Killing fuel/fire to a few cylinders ?
 

I have done some lurking and have seen many posts on this subject. I don't think what I am considering has been addressed before.

I am looking to start a high MPG project and originally was going to use a Geo Metro as a test mule. A cheap/good one is too hard to find so I am going to experiment with one of my Miatas. The Miata currently has a supercharger and gets 20 MPG (horrible). I pulled the SC last night to return it to natural aspiration. The car's stock ECU has been replaced with a fully programmable Link ECU. I can control the injectors, timing, etc on a grid that is looking at the engines vacuum (and boost if the SC or turbo was on there).

So, what I am considering for one of my experiments is to get the car (I4 1.6L engine) up to around 55 MPH and then kill the fuel (and possibly spark) on one or two of the cylinders. I do not want to mess with the valves, pull the pistons, or anything major like that. I just want to install a $5 high amp switch and kill the power to one/two of the injectors. The car comes with 2 coils and a wasted spark system (fires plugs on two cylinders at the same time regardless of which has fuel in it).


Anyone have any comments, suggestions, etc.




Robert McElwee and Red Beast
1991 T25 Turbo @ 15 PSI
Link ECU, FM IC, 9:1 pistons
Over 400 lbs of "added lightness"
www.lightweightmiata.com

Lightweight Miata Forum:
www.lightweightmiata.com/forum

The Miata Trailer Project:
www.lightweightmiata.com/trailer/trailer.htm

I'm probably not the best to answer this question fully so someone who has researched/tried this can correct or add anything that else that needs to be said. The main problem that people run into with doing this that you will be pumping the 4 cyl worth of air and only 2 cyl of fuel past the o2 sensor(s) and the computer will read lean and will compensate by dumping extra fuel. Having an aftermarket computer, if you can correct the A/F through your stand alone or piggyback, then you may have a fighting chance at this. The issue here is that you how will you know if you are actually running lean during this because of the false A/F reading you will be getting? Getting up to speed with 4 and then cruising at 2 may be viable if you can work that out.

You can't correct the AFR to compensate for dead cylinders. The mixture will ALWAYS read lean no matter how rich the other two are running. You have to run in open loop mode while the cylinders are disabled.

I knew that O2 on a "regular" car would be the problem to overcome so I pointed out that my car had the standalone ECU but let me address it again in more detail:

Since my car has a stand alone ECU where I can control every aspect, all problems concerning O2 sensors are eliminated. In fact, I tuned the car a year ago and have not even had an O2 sensor installed in it since that time. I tell the car how much fuel to dump in, it doesn't decide on its own how much to dump. If I am running with an OK mixture on 4 cylinders (as read by a WBO2 sensor) then killing fuel to 2 cylinders will still give me the correct mixture in the other two.

So, what else do I need to consider for this test? Comments/suggestions?

Then it's just a matter of a few relays and you're set.