Since the engine transplant I've had problems getting the engine to rev to redline at WOT without going into “safe mode” at 5000 / 5500rpm. I've read lots of info and other peoples suggestions that have included running weak at those revs, the cold start injector somehow causing the problem, the knock sensor signal being too strong, but nothing that anyone has done has yet fixed the problem.
It seems to come down to this - the knock sensor sends it's signal to the ECU where it is filtered and analysed. If the ECU reads this filtered signal as knock, pinking, pinging, detonation (call it what you will) it retards the ignition timing as a fail safe measure to prevent engine damage. Unfortunately even when there is no knock present the filtered signal can still contain frequencies that the ECU reads as knock and retards the timing. Removing the knock sensor is not an option as the Toyota ECU needs the knock signal when the engine is started.
The solution (albeit an expensive one) is to ditch the OEM ECU and replace it with a standalone which will not require the knock sensor.
The piggy-back Unichip as installed was only able to control the fuelling – ignition timing was still being controlled by the original Toyota ECU. Dastek suggested that they now had a fix for controlling the timing so the car went back to Owen Developments for fettling. I was hoping that the Unichip would be able to retard the ignition timing slightly at about 5,000rpm and then increase it again after 5,500rpm but this turned out to be a non starter. The igniter has a feedback circuit to the ECU - if the ECU detects that the spark has not occurred when it expects it to then it just alters it to spark at the correct time - no matter how to you try to change the timing the ECU changes it back again. With the HKS FCON SZ in the system the engine would not even start so that's a no go as well The FCON works in a different way in that it modifies the signal going from the ECU to the igniter whereas the Unichip works by intercepting and modifying the signals going from sensors to the ECU. It’s a shame that nether of them will control the ignition. A standalone ECU will not have these problems.
While at the Rolla was at Owens I decided to have water injection fitted to see if that would reduce the knock being picked up by the knock sensor. Mark suggested that the most reliable way to do this was to use a boost activated system. A tee in the vacuum pipe to the fuel pressure regulator goes to the pressure sensor which at 6psi activates the water injection pump. Water / methanol mixture (50/50 mix by volume) is drawn from the windscreen washer reservoir and pumped to the injector located in the intake pipework about 6” ahead of the intake manifold. When it cuts in the intake temps drop by about 10*C. Pictures here.
In an effort to get better flow of air/fuel mixture in and out from the engine I spoke to Kevin about port matching the head. He suggested working his magic on a spare head he had available so that when the time came to swap the heads my car would be off the road for a minimum of time. While Kev was working on the head I got hold of a “head rebuild” gasket kit and a new timing belt for good measure. I took the lot up to Kev so he had all the gaskets and seals ready when he came to rebuild the head. He port matched the inlet and exhaust ports and reground the valve seats. The head on the 4A-G series engines are very well designed (by Yamaha) around the combustion chamber area so it’s just a case of taking off any sharp edges here. When the refurbished head was ready to be fitted, the Rolla went to Kev and I picked it up the following day. Pictures here.
It’s difficult to say what performance differences there are, as now there is more air getting into the cylinders but still the same amount of fuel so the mixture is leaner than before. This has been shown by the exhaust gas temperatures now being higher than previously by about 50*C throughout the rev range. There are differences in the way the engine runs – it’s not as smooth at idle but I haven’t given it any hard running with the lean mix. Now I just have to get it remapped to make the most of the better breathing and hopefully release a few more horses.
With the problems indicated above and no likely hood of a fix from any other direction I bit the bullet and took the Rolla back to Owen Developments in Oxford where Mark Harrison installed and mapped the MoTeC. As well as the M4 ECU (the Toyota ECU and Unichip have been completely removed) he also installed a Bosch ignition amplifier (part # 0227 100 200) and calibrated the Toyota MAP sensor for the MoTeC. All the original Toyota sensors are still used. From what I’ve been able to gather this is the first 4A-GZE in the UK to be fitted with the MoTeC M4 so mapping had to be done from scratch as no base map exists. Mark phoned me to tell me that the installation and mapping had gone smoothly - he had tried to get the Toyota igniter to work with the MoTeC but it wasn’t having it – hence the Bosch unit.
Maximum power is now 198 bhp @ 7236 rpm and maximum torque is 177 lbft. @ 4147 rpm. Dyno Plot here . There is no dip in power output at 5000 rpm so that problem has been fixed at last.
Aquamist water injection, port matching the head and the MoTeC M4 has released an extra 11 bhp so cost effective it hasn’t been but I’m still aiming for the magic 200 bhp – maybe I should stick some 106 RON racing fuel in it and see what happens then????????
In the longer term some higher lift cams should release the extra horses to get the power output over 200 BHP – it’s really a question of deciding which cams to use. The choice has to be carefully thought through as forced induction engines don’t respond well to large overlap so maybe getting cams just with greater lift and same duration is the way to go. More research is required before going down that road.
So how does it drive now? I'll quote Mark Harrison (the tuner at Owens) who said
"We've had all sorts of cars in here, but nothing I've ever driven pulls so well so low down the rev range as your engine. Obviously it's not the most powerful car we've seen but, it is the most fun to drive"
This from the man who deals with Mitsubishi Evo's on a daily basis. Now that can't be bad.
I don't remember the power being so readily available when I press the go
pedal. It just seems to make the power much more easily than before and of
course the problem with the ignition retarding at 5500rpm on WOT no longer
exists.
I've driven it normally (well for me anyway) for nearly two hundred mainly
motorway miles and it looks like I'm getting better mpg as well as having more
power on tap. I'm a happy bunny for the time being - until the "more power"
bug bites again.