Version 1 - March 1st, 2001
There is a lot of information available to enthusiasts on how to modify their cars for more power. Some of this advice reflects a full understanding of the engine's systems and concern for its continued survival, while some is based on anecdotal successes and reckless tinkering. Before proceding with modifications to my own MR2 Turbo, I needed to separate fact from fiction to my satisfaction.
The goal of Upgrades Part 2 is to maximize performance around the stock injectors. For more power, I need more spark, more fuel, more air, more detonation protection, and a way to control it all. Of these five aspects, the elbow point in the dollars-per-horsepower curve is thanks to the fuel system: bigger injectors, and some new way of controlling them. 550cc/min injectors cost $150-$300 each (Canadian, rough estimate based on 10 minutes' surfing) and Air-Fuel computers about $600 (alternately, custom ECU programming, $1000). More stock injectors and an auxilliary injector controller are another option, similarly expensive. I want to see just how much safe performance is left in the stock injectors. Let's call bigger squirters the start of Upgrades Part 3.
The purpose of this document is to lay out the facts and assumptions that are behind Part 2 of my upgrade plan, that will -- safely, I believe -- bring my MR2 Turbo to 15psi maximum boost and an expected 270hp flywheel (around 220hp rear wheel). As of the time of writing, I have not proceeded with any modifications beyond what is described in Part 1. When I have completed the planned upgrades, which I hope to do in 2001, I will report on their success.
Power (Acceleration, actually) is only one part of the Holy Trinity of Performance (the other two parts being Handling and Braking). Certainly all three merit study, but this discussion will be limited to Power.
Disclaimer
What follows is a summary of the knowledge that I have accumulated while researching the Toyota MR2 Turbo, and automotive performance in general. Some of my sources are more authoritative than others. I am sufficiently satisfied as to the completeness and veracity of my understanding of the subject that I will modify my own MR2 Turbo based on what I know, with reasonable confidence that I can avoid unexpected engine damage or safety issues. I present the following only in case it might guide the researches of other enthusiasts. I make no claim that any of the following is true or correct; I am not suggesting that anyone make the changes here discussed to an automobile or similar system; I take no responsibility for the consequences of any actions taken by anyone based on my statements. I do not condone nor suggest illegal activity with respect to The Highway Traffic Act or any other laws.
I am not affiliated with any automobile manufacturer, seller or repair shop, nor any parts manufacturer, supplier, seller or installer, nor any automotive publication. I am in no way associated with the automotive industry except in that I own an automobile, and I declare that I am not an expert in any way. The opinions that I express in relation to automobiles and their performance do not necessarily reflect those of my employer, which is similarly not affiliated with the automotive industry.
Following the completion of Upgrades Part 1, my MR2 is running at 11-12psi maximum boost, making an estimated 230hp at the flywheel. All of the major engine components are still stock.
I am experiencing some hesitation at full throttle. I was told to expect this around 12psi by Those Who Have Gone Before Me.
I have reached the limit of some part of the ignition system.
With a 1psi safety margin, I am running at Fuel Cut.
I have reached the limit of a part of the control system.
Surely the original designers of this fine engine set Fuel Cut to 12psi for a reason: I must assume that
I am nearing the design limit of the abilities of the fuel system.
If I am to increase power safely and effectively, I must address the limitations of these three systems before proceeding to increase airflow (and I'll discuss how that will be done, too). And, to tune this monster, I will need more information about what is going on in there. That means more instrumentation (gauges).
And finally, a fifth aspect of engine tuning will be added to the great Detonation equation: heat management. Just as excess weight is the enemy of handling, excess heat is the enemy of engines.
Ignition Upgrades
Since the engine is already experiencing some imperfect performance due to inconsistent spark delivery, some small performance improvement should be realized by upgrading the ignition system. This can be considered a "clean up" of a mess left over from Part 1. These ignition upgrades are a "safe" first step, meaning that they will not push me closer to the detonation point, since the spark timing is not being changed at this point.
The hesitation that I am experiencing at full throttle is caused by the system's inability to deliver a spark (or at least, to deliver an effective spark). With the higher cylinder pressures and richer air/fuel mixture that come with increased boost, the resistance between the spark plug electrodes is greatly increased. The spark will take any easier path to ground that it can, such as arcing to the valve cover through a weakness in the plug wire insulation. This can be remedied by: new plugs, smaller plug gap, new better plugwires. The smaller plug gap decreases the resistance between the plug electrodes (good), but it will also decrease the energy of the spark (hey, some spark is better than no spark). Upgraded plugwires with their superior insulation will eliminate alternate paths for the spark to take.
The stock plugs on the MR2 Turbo are platinum plugs that do not need to be changed more than every 96000km or so. A current theory on the physics of ignition states that copper plugs should perform better than platinum in the most extreme conditions. The fact that copper plugs last only 25000km can be seen as an opportunity to inspect one's plugs more often (the spark plugs are you window on the combustion process, they can tell you a lot). I will switch to copper-electrode plugs.
As discussed in Part 1, spark plug heat range becomes important as power levels go up. The length of a plug's centre insulator determines the rate at which the plug can shed heat into the block. Plugs that run too hot can cause detonation, and will deteriorate quickly. Plugs that run too cold will "foul" (collect carbon deposits). This is a good time to switch to a plug that is one heat range cooler.
Ignition timing. The early-'90s 3S-GTE engines are a strange hybrid of electro-mechanical and electronic ignition. The ECU determines the precise spark timing electronically, but a mechanical distributor is still used to route the spark to the correct cylinder (in older cars, the distributor determines the spark timing, too). The ECU retards spark timing at high boost, and when it perceives knock events. Sometimes spark timing is retarded to the point that the distributor rotor is far past its target contact, necessitating a long arc. Over time this can burn the contacts. So, I will physically turn the distributor cap a few degrees (5?) to accomodate the chronically retarded timing. This will not affect the spark timing, it will only enable greater retard without burning the distributor contacts.
For various reasons that are best explained by Marc Summers, the stock distributor cap is an inferior design. The choice of electrode metal is not ideal, and the gap is too wide (gap between rotor and cap contacts), again promoting burning of the electrodes. Marc Summers sells modified distributor caps whose electrodes have been replaced with copper pieces, and whose gaps are less. I'm gonna get me one o' them.
These measures should be sufficient to correct the hesitation experienced since completing Upgrades Part 1. In preparation for even greater pre-spark cylinder pressures, I will install an aftermarket capacitive discharge (CD) system, which greatly increases spark voltage and energy, as well as giving other benefits (multiple spark events). Simple CD systems are available, as are more complex units that can incorporate inputs from knock sensors and other things. I haven't decided how "big" to go, yet.
In summary, the following modifications will be made to the ignition system:
New spark plugs: copper, one heat range colder Turn distributor cap towards greater retard Marc Summers distributor cap Capacitive Discharge system
Gauges
As I take control away from the ECU, and re-make the original designers' decisions, I need more information about what is going on inside the engine. In order to monitor and tune the fuel delivery, I will need an Air/Fuel ratio (A/F) gauge to watch the mixture, and an Exhaust Gas Temperature (EGT) gauge to monitor combustion temperatures. This level of instrumentation will be sufficient for the extent of tuning that I have in mind. Wide-band O2 sensors, knock sensors with individual-cylinder spark retard, and other unobtainium would be necessary for more extreme power goals.
The A/F gauge displays the voltage on the factory oxygen (O2) sensor, which measures the amount of unused oxygen remaining in the exhaust gases. A lot of unused oxygen indicates either a lean mixture or incomplete combustion; very little remaining O2 indicates a particularly rich mixture. Even though stock O2 sensors don't accurately read any mixture other than stoichiometric, an A/F gauge is still a useful window into the engine's operation. (Professional engine tuners use "wide-band" O2 sensors that are worth thousands of dollars, and they don't leave them in the car when they're done!) An A/F gauge is simple to hook up.
The EGT gauge displays the temperature of the exhaust gas, that it reads from a pyrometer (probe) which is plumbed into the exhaust as far upstream as possible. A hotter exhaust temperature can indicate a lean burn and/or detonation, but it can also indicate too much spark timing retard. A cooler temperature puts the driver at ease, and can indicate good spark timing, but can also indicate an unnecessarily rich mixture. The safe limit, by consensus, is in the neighbourhood of 800-900 F.
It is well documented that, all other things being equal, the #2 cylinder on the MR2 runs leaner than all the others at very high airflow. This is due to the imperfect design of the intake manifold; it delivers more air to #2 than to the others. Therefore the best place to put the EGT pyrometer is in the #2 exhaust runner/manifold/thingie/header/pipe. The challenge will be to find an easy and economical way of installing the pyrometer! Since I have no other reason to remove my turbo and exhaust manifold, I am looking for a way of installing the pyrometer with everything in place. Drilling the hole will certainly introduce metal filings into the exhaust manifold, upstream of the turbine (very bad, potentially). It has been suggested that I drill the hole with the engine running. Perhaps while holding the wastegate open, too? I'm still soliciting suggestions on this one. E-mail me if you have one. The other option is to install the pyrometer downstream of the turbo, in which case I should allow for at least 100C of temperature drop across the turbo (the exhaust gases cool further as they do work to spin the turbine, plus some heat is absorbed by the turbine housing). In fact, assuming that one cylinder is running hotter than the others, I should add additional margin (another 50C?) to allow for the "averaging" of the individual cylinder temperatures.
And because I just don't like the intrusive look of A-piller guage pods, I am looking for alternate ways to mount these gauges. Owners who have a single-DIN stereo head have room for three gauges just above or beneath it. Unfortunately I have the double-DIN Toyota 3-in-1 stereo, and I'm not about to replace it just to make room for gauges. One way that I'm sure will work is to mount the gauges in the ash tray cover of the centre console. I have seen this done and it's a very neat installation. The drawback to this would be that the gauges are far from the driver's field of view when driving, making tuning runs difficult without two people. Something else which I will look into is custom interior work: perhaps some skillful fabricator can create an installation that blends with the look of the factory dash. Another option: some A/F gauges are small and flat rather than round: such a display could mount on top of the steering column, without blocking the tach and speedometer.
And an intake-air temperature gauge would be nifty. Just looking into this now. ATS Racing.
In summary, the following instrumentation will be added:
A/F Ratio Gauge EGT Gauge Intake Air Temp Gauge
Fuel System Upgrades
As discussed, the designers built a fuel system that would still have lots of safety margin in it, at 12psi boost (the point of Fuel Cut). That margin is extra fuel capacity, "just incase": incase injectors clog up or the pump gets weak; incase airflow is somehow increased beyond what should be possible; incase things get too hot, and extra fuel is needed to cool down combustion. That excess fuel capacity is mine for the taking, I can use it to make more power IF I can obviate the need for all that extra safety margin. I am not the worst-case MR2 owner for which the designers must design, the owner who will run the car without proper maintenance and caution. I can take responsibility for my engine's safety, and I can take that excess capacity for my own uses.
At full boost and WOT, it has been measured that the engine is running extremely rich (<10:1 A:F), especially from 6000RPM to redline. Leaning this out will recover some power; some owners do this using an AFC (Air-Fuel Computer). I will take the opposite approach, and will allow the mixture to lean out as I carefully increase airflow.
Before encroaching on the stock safety margin, it will be important to ensure that the fuel system is working perfectly. One of the worst possible failures would be having one partially-clogged injector, leading to one lean, detonating cylinder. The oxygen sensor (and an A/F gauge attached to it) would not register much effect from one lean cylinder. Therefore, an important step will be to have the stock injectors cleaned, balanced, blue-printed and flow-tested (all one process), the result of which will be four injectors that are known to flow exactly the same amount of fuel. RC Engineering in California is well known to performance enthusiasts, but I know that there are similar shops in the Toronto area (there are lots, in fact, for deisel injectors, some of them also do gasoline injectors, but I'll try to find a shop that does high-performance work, specifically).
Good injectors need a reliable supply of fuel at constant pressure (meaning a pressure that does not drop as the flow rate increases). I have just put in a new fuel filter. The stock pump should be able to flow enough fuel for the stock injectors' needs. To optimize the pump's performance, there is a resistor pack and relay that can be by-passed, to keep the pump working on full voltage, all the time. This eliminates a possible source of unwanted resistance in the pump's power supply. The downside of this mod might be a shortened pump life. This should not unduly increase the pressure in the rail, except perhaps at low injector duty-cycle (idle). I will make this mod reversible, and note any change in A/F ratio or driveability.
The next low-buck measure that I could take, and I don't know whether it will be necessary, would be to add a variable-rate fuel pressure regulator. This does not replace the stock regulator, but acts in series with it, to increase fuel pressure at a rate greater than 1:1 with the manifold pressure (above a certain boost level). This mod would necessitate a bigger fuel pump (the stock Supra Turbo pump is a popular choice). The injector flow rate increases with the square-root of fuel pressure: a pressure increase of 50% would only lead to a 22% increase in flow. Injectors working at fuel pressures much higher than for which they were designed will sometimes stick open or closed.
In summary, the following modifications will be made to the fuel system:
Clean, balance, blue-print and flow-test stock injectors By-pass pump resistor pack and relay Maybe: Variable-rate fuel pressure regulator and Supra pump
thesandmage@netscape.netControl System Upgrades
- Bill Wotschak DIY FCD
Heat Management
- 7.5" SPAL fan on IC
- 10" SPAL fans on engine lid
- Aqua-Mist system (if necessary)More Airflow = More Power
- reference MBC to manifold
- turn boost up (12-15psi)
