Technokill: Building a Blown Hybrid, Part 1

This article was first published in 2005.

Over the last 100 or so years, a number of alternatives to the reciprocating internal combustion engine have come and gone. Electric cars were once more popular than petrol engine cars (and electric cars held the first speed records!); the rotary engine is now championed only by Mazda but once nearly every major car manufacturer in the world held licenses for its production (remember the rotary Corvette and Mercedes C111 prototypes?); and turbine cars were developed by both Rover in the UK and Chrysler in the US. But compared with the incredible number of piston engine cars that have been produced, none of these alternative powerplants can be viewed as having been successful. The same may happen with hybrid cars like the Toyota Prius – those that combine a petrol engine with a battery electric system. Long-term, they may go the way of the turbine car. Or perhaps they’ll be more like the rotary – still being produced, but for only a niche market. However, our feeling is that hybrids may well be more than that. Either as a stepping stone to fuel-cell vehicles or as a long-term sea-change in their own right, we think hybrid cars are here to stay. And modifying hybrids? Well, that’s on the very leading edge of car tweaking.

The thing about doing what no-one else has done is that there’s no obvious source of advice.

When I bought my Japanese-import ’99 Toyota Prius, it was with the intention of making some fairly major modifications. That could involve increasing the capacity of the battery pack (perhaps the nearest analogy would be fitting a larger nitrous tank), altering the air/fuel ratio of the engine to provide more power-friendly mixtures, or even turbocharging or supercharging. And while someone has turbo’d a hybrid Honda Insight, no one had performed a modification of this complexity on a Prius.

The first engine modification step was to alter the air/fuel ratio at high loads, a process covered at Altering Closed Loop Mixtures. That resulted in a slight increase in power (long-term it proved less impressive than described in that article), and also showed that the mixtures could be changed at will once the car had been forced out of closed-loop.

About this time a half-cut Prius – complete with high voltage battery pack – was purchased. That then gave the option of adding battery capacity, but the battery pack is an unwieldy and complex thing, with lots of internal control systems and circuit breakers. Certainly, adding battery capacity is nothing like as simple as just connecting the two packs in parallel. Also, the battery pack outputs over 300 volts DC – potentially lethal if not handled very carefully.

Driving the Prius for thousands of kilometres also showed that the battery pack and electric motor worked very well in providing short-term power bursts. Especially when accelerating away from a standstill (electric motors have peak torque at zero rpm!), the electric assist was effortless and surprisingly strong. (Surprising, that is, for a car of this much weight – 1240kg – and this much power - 43kW petrol engine and 30kW electric motor.) In fact the performance downer was really only when the battery pack was short-term exhausted – then the little petrol engine simply didn’t have enough power.

So if increasing the high voltage battery capacity was pretty difficult, what about increasing the engine’s power output by forced aspiration?

Complexities

The Prius engine is a 1.5-litre four cylinder that’s based on the Echo’s 1NZ block. Up top, however, there’s a different intake manifold, electronic throttle and other detail differences.

Most important of these differences is that the Prius engine uses what is called an Atkinson cycle. In this approach to valve timing, the inlet valve stays open for a long time – in fact, even as the piston is well into the compression stroke. This forces some of the intake mixture back out into the intake manifold and so reduces the amount of charge that is trapped in the chamber. The effective compression ratio is therefore much lower than the 13.5:1 mechanical compression ratio would indicate. However, the expansion cycle (ie when the mixture is being burned and the piston pushed down) remains at 13.5:1, which has benefits for efficiency.

Another odd aspect of the engine is its 4000 rpm redline, at which both peak power and peak torque are developed.

But the most complex part of the driveline is the way in which electric and petrol power are combined. The ‘gearbox’ (called the Power Split Device) contains two electric motor/generators connected to an epicyclic geartrain. The engine output is split between the wheels and one of the generators. The generator charges the high voltage battery or alternatively, feeds the other electric motor that in turn helps drive the wheels. This electric motor can also receive power from the high voltage battery to either assist the petrol engine or propel the car on its own. The PSD's gear ratio is a result of the balance between the speeds of the engine, the electric motor/generators and the wheels that depends on how much force is applied by each. This gives the effect of a continuously variable transmission (CVT).

One of the electric motors also acts as a quiet and powerful starter for the engine, allowing it to be stopped and started smoothly as needed. The other generator is used to recover energy from the car during braking, and store it in the battery for later use.

When the driver lifts their right foot when travelling slowly, the engine switches off.

Read it all quickly and the implications of forced aspiration aren’t particularly clear. But take just these questions:

• How does an Atkinson Cycle engine respond to forced aspiration? (The Eunos 800M runs a supercharger with an Atkinson cycle engine – but what is the Eunos’ effective compression ratio?)

• For forced aspiration, is the compression ratio of the Prius regarded as 13.5:1 - or much lower? If lower, how much lower?

• If it is regarded as much lower, is the ignition timing very well advanced to provide as much power as possible with the low combustion pressures? With forced aspiration, would this lead to early detonation?

• If the engine power is increased, will the power split device still be able to function?

• If not, will the control system shut the (electronic) engine throttle, so negating the effects of forced aspiration?

• In that case, could just the low rpm torque be boosted (ie fatten the torque curve while leaving peak power unchanged?)

• If a turbo is used, how will its bearing be cooled if the engine automatically switches off immediately after a full throttle event (as would occur when driving slowly down the other side of a steep hill that’s just been climbed)?

Boost Choices

Of the choice between a supercharger and turbocharger, a turbo gives the highest efficiency. This is because it uses heat that’s otherwise wasted out of the tailpipe, whereas a supercharger is drawing power directly from the crankshaft. So in terms of fuel economy, a turbo is a better choice.

In order that the turbo could survive all the hot engine shut-downs, a separate turbo oiling system could be used. This would use a small high pressure pump, and dedicated oil cooler and reservoir. In addition to the convenience of not requiring that the sump be removed to fit a turbo oil drain line, this would also allow the turbo to be mounted much lower than normal if required.

Alternatively, a system could be configured that prevented the engine turning off in some conditions. This could be done by accessing an ECU input from the air con. The air-conditioning system in the Prius has two modes. When in high power mode, the engine doesn’t switch off. This input to the ECU could be accessed and a timer and relay used so that whenever the engine load had been high, the timer caused this ‘engine on’ request to occur for the next minute. That way, the engine wouldn’t ever turn off directly after a boost event.

However, the major benefit of using a turbo over a supercharger is that the turbo boost can be easily mapped over the engine operating range. By using the Independent Electronic Boost Control kit (see the AutoSpeed shop), the turbo boost level could be set at each engine load point. This means that if the power split device control system has problems with extra engine power at high rpm, the boost curve could be tailored very accurately to take this into account.

The downsides of turbocharging? Firstly a very small turbo is needed – eg of a Japanese Kei class car or something like the Garret GT12 ball bearing turbo. And secondly, the exhaust manifold on the Prius faces the firewall and the top of the engine is also tilted in the same direction – making access very difficult, especially when working at home without a hoist.

A supercharger is much easier to fit. Because of the strange shape of the inlet manifold, there exists space for a small blower to one side of the throttle. In this position, the supercharger can be driven by a modified version of the standard belt drive. A supercharger also develops more bottom-end boost than a turbo, which may be important in the Prius if the peak power of the petrol engine cannot be increased without upsetting the power split control system. Compared with a turbo, a supercharger should also be able to be better matched to the engine by way of pulley diameter changes.

It also seemed to me that the supercharger boost could be mapped by using the Independent Electronic Boost Control to control the action of a large bypass valve, one that would direct outlet boost pressure back to the inlet. This same bypass valve could also act as a closed-circuit blow-off valve and to control when the supercharger stopped bypassing and started boosting. However, I have never heard of anyone else using this approach and it may prove very wasteful of power.

However, the choice was finally made on practical grounds – I found a small secondhand AMR300 supercharger available at the right price...

Next week – installing the blower and the first on-road testing