Mercedes AMG M113, 5L V8 without compressor from factory (not M113K) but compressor DIY-added with lot of mech-work.
Same set of sensors as the V6 M112 engine (just in case it is easier to find docs for that).
V8 engine 15426378 firing order (CL500 or similar chassis)
[M113 engine] 3 valves and dual sparkplugs per cyl (16 sparkplugs total), no VVTI
Engine started immediately after sorting out the mess of ignition transformer connections:
- they were different on paper documentation and real life, no surprise it did not want to start
- on top of that, EC36/31 low current p259 output was connected instead of EC36/33 IGBT to drive 2 igncoils, luckily it did not blow up (nothing was configured on that p259 output).
Ign outputs ... - will be overloaded this way
However, because 1-6 and 6-1 are driven in pairs, 4 igncoils from 2 outputs with opponent cyl hard-wired
- currently 16 ignition transformers can only be driven per crankrot (ign-dualout), equivalent to 32 passive igncoils driven from the 8 IGBTs inside VEMS
- on top of that, VEMS ECU is installed in the engine bay.
- This is outside the safe envelope, it is impossible to guarantee proper cooling and operation while driving 16 ignition transformers at double frequency (equivalent to 32) from the engine bay...
15426378 recommended output connection:
- cyl1 EC36/35 ignch0
- cyl6 EC36/33 ignch1
- cyl5 EC36/34 ignch2
- cyl3 EC36/36 ignch3
- cyl4 EC36/11 ignch4
- cyl7 EC36/12 ignch5
- cyl2 EC36/24 ignch6
- cyl8 EC36/10 ignch7
Mercedes ignition coils have low inductance, usually low dwell length is appropriate for 120 mJ spark energy: Around 1.8 .. 1.9 or max 2.0 msec. NOTE: 1 '''shared fuse for 8 ignition outputs and 3.19 msec dwell (if I understand right) is asking for trouble. PCB trace or IGBT or ignition transformer will likely be damaged.
- IMPORTANT: 3.0 msec is WAY too much for most Mercedes ignition transformers, 23 A was measured by someone: 14 * 23 / 2 * 0.003 = 0.483 J that is 483 mJ or almost 4 times the 120 mJ (not exactly because of losses).
- Assuming engine is running normally (and tuned at the point of interest) decrease dwell and when misfire happens (eg. at 1.8 msec) add +10% (1.8 + 0.18 = 1.98 msec ~ 2 msec) dwell (which is appr +22% spark energy).
Had to do some investigation (or hint, eg pinout) regarding M113 sectrig HALL sensor
- engine harness schematic might also cast some light on this?
- although both GND and output seems to be connected to ECU (camHALL GND not connected to other GND-s in the harness, only inside the ECU)
Primary trigger VR (1072 Ohm)
- Factory trigger seems to be "inverted",
- swapped wires so that falling edges are correct at missing tooth
- verified (slowly rising in tooth-gap)
- [scopeshot here] is good (wire swapping not needed if the same signal), may or may not be applicable...
- Factory ECU pin37, (pin38) are the crankshaft VR input
Primary trigger VR signal is really trash
- some teeth amplitude are 1.6 times greater.
- interestingly, not even the tooth right after missing tooth, but 4th tooth was bigger... Greater in positive direction, even greater going negative.
- most likely interference from ignition (?)
- need to try with ignition disabled, but battery was almost drained, had to stop.
- Most likely Shielded cable is needed. Shield connected at ECU side only, and GND and VR-signal wires inside connected on both sides.
- VR modification: either primtrig LM1815 pin5=5V to disable adaptive threshold, or R181=270k or similar
- signal amplitude is otherwise OK, but high voltage noise is unknown.
- Perhaps a 1k pulldown could be useful at the same time if signal amplitude allows...
Secondary trigger: 1 pulse per camrot Bosch 0.232.103.027 - sensor had contact problem at male receptacles (oxide? even with new heatshrinked receptacle it did not work first
- Verified HALL sensor for sure (with +12V supply, 30..40mA supply current)
- Pulls to GND (with pullup resistor in ECU)
pinout (Verified: pin1,2,3 are: GND, HALL output pink, +12V red-blue)
- pin1: GND from Bosch ECU/pin39 (low current connector) brown-green
- pin2: output pink = Bosch ECU/pin 40 (low current connector)
- pin3: +12V (in factory harness same net as injector shared +12V) red-blue
Similar bosch camshaft Hall sensor's datasheet: [0232.103.037] - All "0 232 103 XXX" sensors seem to bee camshaft sensors.
Big mistery with secondary trigger,
- Ours did not react to hammer moved near and away (when powered +12V, of course)
- Apparently same sensor as in [this m112 video] (and even same pins 39,40 in the factory engine harness, although the factory harness is not actually used), but our sensor does NOT react.
- IN THE VIDEO, middle pin seems to be signal output as assumed (he attached probes reversed in DVM, never mind)... 12V when no iron, and 0V when ferromagnetic tool is close
DBW
- currently factory ECU drives it, but VEMS is prepared (with bridgeoutput driver)
Ignition
- 16 sparkplugs (dual sparkplugs per cyl)
- only 8 passive ignition transformers
MembersPage/MercedesAmgM113/OilTemperature
Engine oil
- Motul 5W40 229.5 oil was perfect (229.3 was not good). Oil temperature sensor was very important and useful.
- Mobil 1 10W60 the engine suffered main bearing failure in very short time... Maybe coincidence but was not used again
- Castrol Edge A/B 10W40 worked, but oil overheated
MembersPage/MercedesAmgM113/LcdPages -wish
URLS
- MembersPage/PeepPaadam/MercedesAMGSL - very similar
- MembersPage/MercedesSelFivePointSix
- MembersPage/AmgSlovakia much more sophisticated
- MembersPage/BmwWithMercedesEngine