S65 AMG misfire
#51
Super Member
Thread Starter
I use standard nylon hard lines. They hold up well, although I cover the ones in the engine bay with high temp insulation.
#53
#54
Super Member
Thread Starter
Oh blimey, didnt realise that pump you linked at runs at 1000psi!!
Not sure if there is any benefit to the higher pressures when we run at the single nozzle in front of the throttle body. most of the meth injection is going to be carried to the middle cylinders, i'm sure of it.
Not sure if there is any benefit to the higher pressures when we run at the single nozzle in front of the throttle body. most of the meth injection is going to be carried to the middle cylinders, i'm sure of it.
#55
Well when you get the droplet size below about 40 microns it becomes a "dry fog". Where the droplets are small enough they will bounce off of most surfaces instead of wetting them, so the fog is more likely to stay in the airpath and follow the air flow/be more evenly distributed. Any water/meth that ends up on the port walls or intake manifold walls/floor does not contribute to cooling the charge, but still does contribute to diluting it volumetrically (and due to that has sort of an EGR-like effect and still does help with knock but not as much as when it also is cooling).
Beyond that, droplets half the size have 8x the surface area for heat transfer so it becomes easier to vaporize a larger percentage of the droplets. I kind of went down this rabbit hole when I was having the misfires with the water/meth spraying, so I ended up building that spraybar down the center of my intake manifold using small garden misting nozzles...which have a much smaller droplet size, but require high pressures to get adequate flow. Generally speaking the smaller the nozzle you go with the finer the mist will be, and on a given nozzle the higher pressure you run the finer the mist will be.
Beyond that, droplets half the size have 8x the surface area for heat transfer so it becomes easier to vaporize a larger percentage of the droplets. I kind of went down this rabbit hole when I was having the misfires with the water/meth spraying, so I ended up building that spraybar down the center of my intake manifold using small garden misting nozzles...which have a much smaller droplet size, but require high pressures to get adequate flow. Generally speaking the smaller the nozzle you go with the finer the mist will be, and on a given nozzle the higher pressure you run the finer the mist will be.
#56
Super Member
Thread Starter
I see. Does that mean you had no luck with spraying through the throttle body.
It does not help that we have a such an odd coil system.
It does not help that we have a such an odd coil system.
Last edited by alexanderfoti; 05-11-2017 at 03:43 PM.
#57
I really wish there was more clear information available about the coils. It's a very low energy system...I believe I've seen documents referring to the transformer only putting out like 55 watts per side. So they are running 12 coils on something like 5 amps of input current. My Buick runs 6 amps for half that many firings. I can only assume that with such high primary voltage they are concerned more with peak voltage capability as opposed to spark energy/burn duration. So it's more of a capacitive style system which I've never been a fan of. I really would love to re-engineer that, but the cost of doing so even with twin tower coils is still pretty high and would still be stuck with the same coil driver setup. Standalone/aftermarket is really the only option that makes sense and at that point you might have to do something on the fueling side as well and eliminate a lot of the factory features as to not **** off the computer and end up in some sort of limp mode.
#58
Super Member
Thread Starter
Aha right yes that makes sense. A dense water/meth fog will put the coils under stress for sure.
I have read that too, 55 watts per bank, in WIS.
I have been reading up on them and am pretty confident I can replace the mosfets and tubes should there be an issue on mine. It would appear getting into them is the hardest part.
It would be rather nice to replace them with the M279 configuration. I wonder if the M279 ecu has a different part number to the M275. It probably does.
I have read that too, 55 watts per bank, in WIS.
I have been reading up on them and am pretty confident I can replace the mosfets and tubes should there be an issue on mine. It would appear getting into them is the hardest part.
It would be rather nice to replace them with the M279 configuration. I wonder if the M279 ecu has a different part number to the M275. It probably does.
#59
I have all the diagrams bundled together somewhere on this computer but can't find them right now, but if I recall correctly I remember thinking the coils were in fact triggered directly by the ECU, not from within the module. So it would be a pretty simple matter replacing them with virtually any "smart coil" (with built-in ignitors) however, I don't think the computer would be happy about no longer having the ion current sensing. To retain that feature, you'd have to still run any replacement coils from the stock module, though you could probably upgrade the drivers to larger IGBTs. But that's somewhat pointless if you're still dependent on a single stock transformer.
#60
Super Member
Thread Starter
Have a look at this
Rather interesting!
Does the m275 still use ionic current sensing for knock detection? It has 4 traditional knock sensors so I am not sure.
Rather interesting!
Does the m275 still use ionic current sensing for knock detection? It has 4 traditional knock sensors so I am not sure.
#61
Have a look at this https://www.youtube.com/watch?v=bfSIr-48rZU
Rather interesting!
Does the m275 still use ionic current sensing for knock detection? It has 4 traditional knock sensors so I am not sure.
Rather interesting!
Does the m275 still use ionic current sensing for knock detection? It has 4 traditional knock sensors so I am not sure.
#62
That video confirms what I was thinking though. It would be super easy to run aftermarket coils since the trigger wires are external and easily accessible. I just don't know how the computer would react in terms of misfire monitoring.
#64
Watchthing the video some more, your problem is your boost control solenoid. On my tuned s600 it will to 95% i just replaced my airbox sensors and I'm getting 19psi. I would replace you boost control solenoid. Its under the driver intercooler, it has a line from the passenger side intercooler that runs to it.
Last edited by g60wall; 05-11-2017 at 05:40 PM.
#65
Super Member
Thread Starter
Watchthing the video some more, your problem is your boost control solenoid. On my tuned s600 it will to 95% i just replaced my airbag sensors and I'm getting 19psi. I would replace you boost control solenoid. Its under the driver intercooler, it has a line from the passenger side intercooler that runs to it.
The ecu controls boost by changing the boost control solenoid duty cycle. When it wants less boost it actuates it a lower duty cycle.
If I attached a "bleeding" boost controller to the air line between the IC and the control solenoid and bleed boost manually, the duty cycle adapts to control to 15 psi. If I open up the valve massively, it adapts.
It is actively trying to regulate to 15psi.
All 4 maps sensors are new.
#67
Super Member
Thread Starter
#70
Super Member
Thread Starter
#72
Well if you're getting knock retard on the spark side of things I would be shocked if you didn't also have the computer reducing boost to combat that as well. I am sure that's what is going on.
The following users liked this post:
alexanderfoti (05-12-2017)
#74
Super Member
Thread Starter
I have never changed them, in the 5000 miles I have had the car for.
I need to look through the previous service records, but I think they were changed 50,000 miles ago.
When you say re tune, do you mean remap?
I need to look through the previous service records, but I think they were changed 50,000 miles ago.
When you say re tune, do you mean remap?
#75
Have a look at this https://www.youtube.com/watch?v=bfSIr-48rZU
Rather interesting!
Does the m275 still use ionic current sensing for knock detection? It has 4 traditional knock sensors so I am not sure.
Rather interesting!
Does the m275 still use ionic current sensing for knock detection? It has 4 traditional knock sensors so I am not sure.