Quote:
Originally Posted by AngelSoulll
For the crank vent system, my only concern is the possibility of the water meth mix going to the crankcase from the small tube going from the intake piping to the valve cover. As you may know that when the car is idling or throttle is partially pressed, that little tube acts as a fresh air feed to the crankcase to make up for the vacuum made by the intake manifold. When in WOT, the same port acts as a vent due to the low vacuum in the manifold cannot vent out the pressure inside the crankcase. So the flow of the fresh air going to the throttle body helps pulling the vented gasses coming from the valve cover though that tube.
Please correct me if I'm wrong. Just need to make sure if I'm on the right track here.
|
It won't spray at idle and you shouldn't need it at part throttle. Check valves on each WI injector will prevent it being pulled under vacuum.
Think of them as an extra set of fuel injectors -- except it would be a mix of water and alcohol. Also, you will want injectors that atomize the mix as much as possible -- even fuel can cause wetting and pool inside the IM if the atomization is poor (and especially when running cold).
But, this is my whole point: If you need WI at 0 boost (i.e., atmospheric pressure) and part throttle/low load, but NOT under high load, WOT or near-WOT, you should play with the existing tune first and not invest the money, time and effort in adding an entire secondary fueling/cooling system...
Moreover, high humidity will actually quench det -- it's water, highly atomized, and it displaces air... high pressure, high oxygen, high heat (which also increases pressure) are the potential problems -- but if you are knocking at part throttle and not at WOT, its a weird tuning foible that can be addressed with your existing tuning software.
I'm telling you, the problem is either the tune (and if so, likely a problem on throttle tip-in, due either to too much spark advance, less than ideal valve timing or overlap, or a lean condition, causing a miss) or you aren't knocking.
Datalog, see what is happening, under what conditions and where and use that as your guide.