Definately a leak in the boost controller. It was bleeding air through so I opened it up. It has two solenoids in parallel and one was leaking so I stripped it down, cleaned it and reassembled it. It doesn't seem to leak now.
A quick bit of chav bating earlier felt a little better on the boost, but I need the dyno so check it.
The chav (de-badged Honda, nearly dragging on the ground with chav Lexus lights and a 5" tailpipe) does not want to see another red 944 [
] [
] [
] He got deafened under the underpass then lit up like a flashbang during a 6,800 RPM gearchange at 20 PSI.
At this point I would like to say that it was not my exhaust flame that caused a certain accident in a French underpass some years ago ...
Anyway. WUF is now pulling without hesitation from 30 MPH in 5th gear. The new ignition has further transformed drivability and it is so civilised to drive off boost around town and on the motorway. Those that have heard (and driven) WUF are surprised by the idle quality.
A lot of the work that I'm doing is not just about peak numbers, but overall drivability. Starting, warmup, and idle are now pretty much sorted and rival and exceed how WUF used to perform when standard 5 years and 50,000 miles ago. It'll be interesting to compare the new map against closed loop in terms of economy and drivability when I've finished.
As for peak power, 400 BHP is but a milestone on a long path. The current engine management allows much further development and I've taken certain aspects of this pretty much as far as they can go and far beyond anything possible with the DME / KLR combination. Two major parts of this are the sequential injection and new ignition system.
The sequential injection has led to a much finer control resulting in smoother idle with large injectors, and a lot more fuel capacity at the top end where it's needed.
From the standard ignition I originally added MSD (multi spark discharge) and a higher output coil. The high output voltage allows a reasonable size plug gap without the spark being blown out by turbulance when running high boost on big power engines. The MSD works up to 3000 RPM by firuing multiple sparks to extend the effective spark duration to 20 degrees of crank rotation. The original KLR knock protection was retained, which allowed up to 6 degrees of retard under knock conditions.
One of the many limitations of distributor ignition relates to what range of ignition timing is available. This is because the distributor relies on the rotor arm being in contact with a stud inside the distributor cap. Obviously this only happens for a few degrees of rotation before contact is lost as the rotor heads off for the next stud. If the ignition fires when the rotor arm is out of contact range (possibly from running a lot of advance), the spark is decreased or lost.
The current system I'm using moves things on to a whole new league. The Wolf engine management creates four spark signals based on the timing map. These are fed in to a 4-channel J&S safeguard that is a more intelligent version of the KLR. This has adaptable sensitivity and up to 10 degrees of retard available on each individual cylinder and spark, with fast recovery time. It also has a knock display monitor that shows when and if any retard is taking place, and how much on each cylinder. This allows the car to be mapped 'on the edge' with high octane fuel (99), but automatically compensate accurately for reduced fuel quality or temperature variations in the engine and combustion process.
These 4 independant ignition signals (correctly timed and protected on an individual basis) are then fed to a 4 channel version of the MSD unit to allow the 20 degrees of burn duration up to 3,000 RPM. The MSD also has the high current drive to run the independant coils on each sparkplug. On my site I posted video to show how 'fat' the spark is now. It has a voltage >40KV and a lot more current. To demonstrate this, if you knock the 'finger' off a sparkplug, the new coils will produce a spark all round the circumfrence of the plug with an almost plasma like quality. The advantage of this is that I can run a larger plug gap (currently 30 thou) and get perfect ignition each time no matter what boost I'm running. No misfires anywhere in the rev range. I can also run colder spark plugs to keep combustion temperatures down because the fat spark stops the plugs fouling during idle / starting. I can also run a much wider range of advance.
Parts of the original ignition system that are now removed are the coil, distributor cap, rotor arm and king / plug leads, all of which can be prone to failure / high voltage breakdown.
As you can see, this is adding a whole new level of technology to the 944. I'm not saying that this is cheap, and it is not intended to be. What it does allow is to replace old worn parts (like the ignition components above, parts of the original loom such as injector / knock / sensor) and hence promote reliability.
Some of the by-products of changing the engine management from the DME/KLR include changing sensor type and locations. One of the most important of these is the air temp sensor.
On a standard car with an AFM, the inlet air temp sensor is located in the mouth of the AFM (air flow meter). For a N/A car, this is perfectly acceptable. On a forced induction (turbo) car, however, this is no longer acceptable.
The air temp sensor signal is used to calculate the fuelling by working out the density of the air. Cold air is dense so more fuel is needed and hot air is less dense so less fuel is needed. On a forced induction car, the air is heated by the (turbocharger). The reason we have an intercooler is to cool that compressed air back down to keep as much density as possible. With the air temp sensor mounted at the AFM (which is itself before the turbocharger) the temperature read by the sensor (and should be used by the DME to calculate fuel) bears no relationship to the temperature and therefore density of the air entering the combustion chamber.
It is believed that the DME reads this temperature once at startup and provides an overall (static until next time the DME is started up) trim for the complete journey, not changing the amount of fuel used as temperature changes throughout the journey. i.e reduces fuel in the desert, increases it in Norway.
The correct placement for the air temp sensor is in the inlet manifold where it reads the actual air temperature and hence density so to allow the injector duration to be altered to compensate. This is a live process and is used each engine revolution. On th Wolf-3D, this is a mappable table.
On the subject of protestion mechanisms, on the Wolf this temperature can indicate that the turbo is working out of it's efficiency range and hence just producing hot air. If this is the case, the engine power and heat needs to be reduced by retarding the ignition timing using a programmable table. On the standard 951 setup, however, this is not possible.
Take the scenario that you are running a K26/6 turbo at 18 PSI at high revs. The turbo is out of it's efficiency range and just producing hot air / detonation. The DME cannot protect from this because the sir temp sensor is not reading the temperature of the air coming out of the turbo (possibly 180 C by then), but the temperature of the air at the air filter (probably neared 20 C). THERE IS NO TEMPERATURE PROTECTION to prevent detonation that may easily destroy a head gasket or much more. This protection was a much publicised feature of the GURU and similar chips, but the above description shows that this is not so. Even if the air temp sensor was moved to the inlet manifold, the signal is not used 'live' to detect the problem, but only read once at startup.
On cars which move away from the purely 'bolt on' to major changes of engine management etc (such as WUF and Fen's MTTS) so much more is possible that the goal widens considerably. A byproduct of this type of development work is that more functionality can be added which is simply not possible with the existing 951 setup.
I have resisted the temptation to join the big capacity boys at this stage because I want to develop WUF on th standard 2.5 Litre capacity at this stage.
At the beginning of the process I had set a personal benchmark figure of 450 BHP. This is getting closer and the current ~400 is with the original head back on while some extra machining is being carried out on the big valve head. Hopefully by the time I put that back on, the boost control issue will be sorted and then I can get some proper mapping done at the power end to produce the headline figures to match the equally important low end and drivability work that is currently progressing well.
WUF is not a track car. It weighs 1,480 KG ! The aim is for a devastatingly fast road car that is not tempramental, retains reliability and economy when required.
It is privately developed by myself, helped by those I consider to be the best at what they do. The engine management, fuelling and ignition systems are developed in conjunction with Bob Boyer at RandomEMS in Seattle. He is a prolific and sucessful 944 racer in USA. He is also the USA agent for Wolf and supplied the wolf setup to Lindsey Racing and is my direct pipeline in to Wolf in Australia. Dave Lindsey has provided many of the parts, including custom fabrications, without delay, and is mine of technical information. Hopefully a lot of the development work carried out on my car will come back as a Lindsey Racing retail procduct for the benefit of all. I use the fully equipped workshops of Motorpreparation and Peter Philips, whom I trust implicitly, a true spannering genius. The machining work is carried out by CTM - charlie has been machining race engines for years and James has produced some spectacular fabrications. CTM will be building my next 'bottom end'. Other technical input comes from the renowned Jon Mitchell - what more needs to be said.
Actual figures are documented on the 'rollers of truth' at Weltmeister. This is the best dyno I've ever worked with - and it's good enough for evo magazine. Chris Davies is exceptional in his tuning abilities and cheerful nature and the person I bounce mapping ideas off. We've proved repeatability of results on may occasions including multiple runs in different gears which the DD always calculates correctly.
Along the way a lot of parts have been dumped (no pun). These have included the LR air intake kit for AFM, LR Dual Port Wastegate, Link 2 kit (biggest piece of crap I ever bought), Bailey dump valve(s). The bits that have survived to be put on other cars include LR MBC and #55 Injectors (on Beaky), MSD6T & coil (going on baby WUF), K26/8 (now on Will's old 951).
The sad news is that Baby WUF keeps wanting some tuning attention next [
] [
] [
] [
]
Rick.
![[FONT=verdana,geneva"]](/forum/styles/default/pcgb/space.gif ". [FONT=verdana,geneva\"]")
