Cylinder Deactivation

[Mudgey]

Hey guys, I have been doing some researching about cylinder deactivation, displacement on demand or AFM... however you want to name it.

Engine: Honda K20a with VTEC and VCT on intake cam (iVTEC), stock form they are 200bhp, and with my supercharger it is 305bhp.

My quest is to get as best FE as I can out of the car, they do a 2ltr version of this engine in the accord that gets just under 50mpg (UK) and I am considering these mods:

1. Regrind of cams to same spec as economy cam, but keeping the K20's agressive VTEC lobe for high rpm fun. The economy cam essentially runs the engine in 12v mode with the other valve opening very minutely to stop fuel puddling. This creates swirl and concentrates the fuel droplets towards the spark plug making it richer there but overall it allows the car to run at AFR's as high as 20:1.

2. Run displacement on demand by regrinding the VTEC cams so they are flat, this would allow me to run only 2 cylinders at highway speeds, unfortunately I will lose top end torque, but this could be made up for by using a big rotrex blower with its massive ramp up of boost to help overcome the restriction at the top head.

3. Combine the above two with the reground economy cam, but use a custom made butterfly valve just infront of the injectors on the cylinders I wish to close off, by now allowing any intake charge into the engine I should still be able to take advantage of the 'spring' effect that GM use on their V8's.

Has anybody tested anything similar? What are your thoughts on the above?

[H-Man]

I say give the ability to over ride the vtec so that you can disengage the cylinders at any speed and reengage them when needed, and have a relay shut off the injectors on the deactivated cylinders.
I have never messed with honda engines, but your idea sounds like it could work.

[t vago]

Is there any sort of VTEC function for the exhaust valves? If so, then this idea might could work.

[Mudgey]

There is two profiles on both the intake and exhaust, this is the same for atleast the performance vtec engines, some just have vtec on the intake cam.

If I went down the dactivation method I would only be able to use one cam profile for the whole operating RPM of the engine of 8800rpm, of course the reason why honda went with VTEC is that they get utilise two ideal profiles for such a vast rev range. If I went with one that favours upper rpm flow then low rpm economy would suffer, if I went with one that favours low rpm torque then the engine would feel like it runs out of puff at the higher rev range, the answer then would be to go with the economy cam profile and run huge amounts of boost, possibly as high as 20-25 PSI from a rotrex unit to give a more linear torque curve and retain my 300-350bhp

Audi run their dactivation system by trapping exhaust gases in the two non operating cylinders, this overcomes the pumping losses in the now 'dead' cylinders but they have a precise system that can operate quick enough and at a specific time to trap the gases, if you ran a vacuum you would, over time, draw oil and gases up from the crankcase which is not a good idea! How I could specifically operate the VTEC at the required moment is still up for debate....

[t vago]

Originally Posted by Mudgey

Audi run their dactivation system by trapping exhaust gases in the two non operating cylinders, this overcomes the pumping losses in the now 'dead' cylinders but they have a precise system that can operate quick enough and at a specific time to trap the gases,

That's how all true cylinder deactivation systems work. Examples of US automakers' entries include Chrysler's Multiple Displacement System and Government Motors' Active Fuel Management (or Displacement On Demand).

Originally Posted by Mudgey

if you ran a vacuum you would, over time, draw oil and gases up from the crankcase which is not a good idea!

My understanding is that the valves on the given cylinder are deactivated during the compression stroke. The final air/fuel mix is combusted as usual, but is not pushed out via the exhaust stroke. This would solve that vacuum induction issue.

Originally Posted by Mudgey

How I could specifically operate the VTEC at the required moment is still up for debate....

Conduct research, and experiment. That's all I can recommend, having not very much practical experience with VTEC.