With clever engineering a PHEV doesn't really need a "whole drivetrain". You can use the fact that you have a motor and a big battery to solve a ton of problems that ICE engines typically have.
To make an ICE efficient, you want to give it a big transmission with a lot of gears so it stays in peak efficiency RP, undersize it, and put a big turbocharger on. This produces a relatively heavy drivetrain that has lag when you put your foot down.
When you have an electric engine and battery, you can take the same small engine, but instead of a transmission, you just steal energy and put it in the battery when you don't need power and give it back instantly when you put your foot down. This removes almost all the weight and complexity of the transmission, keeps your engine at peak RPM, and gives you the responsiveness of an EV.
But the engineering cost to do all that is very expensive and unique, two things car makers hate and avoid at all cost. There is a reason the long discontinued Volt is still the best PHEV around and only BMW i3 has ever had your proposed layout. No one else has been willing to spend the money required to really optimize a PHEV and i3 could use more power getting up hills than the engine could generate, so your battery % could drop with the engine running. The OEMs want a 'good enough' PHEV to qualify for subsidies with minimal spend and allocate all the real resources to BEVs since they see the end of combustion.
Plus it's not just the mechanicals. You still need a gas tank, exhaust, catalytic converter and all the other ancillaries that come with combustion. That all adds to the bill of materials, engineering and assembly complexity, and packaging constraints.
> The OEMs want a 'good enough' PHEV to qualify for subsidies with minimal spend and allocate all the real resources to BEVs since they see the end of combustion.
I think a lot of this comes from seeing PHEV as "worse EV" rather than "better ICE". Car manufacturers see their business as building ICEs and EVs as a weird distraction that you have to throw money into as basically a PR expense (although this has started to change in the last couple years). The engineering cost here is expensive but not especially unique since it could pretty easily apply to the entire ICE fleet. The fact that the Prius has existed for 25 years, but most cars sold today don't have regenerative braking is kind of crazy. It's not like OEMs haven't put a ton of engineering into ICEs, they just haven't considered hybrids as "real engineering".
Except no cars do that. That's called a serial hybrid, which never shipped, unless you count oddballs like the BMW i3, with the optional "range extender".
I read a review of a prototype BMW mini. The turbine (fixed RPM) + batteries seemed like a great combination. Plenty of HP for highway cruising, plenty of battery+AWD for great acceleration. Turbines are efficient, crazy small for the HP, and have few moving parts. The downsize is having a fixed RPM, which would normally make them unacceptable for a car.
To make an ICE efficient, you want to give it a big transmission with a lot of gears so it stays in peak efficiency RP, undersize it, and put a big turbocharger on. This produces a relatively heavy drivetrain that has lag when you put your foot down.
When you have an electric engine and battery, you can take the same small engine, but instead of a transmission, you just steal energy and put it in the battery when you don't need power and give it back instantly when you put your foot down. This removes almost all the weight and complexity of the transmission, keeps your engine at peak RPM, and gives you the responsiveness of an EV.