SICAMG

So last year this problem reared it's head one time and didn't do it again until today.
What happens is there is no fuel enrichment under boost. Will spike WAY past 12.5 AFR and much higher. My fail safe will shut off the compressor at 12.5 and higher so there is no way any damage can happen....which is the reason I put the fail safe in.
With out the fail safe she will peg the AFR.
The only way to cure it is to stop ,shut car off,restart. No more problem until who knows when. Could be next year for all I know.
It's almost like the ECU does not see any map readings but it is reading on the obd gauge.
Fuel pressure is perfect by the way so it is some type of strange glitch and is "cured" by cycling the key.
Anyone ever see this???

PieRat

I had a problem with car doing similar but it was when cruising. Essentially something would trip the ecu then all my fuel trims woild go to 25% at idle. . I had to rwstart the car then all would be fine for awhile. At first it was a long time between eventz and then just slowly happened more and more. Took it to dealership they found intake leak(metal hose to surge tank loose). Basically i think what was happening is the car was getting invalid inputs overtime so it tried to switch to a backup map that wasnt tuned for the mods i had..

maybe your having a similar issue?

SMed08

What fail safe is this? Interested.

SICAMG

AEM FAIL SAFE.

SICAMG

Ok so it happened again. Scanned for codes and see a - B28 - Implausible Pressure Sensor signal. Code is twice and problem happened twice.
Again I can correct it if I just cycle the key........ then who know's when it will happen again.
Already put a new Mercedes Map sensor in a while back.
Anyone see this before?
Also wondering if my Aeroforce scan gauge is doing anything to effect it. Has locked up on me a few times displaying boost,ait and timing.

ajm55

PieRat

Booost leak IMO. Do you have stock TB?

SICAMG

Stock throttle body. No not a boost leak. i would think a boost leak would just show a lower boost level and not go lean like that.Plus it has been a year and a few race's since this started to happen. Everything seems to register on DAS correctly and will not trip a light.

drothgeb

x2 the cars are speed density controlled. A boost leak would just show the MAP lower boost pressure and it would drive accordingly.

Have you spliced into your MAP wires at all or are you any where around 17 PSI? I hunt for a dirty MAP connection, bad splice, bad wire, etc that might rarely give you an open or short or something? I found this random blog for a turboed E63. Different car but same principles. All of the trouble shooting starts with bad sensor, then tracing out signals/wires etc it looks like. Written kind of poorly but maybe you can find a good idea in there.

https://mercedes-benz-worldzz.blogsp...ble-value.html

Also, you could make a test rig using a ball or bike pump and a pressure gauge, pull the MAP, use like 1/4 or 3/8" tube to connect the pump to the map and pressurize it up. See if the displayed pressure matches the test rig pressure(I think you can probably do that with the car key in the on position but car off?). I did something similar to set my $20 pressure switch for my methanol system.

SICAMG

Hey DJ
No there are no splices and the sensor was changed last year.
No 17 lb limit at all. As soon as it happens all you have to do is tip in to the throttle ...it's lean. It doe's not see boost at all but drives perfect.
Also , for what ever reason I am thinking the OBD gauge may play a role some how because it has locked up and the parameter's get stuck and do nothing and you have to unplug it and plug back in.
Now that I am thinking outside the box with this the next time it happens I will try and log boost and see if it registers. If it doe's the map is talking but the ECU is not listening....doe's that make sense???

drothgeb

Yeah, I hear ya. That's a bummer. Has the scan gauge freaked out every time you get the lean condition?

Manifolds are at the shop. Should have them back next week just in time for track night on friday. Wont have changed anything else after all so I'll have back to back trips to the track with no other changes, as well as logs. I'm the guinea pig haha. Also after taking a close look inside the manifolds that extrude hone might do some real good. Those things are ROUGH.

SICAMG

They are rough and notice the tight squish to make the bend.The cheap way would be cut them open port them than weld close.But for perfection ....Extrude Hone. 800.00 bucks though !!!!!!!!!!!!!!!!!!!!!!!!!!!!

E55amg0220

Hey brother this is the same thing i experienced when i had the 82mm tb on my car .... i mean EXACTLY! I know you say you have a stock tb but go read my threads on the 82mm problems they are the same as what your experiencing. Sadly i feel this weird backup map needs to be found an addressed but the usa guys since over in europe they tell us about this map and dont havr any of these problems.

PieRat

I had 82mm hiccup like symptoms it was vacc/boost leak from stoopid loose metal vaccum hose that goes over valve cover..

From what i understand...
The car calculates throttle position, map iat, fuel and o2 and if they dont add up over a period of time it trys a backup map. In backup map mode my fuel trims would be +25 +25 and car would shke lot at idle and jerky on ingear deceleration. If i went closed loop (full throttle) it was fine.. Turning it off and on would work til i triggered the backup again.. if your sure u have no intake leak ur TB might need recalibration or replace or any of those sensors or maybe injector IMO.

SICAMG

All right I am going to go over everything Saturday. It is just strange that almost 9 months have gone by since it did it again and now twice in a row.
Soon the engine comes out for some work but would like this problem nipped. I will update Tomorrow.

sjc246

Sounds like you're over-boosting. I had the same problem with a pulley combo that was a bit too aggressive. Once the ME sees over a certain voltage from the MAP sensor (something like 4.6 to 4.7V I think) it assumes that the sensor is bad (as this value should never be seen with anything near stock boost levels) and completely ignores the sensor, calculating fuel injection based on throttle position and pre-definited maps (which assume stock boost levels). When it happened to me, I would see AFR exceed the range of my wideband (over 18:1) for a split second before I let off (no failsafe, so I'm pretty lucky). Even if your constant boost isn't over 17psi, it's still possible to have spikes that high for a fraction of a second, and you won't see that with OBD 2 logging or even on a boost gauge. If you data-log (with a high enough resolution) from the stock MAP sensor you should see it, or if you happen to have an oscilloscope, you can hook that up to the MAP sensor, temporarily. If you do it consecutively in two drive cycles, it will log a fault for "MAP sensor short to positive" or something like that, but if it just happens occasionally, you won't see any faults. Assuming it's just spikes, you can safely eliminate the problem using a voltage clamp to cut off the spikes at 4.5V (google split second VC3-100). Some people argue that (with a good tune) you can even use the MAP clamp to run higher boost (over 17 psi); I wouldn't trust that, myself, but using it just to cut off spikes should be perfectly safe.

SICAMG

This is interesting info as I normally only see a little over 15 lb spikes...that I can see.. The odd thing is how random it is. I will check with my tuner about this first before the clamp. Thanks for the info.....I'm on it !!!!

If I can not get a tune sent to correct it I will find out exactly what the maximum voltage is and clamp it at that. Just looked up the sensor you mentioned. Looks simple enough!

sjc246

I was having the problem nearly every WOT run with my 77mm upper (combined with a larger crank pulley). With the stock upper pulley it never happened (around 14 psi peak). With your 80mm pulley (thanks again, by the way), it has only happened twice and only in third gear (although I have subsequently been avoiding full throttle in 3rd until I have it fixed). I'm seeing about 15-15.5 psi (continuous) in second gear, but obviously it is spiking much higher to cause the problem. I just bought the MAP clamp I mentioned and once I install it (hopefully tomorrow), I'll try to get a scope trace of the MAP voltage before and after the clamp (my logging resolution is too low to catch the spikes, I think). I fully expect that the voltage clamp will fix my problem.

Assuming that this is caused by boost spikes (which I am fairly certain of), a tuner would need to change the fault threshold (or remove the monitoring for that specific fault entirely) to correct it. While this is theoretically possible, I seriously doubt that there are any tuners (in the US, at least) that would attempt it (or even know where to start). Even if I found someone willing to try, I wouldn't want my motor to be the guinea pig for it.

There may be other options (besides Split Second) but I couldn't find much else (you want something that doesn't alter the MAP signal at all until it exceeds 4.5V). I haven't tried it yet, so I can't say for sure how well it works, but I was very impressed with their shipping speed (I ordered the morning of May 1st with the cheapest shipping option and it arrived on May 3rd from CA to NY). If continuous boost was actually reaching (or exceeding) the limit, you would probably be better off with a piggy-back to slightly decrease the MAP signal across a range as it neared peak. This would also allow you to do your own AFR tuning (once you had a decent baseline dyno tune), however, that's a lot more work and a lot more risk than is necessary if you're not actually exceeding the limit continuously (and much more expensive, too).

SICAMG

So just so I understand. In my case I take the car for a test run with a buddy of mine ,just to monitor some stuff and car runs like a champ. A seperate gauge reads 16 lb's of boost but the OBD shows 15. Just saying.
Anyhow I park the car for a few hours and than leave to go home. As soon as I tip into boost on the highway,I realize there is a problem,lean out, and never even got into it hard at all.
Stop, shut engine off,restart fixed. So is this how it happens or does the car have to be running,see over map conditions than react ?
I still find my deal a little strange. Also it never did it with that 80 mm either but could be a fluke.

By the way what was the power lose with the 80mm over the 77. Big difference?

SICAMG

Any updates on the Map Clamp?

sjc246

So far so good. One of these days I want to put a scope on it so I can actually see the difference (before and after the clamp) but it definitely seems to be working.

SICAMG

So you just wired this in and set the voltage "out" configuration. How did you monitor that...with Stars ?

Also can you give me a good example as to how your problem would manifest? Was it while driving under boost, than limp mode lean, or the next time out it did it?

Thanks

sjc246

Temporarily connect the 5 volt reference wire to the signal input and adjust the output is whatever you want it to be (I set it at 4.5V). Use a DMM to measure the voltage to set it.

My problem would happen whenever the boost signal spiked too high (with the latest pulley, that was only occasionally in 3rd gear, with more aggressive combos, it would happen in 2nd more easily). Once the boost went too high, once, it would run in a sort of limp-home mode (no MAP sensor) until I cycled the key.

SICAMG

Cool . I will order the clamp and look for the wire schematic for the wires and hook it up.

So just to be clear this clamp is before the sensor on the input ?? I would think it would be the output. Not totally getting the concept .

sjc246

It's wired between the MAP sensor and the ME. It takes the 5V reference supply and ground for its own power/ground supply and intercepts the signal (you just tap;into the power and ground wires, but you need to cut the signal wire). The wire out from the MAP sensor is used as the input for the clamp and the output from the clamp is connected to the ME. The signal coming out is the same as what goes in until the signal going in exceeds the set voltage (at which point, the output signal remains at that pre-set maximum voltage). In theory, you can install it anywhere between the ME and the MAP sensor, but seeing as the ME is in a nice weather-resistant enclosure with easy access to the wiring and plenty of extra space, it's pretty clear where you should install it. Make sure you understand what you're doing before you install it: if it's not installed correctly, it could very easily cause major problems.