VW quantifies MPG gain of 4- to 2-cylinder deactivation at 31 mph = +16 mpg

[t vago]

Originally Posted by pete c

My question was how much of a load the deactivated pistons are. I realize the airspring idea means they are a small load, but, they are a load, none the less.

As long as the deactivated cylinders are not pumping air, they should prove to be a negligible load. Piston ring blowby, and friction losses, will undoubtedly make the load greater than zero, but it's going to be very small.

[pete c]

I guess there isn't enough of a load to justify the expense of an engine that would completely shut down deactivated cylinders

[Occasionally6]

Within the wider context of total engine losses, not much. If you have ever rotated the crankshaft (by hand) on an engine with the head removed you'll know it's a pretty easy thing to do.

Back in the late 70's/early 80's there was a lost motion rocker system, implemented in the Cadillac V8-6-4. That was tested on a few engines with a number of different cylinders being deactivated. Apparently, by the contemporary standards, engine vibration (and noise) wasn't a huge issue with them. These days, I assume that is more of a problem. For example, the cylinder deactivation in the Honda Accord is accompanied by active engine mounts.

What would be interesting is linking the cylinder shutdown with an electric motor connected to crankshaft, IMA style. The motor would be used as an electric "flywheel", buffering the wider spaced or uneven firing impulses. (The IMA motor is already used in this way; the cylinder deactivation is the additional part.)