John,

By the way, your PM box is full, could you kindly let me know if you've posted the rods yet <img src="/forum/images/graemlins/tongue.gif" alt="" />

Cheers,

David

The Porsche 993 RSR uses the same Bosch Motronic 2.10 and does not use either MAF or MAP sensors.

Another thing when tuning the MAF signals, when the idle signal is lower than 6 CFM (around 0.45v), the injector light will come on and the throttle response will be very jerky. The easy and quick way around this isssue for those who are running uprated injectors, is to use a POT to fine tune the baseline signal of the MAF so that it will run smoothly, of course, the alternative is to buy SAFC II or Unichip, or get a re-mapped chip. <img src="/forum/images/graemlins/laugh.gif" alt="" /> For those interested, kindly let me know, I'll send you the autospeed article (Issue 216, 5-Feb,2003) of how it works and some of my results as well <img src="/forum/images/graemlins/tongue.gif" alt="" />

Ah - looks like you've sussed the second part of my problem - I have a permanent injector light - resetting the error code with a code reader clears it, but its back within a few seconds of running. Also, the low speed / low throttle running is utterly horrible - very jerky coming on and off throttle (in fact the throttle appears to have a mind of its own - it can accelerate and decelerate without moving the throttle pedal)



PTS tried repeatedly to sort out the nasty low speed running via the Unichip map, but didn't manage to get it right. I have a feeling they were being prevented by the wayward lambda integration - any thoughts or suggestions?

Back to basics - if the reason for the jerky response is an idle signal under 6CFM, is it because the Unichip is intercepting the true signal and decreasing it? There's absolutely no reason why my engine will consume any more or less air at tickover or low engine speeds - there's no boost and its still 2 litres. The only engine mods are flowed head, pistons and these fccuk'in 16vt injectors

Due to my very basic knowledge of fuelling, I'mm still struggling to get my head round why we can't just dial out 14% of fuel across the entire range, which will reduce the effective flow of the 384cc injectors to 330cc - thus fooling everything into thinking the injectors are still standard.

I'm obviously missing something

Back to the top - if I disconnect the lambda, do I need to do anything else?

You can't just dial-in an arbitrary fuel reduction because there are a load of fuel maps that you will be affecting, and it may not be appropriate for all of them. If you get dodgy throttle response whilst driving it is likely to be the throttle enrichment map which it is probably very difficult for the Unichip to do anything at all about because it cannot easily sense when this map is being used and when it isn't.
If the idle is a problem because of the signal you could get you unichip idle rpm increased so that it is within that threshold.
John

That's exactly the same with uprating the injectors/ fuel pressure without mapping, how does anyone know the actual influence on the overall mapping? The thing to do is to try to get it right with the stock lambda, which gives you an even worse result. Even with unichip, as the ECU will still try to compare the resulting AFR reading from the stock lambda, therefore, it will always try to correct the readings, that's why resetting the ECU will solve the problem for a while, then it will come up again. It's worth noting that with every reset, the ECU also will try to determine the best ignition advance that such setup allows, that adds even more parameters to the equation.

Altering the idle mixture will either shift the entire fuel map to the right or left, which will inevitably give an incorrect mixture, which requires a remap with wideband lambda. Anything mapped with the wideband lambda on the RR and then eventually running on the stock narrow band lambda will eventually give all or some of the problems experienced, i.e. hesitation, hunting idle, decreased throttle response, etc. If I can turn the MAF signal to +/- 50% and run the car in WOT without blowing it up, there're gotta be more load sensors, other than the MAF alone (e.g. TPS, air temp,), so if the MAF signal goes haywire, it will refer to the predetermined value (crippled mode).

All of the previously mentioned methods have been tested for more than 2 years now and no problems encountered, especially now I've got Barnacles Widget to check for error codes and other logs. <img src="/forum/images/graemlins/smile.gif" alt="" />

Here's a question I've been meaning to ask, and this is good place as any.

We all know that the MAF is ignored once it reaches its 5 volt limit, and the ECU fuels according to its inbuilt maps.

Now surely the Coupe must have a MAP sensor somewhere, to know how to fuel 300bhp+?

The throttle position sensor will be maxed at WOT, the MAF is maxxed at 5 volts, so somehow the Coupe must be able to tell the difference say between 1 bar (300bhp) and 365bhp (1.45bar) say in my case.

So it is relevant to the discussion that the ECU must have a way of determining airflow rather than the MAF?

Joe <img src="/forum/images/graemlins/smile.gif" alt="" />

No it doesn't have a MAP sensor. But it does also take signals from the barometric and inlet temp sensors as well as throttle position. The car cannot tell the difference because it has no signal it can use to differentiate.
But your example has another variable - you are running less boost and therefore there is less pressure at the fuel rail, and therefore less fuel is supplied.

Thanks John, was going to pm you, I knew you'd know the answer <img src="/forum/images/graemlins/wink.gif" alt="" /> <img src="/forum/images/graemlins/cool.gif" alt="" />

I've always wondered how it could fuel differently between 1bar and 1.45bar, so the difference in fuelling is purely down to the rising rate regulator then.

However,.. if this is the case and the ECU still is unable to tell the difference between 300 and 350bhp, then you should see that injector duration becomes maxxed out at 300bhp <img src="/forum/images/graemlins/idea.gif" alt="" />, and the only way extra fuel gets in beyond this is via the rising rate regulator.

However, I'm sure a few of us have seen the injectors not quite maxxed out up to ~330bhp, so I'm still confused as to what is stopping the ECU maxxing them out?

regards

Joe

You have to admit it sounds rather primitive though!

I was hoping for something more elaborate and "intelligent" but this one seems to resort to mechanical regulators keeping the pulsing the same <img src="/forum/images/graemlins/confused.gif" alt="" />

Oh well...

V.

The ECU can't max them out because the settings in the actual ECU itself are not high enough. Also the manufacturer will probably try to ensure the car never runs more than 80% duty.

In an attempt to find the cause of the horrible on/off low speed fuelling, I hooked up my code reader and went for a drive.

The problem is that at tiny throttle openings, the car will either accelerate or decelerate, but wont hold a steady throttle. In low gears (ie trickling at 5mph in 1st gear) its quite nasty - just like pressing and lifting the throttle pedal repeatedly, or even as bad as turning the ignition on and off.

The code reader showed nothing at all:-

The TPS voltage didn't move when the fuel was dropped
The TPS angle stayed the same
The MAF figure stayed constant - occasionally dropped below 6 cubic meters per hour, but not for long
MAF voltage was constant
The only weird aspect was what I already knew - when the throttle is being shut for me, its as if the fuel was being cut - the O2 voltage was dropping as low as 0.03v (should bottom at 0.1v?)

My guess is that a combination of the lambda and MAF is telling the ECU that it should be decelerating, then once the fuel is cut, ity changes its mind.

I've disconnected the ECU - I'll be disconnecting the lambda in a few minutes - fingers crossed that it works

BTW - another point that I noticed - the MAF maxxes at 408 cubic meters per hour - this was at about 4,000 rpm at full throttle - basically its a bit like when the lambda goes open loop - this makes me think that it may be possible to disconnect the MAF as well - I'll try it

Ah yes - I've also spoken at length with Paul at PTS and also Paul Wilkes at Dastek in Scotland

Dastek are developing a new Unichip, but they think they may know what needs to be done to correct my problems - Dastek asked for PTS to get in touch, which I have asked them to do - just waiting for a diagnosis

by dis-connecting the MAF the ECU will not have any indication of air flow into the engine.

so lets say it resorts to the Richest fuel map it has, that would mean that you would need to run your higher boost setting all the time, turning the boost down would result in serious overfuelling.

im really not convinced removing the MAF will help, you cant have no MAF or no MAP sensor. i favour the lambda option.

nigel have you spoken to paul about it, what are his opinions of removing the lambda?

can the unichip offer him enough adjustment to give good fuelling at idle or low revs, or will it be overfuelling?

Disconnect the MAF plug, if it stops doing what your describing then 99% sure its a dodgy MAF. That test should only be done on idle though imo.

Paul seemed to think that disconnecting the lambda would be fine, as the base map would be OK for low speed running (together with the Unichip adjustments on top) and the top end would be unaffected anyway once the ECU goes to open loop

Paul should also be able to spend a bit more time on low speed mapping to ensure that there's no under or over fuelling issues at low revs / low throttle / low load

Update:-

The car ran OK without the lambda - it took a couple of turns to start it, but it settled very quickly into a steady and very smooth tickover. Looking good. Injector light came on straight away and didn't go off. My AFR meter showed two or three LEDs, suggesting slightly lean. Throttle response seemed very good, with none of the usual hesitance coming off tickover.

Drove off and instantly noticed that the AFR lit all bar one of its LEDs - never seen the ninth light up, ever! Even trundling at part throttle in 4th gear at 30mph showed full rich. It occasionally dropped down, but soon went back up again. Lifting the throttle killed all the LEDs.

The hunting and jerkiness is still there <img src="/forum/images/graemlins/frown.gif" alt="" /> - maybe a tiny bit better, but that might just be a placebo effect caused by the smoothness and extra throttle response.

At least I now know its not the lambda causing the on/off effect.

Full throttle is totally unaffected, as I would expect.

I've reconnected the lambda, as I really don't fancy washing the bores and wearing my expensive pistons and rings.

I have a feeling that the Unichip is screwing the base map after PTSs attempts to sort out the low speed running. Maybe another couple of hours on the rollers without the lambda and they could sort the mixture out.

[Jim Royle mode] AR$E [/Jim Royle]