Differential Supercharger

Superchargers are mechanical air pumps allowing an internal combustion engine to be more powerful by force-feeding said engine more
air than it can inhale on its own.

Upside is instant boost…downside is fixed boost:no real way to tailor the precise level of boost needed at that particular moment.

(Setting aside turbos and Comprex for this discussion.)

Could a differential gear be added to the drive between the engine and transmission?

Thinking about the gears in an open-differential axle got me thinking: imagine one wheel on a pathch of wet ice and the other wheel on dry asphalt. Due to less resistance on the slipping wheel it spins freely, leaving the other wheel inert and not moving.

Now imagine a differential between the engine and transmission. At cruise, torque flows from the engine to the transmission, with minimal torque going to the blower. But under heavy acceleration or increased load due to the wheels going through thick mud or something, engine torque encounters load at the transmission and spills extra torque to the blower, instantly producing more power and torque to overcome the load.
Does such a supercharger system exist?

The differential drive mechanism you describe splits movement, not torque; its behavior is described in terms of the RPMs of its input/output shafts, and torque doesn’t figure into that.

As such, the sort of device you envision for providing an adjustable amount of shaft power to the supercharger is unworkable.

What might work instead is a continuously variable transmission, which can adjust the drive ratio of the supercharger, running it at low RPM when little or no boost is required and jacking the RPM up when boost is called for. However, one of the benefits of superchargers (as opposed to turbochargers) is the lack of lag; I don’t know how fast CVTs can shift, so if they’re slow, the lag might become a problem.

FWIW, installing the throttle body upstream of the supercharger does a pretty good job of controlling manifold boost.

it would not be controllable, AFAICT. the load applied to the engine by the transmission would swamp the supercharger’s load and the car would just sit there while the blower spins. Think of a car with an open diff on a slick surface. once one wheel starts to spin, the other sits there and does nothing and all motive power is sent to the slipping wheel. As Machine Elf said, they’re not really torque dividers; they just provide a mechanism for drive wheels to turn at different speeds.

what might work is something akin to the power split device of the Aisin-style eCVT in the Prius and earlier Ford hybrids.

in reality, you’d just use turbochargers which already are demand-based in how they deliver boost.

Variable-speed superchargers go back to at least the Merlin engine (as used in the Spitfire) in 1938.


It had a two-speed gearbox.

I believe this is what a blow-off valve or other method to reduce intake plenum pressure would be used for. Turbos are going to out-sync with engine demand as you shift, brake, accelerate, etc. They design for what the engine needs and the type of driving. I.e. on a circle track at a near constant load you can probably design for a narrow margin, where as a daily-driver in stop-and-go traffic would have a much wider margin.

For some reason I wanted to believe that for some time there was a variable or multi-speed supercharger that was commercially available. Googling I found some recent stuff though which is akin to what you guys are talking about:

But, I imagine people have thought of similar approaches, and when you look at the complexity and cost vs the benefits of a variable drive system, releasing the over-pressure is an easy and effective approach.

The OP poses an interesting question.
It could be managed as described, however: Transmissions take care of this situation with a far less sophisticated techno-mechanical solution, as suggested by Machine Elf.

Somebody was thinking about this down under a few years ago. Check this out at 1:03.

Yes, it’s complete fantasy.:wink:

Fantasy is a good description.

Found it!

Popular Science article in Aug '65