The problem with hyper expansion is that it requires to increase the size of the engine for a given output power, on the other side increasing the compression ratio allows to increase efficiency without requiring over-sizing. So it is doubtful that this engine will do better in term of efficiency than a turbo charged downsized engine with direct injection and that would be smaller...

But Roger Pham can certainly explain this better than me

I enjoy mechanical engineering experiments; this one is especially interesting. However, the golden age of the ICE is now fast becoming a past era as battery powered electric motors are fast becoming the future of transportation.

One interesting possibility with split cycle engines is that they allow cooling the charge after compression so that combustion can be accomplished at much lower temperature than in conventional ICEs. This engine apparently does not take advantage of this opportunity. Likewise, it should be noted that asymmetric compression/expansion and turbocharging can be made relatively easily on a conventional ICE by using Miller system (utilizing the Atkinson cycle). This would be an aggressively downsized engine that also has high efficiency at high loads, which increase the potential for hybridization. Thus, the basis for comparison should perhaps be such an engine. Note that this kind of engine is not (yet) in production. The 1.0-liter Ford 3-cylinder engine could be seen as basis for this concept. It is aggressively downsized but it does not have the turbocharging available for employing a Miller system, implying a further development potential. Note that one feature of the Miller system is that it cools the charge to a lower level at top dead center than a conventional engine, with considerable advantage for thermal load; a known problem with aggressive downsizing. Split cycle engines could have similar advantages. I am looking forward to reading the SAE paper on this concept and, of course, compare with the Scuderi engine.

@Lad
If we just sit down for the next 100 years and wait for cheap and good batteries, we might see the golden age of battery cars. Those of us who want to buy cars in the meantime are happy that the development potential of ICEs is far from being exhausted and in addition, hybridization will take us one more step forward. Recall that both energy use and CO2 emissions in the 2030 US perspective will be lower for a gasoline hybrid than a battery car (as shown by Heywood et al. at MIT).

Why would you cool an air charge immediately after compression, then heat it by burning fuel in it?  Just burn less fuel, it's more efficient.  If you need a stoichiometric mixture, use a diluent.

Intercooling makes sense; it cuts the back-work of the next compression stage.  Cooling between compression and combustion does not.

What I'd like to see is a split-cycle engine with heat regeneration.  That would break efficiency records.

Peter XX,

'lower' is not good enough. We need to get to zero.

Three things here:

1. The compression ratio of this engine can be varied by changing the phase difference between the compressor and expander crankshafts.  The peak compression ratio can be limited as desired.
2. Differential compression vs. expansion ratios can be produced by late closure of the intake valve.  With current VVT systems, the change can be any value between the mechanical limits.
3. To make up for lower volumetric efficiency at higher differential expansions, turbocharging can make up for the reduced air charge.  Diesel turbos regularly operate at higher output pressures than the input exhaust gas pressure; this would just harness the energy better.  Hyperexpansion doesn't preclude turbocharging; any operating point which uses a turbo waste gate or equivalent could have used hyperexpansion instead to turn the dumped energy into output.

A turbocompounding system is an extra element.  This is not to say that it may not be worthwhile, but if it could be replaced by a hyper-expansion engine it raises the question of what purpose it serves.

But at higher cost for electric turbocompounding vs overexpansion engine, so Tour engine can be more advantageous if the exp:comp ratio can be optimized depending on the engine's mission, whether in an HEV vs PHEV vs non-hybrid. I would stay away from hyperexpansion on non-hybrids.

Peter_XX,

No, current electricity production does not give greater CO2 emissions than conventional fuels, where did you get that idea from?

Average fuel consumption of cars sold today is 24 mpg, resulting in well-to-wheel CO2-emissions of ~275 g/km. A 5 km/kWh electric vehicle on the US grid mix is about 120 g/km (assuming 600 g/kWh).

But then you can point to the Prius and say that it has half CO2 emissions, or ~130 g/km. Yes it is tempting to compare the EV to the most fuel efficient petrol car on the road, which represents 2% of sales, and forget about the other 98%. Another problem is that 'the EV' does not exist yet. There are less than a handful available at the moment. So any comparison is highly subjective.

Even if the announced 54.5 mpg CAFE standards for 2025 come true, then you'll have to assume that the us grid mix is still the same 13 years from now to argue that the EV is worse than a conventionally fueled car. This is highly unlikely. And even then, the petrol car wins by only a very slim margin.

On the individual level, many people link the purchase of an EV to the installation of a PV system to offset their extra electricity consumption, thereby tilting the playing field very much in favour of the EV.

The essence is this: depending on your assumptions, the EV today can be slightly worse than some petrol cars. Otoh there are ample circumstances, some under your control, some not, where it is significantly better.

Ok now, the point I was trying to make is that a petrol car can not get to 0 emissions. You can not keep increasing the mpg's forever. It will become exponentially more expensive to wring out yet more metres from a droplet of petrol. If you want to get to 0, you'll need something fundamentally different like an EV running on 100% [plug in your favourite clean technology here] electricity.

Finally, the most important argument for buying an EV now is that the technology is still in its infancy and needs to be developed before it can be mass marketed. No significant EV sales are to be expected before 2025. Sure, you can wait until 2030 or so, but you can't skip this early adopter phase and begin mass marketing out of nowhere. If you start in 2030 with the sale of EV's you're too late to 1) develop the technology and 2) ramp up mass production and 3) replace the existing fleet. That whole process takes at least thirty years.

@Engineer-Poet It is nice to see that you always have a different opinion than me.

If you have accepted the concept of intercooling, you should realize that an even colder charge is more beneficial from a thermodynamic viewpoint.

Intercooling and reheating will always decrease the thermal efficiency unless they are accompanied by regeneration. This is because intercooling decreases the average temperature at which heat is added, and reheating increases the average temperature at which heat is rejected. Therefore... intercooling and reheating are always used in conjunction with regeneration.

I see Anne polished off your other claim so I don't have to.

Peter XX doesn't merely forego quoting the relevant portions, he foregoes linking the report at all lest he open himself to criticism on the specifics.  This is handwaving carried to extremes.

Mr. XX, I'm well aware that turbocharging and intercooling increases the volumetric efficiency of an engine.  Whether this increase the thermal efficiency depends on the average vs. peak output, among other things.

you should take as homework to find out and explain to everybody at this site why intercooling increases the efficiency of an engine. After that, you should never question my comments in the future!

@Peter_XX,

Link the exact study that is the basis of your claim. Heywood has done more than 1.

I have shown you that the average car now emits twice as much as an EV. There's no way around that. And the recent Heywood studies I assumed might be the ones you're referring to do not challenge my figures in any way. And I am not interested in cherrypicked models (Prius-only scenario's). It's what people buy that counts, not what is available.

Oh, wait, I think I remember when he did link to that study.  Conveniently for him, it's behind a paywall.  That's how he can assert that it supports him; nobody is going to pony up forty bucks or so just to get the goods to prove him wrong.