@ Ringo,
I regards to your question, I think it is not so easy to answer, but to start somewhere, let´s see if this makes some sense.
With the current KERS (60kW) it was mentioned, that it is worth ~0.3 sec/lap.
But this assumed (in 2009) that the car would be still at the minimum weight, and that KERS would simply mean less ballast.
(see the interview with the Renault engineer, which I posted in the other thread)
On the same token 10kg are ~0.3 sec a lap.
Therefore if the car would be more then 10kg heavier with KERS there is a net
disadvantage.
It´s worth to remember that the 60kW are only deployed for ~5 sec a lap (6.7 sec max), as the systems are not powerful enough (by the regulations) to store all the required energy for an 60kW extra for the rest of a lap (~40 )sec.
Let alone +120kw for 40 sec. which would require a 4.8MJ KERS storage capacity, able to be charged in ~12 sec.
(in the other thread we ended up with 150-185kg extra weight for such a unit, and current technology)
But let´s take the system used in 2009 (2011) and the shorter time frame, and assume it would be possible to make such a unit at the same weight, you may want to consider a advantange of 0.6 sec in overall laptime or 0.3 against the guy with a 60kw KERS.
Trying to simplify the things a little bit 0.3s/60kW = 0.005s per kW/lap so, if we use your model, the next closest car (let´s say 2nd place vs. 1st place) would have an advantage of 0.013s per lap to the car infront of him.
Not all that much, if we consider that we talk about two cars (let´s say two RBR´s), which are otherwise equal.
If we look at this years performance and take Abu Dhabi as an example.
In qualifying Vettel´s time was 1:40.318 (in Q1) vs. Trulli 1:43.516 (Q1).
(I choose to use the Q1 times for all, to have the track conditions as equal as possible, and I choose Trulli as reference to eliminate influences from lack of driving (Klien), just tried get a snapshot of car performance)
Therefor if we say, that this would be the performance difference of the different cars, it would come down to ~3.2 sec a lap.
If we take the 0.3 sec/lap for 60kW extra KERS, this difference would come down to 2.9s/lap.
I don´t think, it would make a whole lot of a difference in overall terms.
Maybe it means, being one time less often laped during the race, for the slow HRT car.
But even with your system, it is unlikely that it will turn the HRT car into a winner.
The underlying performance differential if still to large, IMHO, to make your system work as it is intented to work.
If the cars, whould be more evenly matched to start with (all drivers driving a HRT or RBR etc), the system would work much better.
Maybe your proposed system would work better in a "spec" race series, where the cars/engines are closer together, such as GP2/GP3, Indycars or V8Supercars etc.
The calcs above, are not "all that dead serious", just wanted to start somewhere. So I´m happy if someone comes and proves me wrong, or proposes a better calc - no problem at all.
Nevertheless, I still see some merit in the proposed system from an entertainment PoV, to help closer racing.
But I understand, and agree, that it is a bit artifical for the purists, and that some people will oppose such a system/proposal.
P.S.:
I´m not all that much into video games (maybe being a bit old school
), and at the moment, I don´t have access to a full lapsim package, but if some of you want to test the effect of the system, you may just alter the engine power according to Ringo´s proposal, and see what happens laptime wise in your video game.
After all, it maybe just evens out the percived power difference between the engines. (aka the Renault vs. Mercedes argument)
Is that what it comes down to,..... car failures?
Where does the time go! 
