ignition amps have been tried and didn't work..theres been 2 very competent people going over every possible thing it could be from uprated motorsport coil packs to ignition amps and testing every possibility and every part replacing everything more than once ..although everyone has ideas what it may be,you can be sure its been tried before getting to a point where john has had enough..the point really is that the type of cars pushing 500+ throw up so many different issues and power comes on so brutally on some that it really is unchartered teritory ..having to manufacture bigger and bigger maf housings, massive injectors etc and alter the timing to the degrees these cars need along with the amount of gasses flowing and all the other parameters like more air... more everything...is it not so that it is more likely than not that there will be issues as yet not seen on the likes of nigels which is a pretty straight forward set up with progressive power and a medium sized turbo?
pushing the boundries of any machine or system shows its weak points and issues up..there is no question about fleas competence and abilities but there was always going to come a point when the ecu became an issue and threw something up that unless every code and feature was fully understood couldn't easily be sorted
or am i talking sh1t again? it seems pretty likely to me it was always going to happen at some point

can someone answer this question ...has anyone yet cracked that ecu completely? is it an open book or are we playing with just the few bits of it we have managed to decode?

if as i suspect we know only part of whats going on then i am suprised there has not been more issues

as i have mentioned i have been told by a few people with experience that it becomes a problem for the real big power cars..they were not saying you could not get the fuelling oor ignition etc right but that the standard ecu threw up too many issues on these cars that were too much of a problem to easily iron out so they went to standalone getting rid or the maf for a map set up and a system where evrything could be seen ,read and altered ..the fact is they work and work well..the real proof that this is right or wrong is for john to go standalone and see if its all straight forward and easily mapped..

and pray it was the ecu all along.....

Suba,cdi systems may be worth trying,in all fairness i have come to the end of my knowledge as far as the car goes.
What happens internal to the Bosch ecu is way out of my experience.
But i think i have written out everything so far.
Barnacle,i by passed the coils by using the Bmw 330 coils.
These fitted perfect and had the same internal .01 resistance on the earth trigger side.
With the coils held next to the original ones i found the spark to be more intense on the Bmw coils.
However i had to gap these right down to 0.5 on the Ngk's to be able to hold the boost consistently.
But still after 5-10 mins the 1.6 setting would induce the ignition misfire.

The weird thing was that it bought up a crank sensor fault at low revs.
So something in the ecu didnt like the coils.However they held upto 1.8 bar briefly before the issue.

As far as cdi systems go,there as costly as a stand alone and imo i dont think the ecu is capable of holding such boost.The stand alone has built in ignition amplifiers.,

I have even tried going up the revs slowly,so 5-6-7k then applying full throttle and the effect is still the same.
Unusual one


Great point Rob,behind the scenes with this problem theres some very experienced people who know the product very well and we have all being trying to sort this.
In my case its ok to generalize but i am not running anywhere near the same setup as another coupe.
Similar in cases but again not identical.

Theres a load of people that have been contacting me for the last 4-5 months keen to see my solution to this problem knowing all along that at best my car is underpowered and at worst its unreliable.

you can get hold of 2nd hand Hks twin sparks for a few hundred quid - no idea if they will work on the 5 pot though. If you want to try one I have a harness / loom for one that you can try - though this may be specific to a rotary...

I appreciate the offer Simon,but i ordered the stand alone on Friday,
So as soon as i have the car back i will wire up the new unit and post the results.

Suba, these ignition amps have been tried with someone with exactly the same misfire and it didn't work...thoough funnily enough Flea cured the misfire on that car even though it had exactly the same symptoms as Johns.

I'm not sure that statement is correct - IIRC, the ECU doesn't ever get to know what the boost is

If the BMW coils temporarily solved the problem, then I'd say you've found the most likely source of the issues. It would seem that you're experiencing the same problems that I had for quite a while when Perfect Touch were trying to get my car past 400bhp. On more than one occasion, I've had to leave my car with them, due to a complete breakdown of all five coils while on the dyno. New coils (and a reduced plug gap) sorted it, but the misfire came back eventually

Thinking about it, a BMW 330 coil would be no more powerful than a Coupe coil - the fatter spark was probably down to the relative age of the BMW coil

Speak to Barbz - pretty sure he's looked into direct replacement coils

I think you are correct. There is no sensor to tell the ECU what the boost pressure is. The ECU does control the Pierburg valve, so with the right values in the map (i.e. Pierburg not bleeding off pressure from the wastegate actuator at any time) the turbo will run uncontrolled and produce the maximum it is capable off.

Its not the fact the the ecu is sensing the boost, its more likely that will high levels of boost there more volume of air/fuel going into the cylinder resulting in a higher pressure in the combustion chamber when compressed. This higher pressure requires more power for a good spark, or if its not available a poor or no spark is produced

I have spoken to and seen Barbz and theres something in the pipeline for the future weeks.
What i do know is 3 people running good power have seen identical high boost misfire issues that were cured by the update of the ecu.
In my case it did the same for some time but then it came back.

I have no idea if the ecu can monitor boost,in theory it doesnt.
But the boost will determine quicker spool up and more rapid pulsing of the coils.
The BMW coils caused more issues that they solved.The temp cure of boost lasted 2-3 hard runs on high boost followed by some weird remote faults.

One thing someone has pm'd me about from abroad who had a similar fault is the biometric sensor that monitors the ambient temps.
My equipment doesnt have a parameter to monitor this in live data.
And it has never been replaced or touched.
I would think this would cause a fueling issue which as said above the fueling is perfect,even with the misfire present.

I think that the internal ignition amplifiers in the ecu are incapable of supporting the charge thats required for the sudden advance.
Or they are assuming the coils are at fault so are shutting them down.
Again dont quote me on this as the ecu complexities are new to me on this.So assume this may not be the case.

Starting at the bottom:

- the ECU performs no monitoring on the state of the ignition coils; it just provides a ground and lets the coil get on with it. If it doesn't spark, it will show as a rich mix at the lambda sensor which will probably be ignored as you're likely to be on full boost - i.e. mapped, not calculated.

- The ignition amplifiers in the ECU are darlington (high gain) transistors and while they won't be particularly high speed, even at full speed, they are switching at only 60Hz or so. When the coil is driven low, energy is stored in the inductor of the coil; when the input is released, the flyback voltage gives the spark. They're not being stressed; the limiting factor is the actual voltage generated by the *coil* and the plasma breakdown resistance in the cylinder.

- There is no ambient temperature sensor (except for the A/C) but there is a pressure sensor; it's on the back wall by the brake reservoir.

- The ECU can't monitor boost pressure; it has no sensor. But it can keep an eye on the rate of change of revs, and of the engine load. I believe that this is how things are slowed down in a standard model in the lower gears - but it will do it by bleeding pressure to the wastegate, not dropping sparks. I don't know if your chip has removed this functionality.

- All ECUs keep track of the position of the crank/cams by extrapolating a timer between pulses from a rotating sensor. In a *very* rapidly accelerating engine, it can happen that the reference changes faster than the ECU is expecting; this makes timing of phase-important things like the sparks slightly delayed, which is probably self-correcting. Injector timing doesn't matter particularly; the injector pulse is far longer than the opening time of the inlet valve at high load. But the change in timing doesn't make a significant difference to the operation of the spark coil; on a spark to spark basis the change is probably less than a millisecond.

It might be down to the algorithm in the crank sensor and how the Bosch ECU determines position. The cranks sensor is so sensitive that you can see the crankshaft slow down and speed up on the compression then firing stroke. You can clearly see this in the example below. It's an R6 engine we use for racing in a single seater formula:



Really the minimal trigger wheel for a race engine is 24-2 anything below this and the engine will blow up. (seriously we map engines to 0.1 degree of spark right on det, you have to or you won't make the power.) We do this in a test cell and the ECU is so accurate we can see the engine out of phase when the rubbers begin to wear on the dyno's drive shaft inputs. The ECU must perform all its calculations and inject the correct amount of fuel and spark at EXACTLY the right time. On a race engine you can tell if you change ECUs. You shouldn't be able to but seriously you can. The crank signal is the BOSS and a lot of effort goes into getting this correct. I now only work on motorsport ECU's because they are so far advanced in the right areas (performance).

Below is a statement from one of the ECU manufacturers we use:

"Our core business is race engine control. In a motorsport environment time is everything and everyone is under immense pressure. Within our ECU, the main function of the 450,000-gate APA450 ProASICPLUS FPGA is to derive engine position from the engine's crankshaft trigger wheel signal. The FPGA generates CPU interrupts based on abstract crankshaft angles, rather than the physical trigger wheel tooth positions used in traditional designs, providing increased flexibility and accuracy. ECUs commonly schedule fuel and ignition actions as timed future events based on engine operating conditions at the point the scheduling code is executed. Changes in operating conditions that occur before the event can result in angular errors, and the scheduling code is often highly dependent on the crankshaft trigger wheel tooth pattern for the engine in question. In our ECU, the ProASICPLUS FPGA allows the scheduling code to be independent of signal patterns. The FPGA can also schedule events and continue to tune placement by monitoring engine operating conditions until the event occurs. This increases code efficiency and flexibility and improves control accuracy under dynamic conditions."

This sensor is so important yet people underestimate it and assume that it's just a crank sensor (don't worry i did too until about 3 years ago). The reality is get this algorithm wrong and you will encounter all sorts of control issues at some point. This might be the underlying issue of the Bosch ECU but then again it might not. The raw truth is that it will only be proved if you understand how the Bosch system FULLY works in other words fully dissemble the code and UNDERSTAND its limits.

John, When you change your ECU IF your problems disappears then consider the above statement as gospel. Good luck and keep us all updated.

Just to add,
This thread has gone a bit off track.

This is in no way an issue with Flea or Fc Performance this is my frustrations with an on going issue.
My car and its build like any have caused complications.Not many have straightforward runs in tuning despite technical ability or experience.

My car has been faultless at lower boost and still beat some bloody quick cars round the track including Fleas monster
The few times i took my car to Flea i had a engine with bad oil pressure.
The second i used a clutch that i knew was being put through more torque than it was ever designed to cope with.

I have also had issues with ecu failures,surging due to a manifold i fitted and a weak actuator.
Again nothing directly down to the map or ecu but remote faults.
My original post was to question the reliability of the stand alones and there issues.

So far lets not forget that Flea is the most famous direct ecu tuner on the coupes both in the Uk and abroad.
To date not one engine has failed as a direct result of his actions.

In my case i wanted to explore the pros and cons for stand alone as Flea isnt 10 mins away.And although anyone can sympathize with the distance issue its not a problem caused by him.
I have used him several times,and have used him directly to tune most of the cars i have built for customers.

My frustrations and impatience are my issues and shouldnt be confused with the person at the end of the mapping.
The cause of my fault could and may be a direct result of something i have done or something thats failed on the car.


I have a few things to try in the next few weeks then i will drop the car down with Flea to see if we can bang our heads together and see what exactly is causing my issues.


So on that note,please lock the thread.