roughly 1mph (or 0.1 seconds) given identical launching. I wouldn't use it to claim a power gain as opposed to more of a power gain confirmation from what I actually saw on the dyno.
And lastly, just to correct myself after looking at the pulleys article in Part 2, it was 5whp peak-to-peak gain with an 8whp max gain (although we're not talking a totally smooth graph here, so give/take 1-2whp there). And all pulleys are underdriven:
Saturday, November 01, 2014 8:10 AM
I did the math some time ago about the inertia of a flywheel. 4lbs of aluminium on the peripheral section (4" to 6" zone) of a flywheel can store approx 50W at 5000rpm+ speeds. That is approx 0.07HP.
I do agree removing mass there "does something". Yet I know it does not add power. it still increases performance though. I have nothing to prove it, but my guess is that reducing inertia allows the engine to reach higher RPM in the same time frame, thus reaching higher power. Meaning when you could only rev to 4000rpm before, you can now reach 4500, and the power output will be what your engine can give at 4500, not 4000. You now have "more hp", yet engine power ouput, as it is expressed in HP@rpm, is still the same.

Now clearly off topic, but you might find it interesting. An important thing to consider when fiddling with flywheels is that a standard engine power output is a sine wave, not a line. The flywheel is used to absorb the peaks and release that absorbed energy during the creeks, so that it looks like a line to the transmission. Like a capacitor in electrics. If you remove a lot of material, you will feel that sine-like power delivery when clutching. And so will your transmission, meaning it will wear faster.The tyres will feel it too, and they will heat up, wear and lose traction faster too. On motoGP bikes they reached that tyre slipping problems some years ago, so they went to the big bang architecture to reduce the frequency of that sine wave. And it worked ! They now have a single power surge, followed by a long "no power" cycle.
Saturday, November 01, 2014 10:17 AM
Alternator current is dynamically controlled to keep the system voltage at a certain level as electrical system impedances and current flow both dynamically change, so the alternator current isn't constant for a certain RPM and voltage. Because the alternator current isn't constant, the alternator EMF isn't constant, and therefore the engine load is not constant. If you can get the same required current from the alternator at a lower RPM, underdriving the alternator isn't going to present a problem.

That really means that the underdrive pulley debate can be summed up into one simple and fairly obvious fact: Either you're drawing more current than your alternator can supply at a certain alternator RPM to keep the voltage above a certain level, or you're not. If you're not, you aren't going to be affected by the underdrive pulley. If you are, your output voltage will be affected. Because the factors governing current draw dynamically change, and alternator sizing and accessories are different from vehicle to vehicle, an across the board damning of alternator underdrive pulleys doesn't hold up. Some vehicles may respond perfectly well with no change, and some may respond terribly.
Sunday, November 02, 2014 7:19 PM
Love Bimmers but stuff like this keeps convincing me to enjoy them vicariously
Pablo Mazlumian
Pablo Mazlumianlink
Monday, November 03, 2014 6:39 AM
Crousti: Yes these things don't add hp but they "seem" to when testing released hp to the wheels (or allow them to accelerate faster), since that of course is where the measurement is taken. And thanks again for the technical tidbits! Good to have on this page.

@Fabrik8: E36 guys are fortunate in that you can get a 140amp alternator specifically for this reason, which I had on my previous project. Unfortunately, I've searched for anything greater than 80amp for the E46 and I cannot find anything.

@CTK: rule of thumb for me is to not have a project car as a 100% daily driver!

Thursday, December 04, 2014 6:12 PM
I've read your articles all the way back to the ec days. Just to give you a heads up. Be wary of that alternator. The Bosch remans have a terrible success rate in E46 M3 enthusiast, track and race cars. Life tends to be slightly longer than a fruit fly. I even installed one in a car that was no good out of the box, brand new. Guys are having much more success pulling from parts cars. I know one Bosch shop that won't install them in any customer car. If any weirdness pops up, that would be my first place to look.
Pablo Mazlumian
Pablo Mazlumianlink
Tuesday, December 09, 2014 9:45 AM
Thanks for heads-up, Duncan!
Post Comment Login or register to post a comment.

MotoIQ Proudly Presents Our Partners:

© 2018 MotoIQ.com