Let's clear some things up here so we can all be in the same mod of thinking.
First let's agree on why engine fire BTDC.
1. As has been said, as the engine picks up RPM more timing advancement is needed because combustion is not instantaneous. The time required to combust at 2000RPM is that same as 10,000 RPM.
First off what I have discussed sofar has nothing to do with ignition timing. We will need advance, we just don't know how much, and it is totally irrelivent in my design. At this point I don't really care what the end timing is going to be, I'm just trying to figure out if the design is solid.
2. By starting the combustion process early, we get an increase in power. Image this. You have a door and I have placed a device on it that measure how much force you open the door with. First I have you open the door as hard as you can and you produce X amount of force. I then push on the other end of the door as ask you to open as hard as you can. You push as hard as you can, but the door doesn't give because I am preventing it from opening. Then I let go of the door instantly which allows the door to open a result in Y amount of force being produced. Which is higher in force, X or Y? Y, because Y had Z amount of potential energy (PE). Y is basically equal to the X force (because both times you opened the door as hard as you could) plus the addition of the Z (the PE force). Make since?
Now, here is why I see this floating cylinder not working. If you had your whole power stroke in 90 degree, you have less of a degree rotation in which to transfer the power produced from the combustion. Once the piston starts to move up you have stopped the power stroke. In our engines, we have 180 degrees of rotation to transfer the power from the combustion into mechanical rotation. If you start the power stroke at a horizontal position, that means you would have to start it 90 degrees later then it does now. This means that you have only 90 degrees to transfer the combustion force. This might work at low RPMs, but will not work at higher RPMs.
Don't think of the power stroke in degree's think of it in inches. Does a shorter stroke engine suffer from it's shorter stroke? Or is it just a smaller engine.
Now the two piston per cylinder which has been brought up has no real advantage in my mind either for car applications. First, understand that even though you have two pistons per cylinder, you are exerting the same amount of force on them. You take X amount of cubic inces of air/fuel mix and you will produce on Y amount of force. The two piston design is just applying that force in a different way, not a better one. Also, this will then cause you to have ½ the cylinders WHICH IS A BAD THING! If you had a 302 CID V8 and a 302 CID inline 4, which would be better? Well, the inline 4 might have less parts and less friction, but that is not worth the vast increase in engine vibration that you would have. More cylinders results in smoother engine running.
He is what I think is the other reason. Car companies do not want to have to go through the work of designing a radically new engine block. Back in the 70’s, Sandia National Laboratories designed a variable stroke engine (in other words, variable displacement). The engine was EXTREMLY fuel efficient but was not produced because it was too hard to produce at the time, and car companies did not want to spend the money to re-tool factories for the block.
And on the other hand, Mazda used the Wankle