When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
So just so I'm on the same page....there is no problem with the blower belt off and no problem under full acceleration correct.?
Blower belt off is smooth and underpowered, as expected. Full acceleration feels nice and strong although it hasn't been on a dyno. If the TB is opening up way too early it causes these issues at part throttle, but kickdown would be wide open anyway so that would explain proper fueling and power under those conditions.
My thought on the smooth driving with no boost is that it would just act like an N/A car, even with weird throttle positions it would still just be air-in and air-out without forced induction to make things exponentially more noticeable.
To me that leaves crazy internal ECU fault, or bad wiring somewhere? I wish there was a way to mount a tiny camera inside the TB while driving haha.
Any ideas?
The TPS signal(s), give you all the data you need, to determine throttle blade angle, unless......that information is corrupt. The MAP is not under boost pressure, with the belt off, also.
Full disclosure: working with multimeters is not my strong suit, but I get by. I back probed the TB this morning and checked for power supply. Now, according to the post pic below, this is our wiring pinout for the TB. First oddity: I got 4.76 volts at pin #6 with it unplugged, so this is my power supply. Ground is pin #3. So already this info below is wrong (and could have just been mis-posted). Since pins 1 and 4 are mirrored on the opposite end, you'd almost think it was plugged in backwards but the flat end on one side of the cable makes it impossible to do that.
Then I back probed pin #4 with it plugged in, see video below. It goes up to about 2.5v (about 30% through the pedal stroke, when the butterfly is now wide open) then drops to almost nothing. Then on the way back, when the butterfly starts to move back it spikes up then comes back down. Both times I noticed a momentary jump of voltage in the wrong direction too.
I haven't hooked STAR back up, but I have done the TB voltage test with it in the past and it passes just fine, so the ECU thinks everything is good (and also explains no codes).
I just watched my video again. How on earth is it possible for the voltage to go UP as it's opening, and then also go UP as the TB closes?? I think this car is haunted.
*Edit: I just realized I have WIS on my laptop, and looking at the wiring diagram the power and ground should be pins 1 and 4 as the posted pic indicates. So I have no idea why I'm seeing something different. I can confirm I have 5v power to pin #6 with the harness unplugged, directly contradicting the wiring diagram, confirmed by location as well as color. I'm assuming this is sent by the ECU, so is this enough to condemn my ECU?
Last edited by GinDistiller; Nov 21, 2020 at 02:17 PM.
I decided to do some wiggling while the multimeter was plugged in, now getting random voltage readings, all under 5v, on multiple pins. This tells me there is a wiring break somewhere. The connector itself looks fine, and the chances seem very slim of the ECU sending voltage signals to multiple wires that its not supposed to, so this seems like the best course of action.
I just don't understand why this wouldn't set any faults.
I'm leaning towards a bad ECU. Yes, I'm getting weird voltages at the TPS harness, but a wiring issue/short would be rare to affect multiple pot signal wires, meanwhile the throttle plate when moved by the gas pedal does in fact move forward in a smooth, linear fashion, just not in the right spot.
STAR shows the voltages correctly, topping out when the throttle plate does so mechanically, and the accelerator pedal voltages are a smooth sweep all the way to kickdown.
This means that the learning process for the throttle stops is not being done correctly, and I can physically watch that happen as it only opens part way then shuts again during the learning process. The other test in STAR for proper signals to the potentiometers requires an ECU wire harness adapter (which I don't have), but says that if the voltage isn't right, the only remedy is ECU replacement...
So here are my options as I see it:
1) Buy a used ECU and have someone program it (or maybe my STAR unit can do this??)
2) Buy the $100 special adapter, test the circuit per STAR to confirm the ECU is needed.
3) Pull the supercharger so I can inspect the entire wiring loom for breaks or corrosion.
We know it can't the TPS itself, since I've tried two and one was known to be good on another car. The accelerator pedal was replaced last year, and voltage sweep and values are spot on. To me that just leaves the above possibilities.
I've replaced one E55 engine harness due to corrosion. It was deep inside the rubberized part of the harness that lays in the valley and wraps over the timing cover, underneath the supercharger of course. That car had 165k well cared for miles. The wire insulation inside was flaking apart. However, that car had a few odd fault codes and ultimately an injector that was stuck open. The ECU was also shorted internally from this and required a new one, purchased and programmed from the dealership. Those "we fix ECUs" places have about a 90% failure rate in my opinion.
I can have my supercharger out in about 30 minutes, however, checking the harness requires cutting the rubber shielding open which is quite invasive and will likely damage the wires in the process. Last I checked a couple years ago, there were a few harnesses available from MB for around $600 retail.
Last edited by E55Greasemonkey; Nov 21, 2020 at 06:30 PM.
I've replaced one E55 engine harness due to corrosion. It was deep inside the rubberized part of the harness that lays in the valley and wraps over the timing cover, underneath the supercharger of course. That car had 165k well cared for miles. The wire insulation inside was flaking apart. However, that car had a few odd fault codes and ultimately an injector that was stuck open. The ECU was also shorted internally from this and required a new one, purchased and programmed from the dealership. Those "we fix ECUs" places have about a 90% failure rate in my opinion.
I can have my supercharger out in about 30 minutes, however, checking the harness requires cutting the rubber shielding open which is quite invasive and will likely damage the wires in the process. Last I checked a couple years ago, there were a few harnesses available from MB for around $600 retail.
Yeah I'm having the same thought, I'd hate to chance an ECU and then have it damaged again from a wiring issue that I can't see. I took off a surge tank and jammed a hand down there to try and wiggle it and see if anything was fluctuating but too much of it is buried. Mine has 155k on it, and I have enough of the coil harnesses secured with zipties from brittle plastic that it might be nice having a new harness anyway, I found one dealer that has one for $550.
I definitely wouldn't chance trying to have someone 'fix' the ecu, and I'd likely try the harness first anyway and see if things clear up. I've gotten quick at pulling the SC with my engine hoist, I'm just bitter about it haha.
Do you think it's a safe bet though to have a used ECU recoded to my car? I sent an email to Tony at RaceIQ asking if he did that kind of work, and I've seen a few places advertising that service for about $200. I have STAR w/ developer mode, however I have no idea how to use that feature and I think you still have to have a subscription to the dealer network to get new SCN coding.
If you really suspect the harness you can bypass the engine harness and make your own wires between the ECU and throttle body. Just de-pin each end and leave those original wires hanging loose. The female pins can be purchased from the dealer and you would then use a crimp tool and your own wires. However, this is an advanced diagnostic step and I don't know your abilities. Before that you should first verify all wires in the suspect circuit with a continuity test not only from point A to B, but for shorts to each other, ground, and positive.
Correct on the SCN coding, dev mode won't help you. Other than a new ECU you may try a used one and have the VIN wiped by an outside company. Then a dealer would have to do the basic programming. This doesn't always work right and by the time you pay for all that, and the hassle factor, you've almost paid for a new ECU.
Last edited by E55Greasemonkey; Nov 21, 2020 at 10:02 PM.
If you really suspect the harness you can bypass the engine harness and make your own wires between the ECU and throttle body. Just de-pin each end and leave those original wires hanging loose. The female pins can be purchased from the dealer and you would then use a crimp tool and your own wires. However, this is an advanced diagnostic step and I don't know your abilities. Before that you should first verify all wires in the suspect circuit with a continuity test not only from point A to B, but for shorts to each other, ground, and positive.
Correct on the SCN coding, dev mode won't help you. Other than a new ECU you may try a used one and have the VIN wiped by an outside company. Then a dealer would have to do the basic programming. This doesn't always work right and by the time you pay for all that, and the hassle factor, you've almost paid for a new ECU.
Thanks, and that all makes sense. I thought about bypassing, the wiring schematic outlines wire sizes and locations etc and it's something I could handle, but here's the kicker: If we're pointing to either the ECU or wire harness, if it happens to be the ECU then I can't in good conscience replace it without wondering why it went bad. Which would be a wire harness issue, possibly an intermittent one that might not show up when I test it initially (as it would be cold, who knows what happens when it gets nice and hot down there). So I'd get a fancy new ECU and then worry every day about another failure.
So, I should just order and replace the harness. If it still has an issue, then we can point to the ECU. If it's fixed, then yay! I don't have to buy an ECU.
As a footnote: The channel under the supercharger had a bit of oil in it after I bought the car, enough that I think some of the loom was sitting in it. I cleaned it the first time I had the SC off, but it makes me wonder..
That is one way of approaching it. Keep in mind, a new ECU from MB has a 2-year unlimited mileage warranty. Replacing the engine harness is about 1 days worth of work for an experienced tech.
Also, don't overlook your fuel system before you condemn the ECU. Low fuel pressure, and/or large swings in fuel pressure (i.e. faulty dampener )can play havoc with these cars.
Last edited by E55Greasemonkey; Nov 21, 2020 at 11:09 PM.
I'm sorry, but this is painful. Talking about throwing parts at a problem, and guessing. Properly diagnosed, is the only way to repair high tech, modern cars. I would take to a independent Mercedes specialist, that has the right test equipment, and experience. There a few in both Snohomish county, and Seattle.
That is one way of approaching it. Keep in mind, a new ECU from MB has a 2-year unlimited mileage warranty. Replacing the engine harness is about 1 days worth of work for an experienced tech.
Also, don't overlook your fuel system before you condemn the ECU. Low fuel pressure, and/or large swings in fuel pressure (i.e. faulty dampener )can play havoc with these cars.
Agreed, but fuel system is perfect. Dampener done in 2018 and the pumps this year, pressure has been just fine when I've tested it.
I'm sorry, but this is painful. Talking about throwing parts at a problem, and guessing. Properly diagnosed, is the only way to repair high tech, modern cars. I would take to a independent Mercedes specialist, that has the right test equipment, and experience. There a few in both Snohomish county, and Seattle.
Normally I'd agree, I spent 10 years managing an independent specialist shop, and if this was a customer's car I would insist on that route so that we'd have a clear conscience on repairing parts. Where I'm at now though is a 100% verified problem between the throttle body and ECU, and I have to address that first before any further testing. Leaving me only two options: Replace the ECM but worry about why a rare failure like that occurred, or replace the harness and see if the problem resolves.
I just can't see the justification of spending 2-3 hours of diagnostic time (about the same price as a new harness) to get me to the same conclusion. Plus the car is 'modified' and I've known too many techs who start going down the rabbit hole or have pre-judgement and spend more time grumbling than treating it like another patient.
Another tip for you, especially concerning replacement of the engine ECU, which rarely goes bad, is expensive, and not returnable. Ask yourself what the next step will be if the ECU doesn't fix your problem. Sometimes that helps you think beyond what you see at the moment.
Another tip for you, especially concerning replacement of the engine ECU, which rarely goes bad, is expensive, and not returnable. Ask yourself what the next step will be if the ECU doesn't fix your problem. Sometimes that helps you think beyond what you see at the moment.
My plan is harness replacement here, then if that doesn't fix it I'll take it in to the dealer since they'd need to do the ECU anyway (if it was needed). And I'd have the backing of their diagnostics should they recommend it and it still doesn't fix the issue.
I can get the harness for just under $500 and it's an afternoon of work for me so not a big deal, and it will be nice having refreshed connectors as so many of mine are brittle and some have cracked (I've got zipties around most of my ignition coils as a backup since those plug clips were broken off before I bought the car).
Another tip for you, especially concerning replacement of the engine ECU, which rarely goes bad, is expensive, and not returnable. Ask yourself what the next step will be if the ECU doesn't fix your problem. Sometimes that helps you think beyond what you see at the moment.
So true, I can tell you that selling a new ECU to a customer is the last possible fix and I make absolute certain that it has be a failed ECU before I sell one.
It's rare that they go bad so I make sure with all the testing as they are expencive and time consuming to replace.
Now it appears to be the ECU. I figured I'd do some more testing since I was planning on replacing the harness anyway, I spliced into the wires at the ECU to bypass the harness and that also gave me some nice 'meat' to read voltages vs back probing with a paperclip.
The short story is that the TPS supply is only receiving 1v (ignore the (-), I had this swapped around when taking the picture, I really need alligator clips that have different colored wires haha. And I know it looks like a rat's nest, but all connections were rock solid so I can trust the readings).
Per WIS (below) their possible causes are cable and ECU, and since we bypassed the cable that leaves nothing else. My readings on the potentiometer values are all over the place but based on my understanding the power supply should be a steady 5v at all times, and the TPS uses this as a reference, so the fluctuating voltages on the other wires can't really be trusted. Plus it acts the same on both of my throttle bodies.
I know you guys chimed in to keep me honest and open to other possibilities, but as a tech would you suggest anything else at this point based on these findings? I can't seem to think of anything. Ugh.
REPLACE all 2 or 3 fuses for the engine control unit. Do not bother checking them. A voltage drop across one of these fuses can cause similar problems, I've seen it and it was after many hours and parts thrown at a car.
REPLACE all 2 or 3 fuses for the engine control unit. Do not bother checking them. A voltage drop across one of these fuses can cause similar problems, I've seen it and it was after many hours and parts thrown at a car.
*Pushes aside children, runs to the garage to rob fuses from the other car*
Nope, no change unfortunately, there are two according to my little sheet (15a near ECU and 7.5a inside cabin), still the same readings.
Mercedes SLR McLaren 722 S Is Extremely Rare Example Modified by McLaren
Slideshow: A one-of-one U.S.-spec Mercedes-Benz SLR McLaren Roadster became even rarer after a factory-backed transformation at McLaren's headquarters.
