john banks

Some of the colour resolution is lost in the .gif compression to 8 bit. Stephen came up with this ace way of colouring your maps - in this example the ignition timing map is overlaid with actual colourings of lambda voltage from a drive so the bits you need to alter are easily spotted from the colour key Didn't show you the knock as it was zero everywhere. RPM is down the side, load across the top. Note the reasonably nice progression from lean to rich as load increases - still some work to do on it and still quite rich.

[Edited by john banks - 5/15/2002 11:35:41 PM]

Cosie Convert

I want one of these

WREXY

Very interesting John and a good easier way to map with the colours. Let's see if I got this right. So where the mustard coloured areas are, is where it's running rich ay? And where there is purple colour, it needs a bit more fuel?

Cheers,

Wrexy.

john banks

Almost WREXY. The purple area has already had some fuel added to it as it was lean before. You don't need as much fuel at lower revs/loads. So it is about right for now. Colonel Mustard needs a beating though

WREXY

So would it be right to say that you have to aim for the yellow colour of the 881-900 section? The brown colour of 861-880, would that be regarded as a tad lean?

What do the numbers on the actual table mean like the 12,14, 0B 07 0C etc?

Cheers,

Wrexy

john banks

That is where I am shooting for now 8-9% CO. This is more like 9-10% CO still.

Numbers are in hexadecimal degrees.

WREXY

Cheers m8! I'd love one of these too. I'd need to be shown what to do, but I'd still like one.

Cheers,

Wrexy.

JenisonWRX

You say your knock was zero...
does that mean your knock correction was 0 ...
or the signal from the sensor was 0 ... I would assume the first to be the correct statement?

john banks

Both were zero. Old shape tends to run negative knock correction = retard whereas new shape tends to run positive knock correction = advance from the the base map.

iain atkins

John

Would this work on a MY97 UK ECU, if so i could be mildly interested.

Are you selling the programnming software or the re-programmed ECU?

I could also be very interested in these for RACE-TEK.

Cheers John


Iain

PS. you can mail me offline if you wish

john banks

Only for MY99/00, later models hopefully to follow no plans for earlier. I am offering a custom remap in Scotland for people that bring their cars to me and are distant from Uxbridge and Ecutek.

[Edited by john banks - 5/16/2002 12:39:54 PM]

john banks

Some more for you

The standard AE801 ignition map is shown on which my present map is based, and also my present ignition map - note the bit added in low down, and note how the column headings are changed to increase the advance a column at a time. However, the standard car runs a bit of knock correction typically in the upper midrange - in advancing the map this area needs some work to retard it again.

Notably on a cold wet evening with the same map there was no knock correction whatsoever, then as much as -5 degrees today in the heat. You don't want to leave it like that, but the ECU is coping admirably getting the map right on the move. Some standard cars do run this much when they are choked up with cats and NUL fuel. I suspect it will not retard to the same level if I just take out a few degrees at this spot.

There seem to be characteristic weakenesses in the original map - a lot of retard and leanness during spool up, too much advance in the upper midrange - at this point airflows can be very high and the turbo working very hard.

Looking at Delta Dash logs, I can see that the only time the -5 degrees occurs is roughly in the 4500-5500 RPM level at part throttle. Most of the time the logs show occasionally up to 2-2.5 degrees retard in this zone on full throttle, but the fuelling was always over 9% CO according to a (hot) lambda sensor. So retarding the ignition would be a good cure. I may also drop the boost a bit at this point.

Notably at no point during all the logs even with the -5 degree knock correction was the actual on-off knock signal output activated - you'll know what I mean if you use Delta Dash.

[Edited by john banks - 5/16/2002 6:34:18 PM]

dowser

John,

But if the knock sensor wasn't active, why was it retarding? Has it not done it once in this rev band and then 'learned' to retard?

The software looks bloody fantastic

Richard

john banks

It is reset when you remap. There was no knock signal. Presumably it only counts a knock signal if it can't retard, but it is slightly new territory - all we do is read what it puts out - the knock signal/correction need interpretation as to their meaning. Speculation - perhaps it keeps a count and transfers to the low octane or lo/hi det maps.

dowser

If that's right, then maybe there's some value to be had from isolating the knock sensor a 'little'

In the interests of science, could you lean out the map enough to generate real knock?

What are the 'Map B' and 'Comp Map' representative of?

I've also always wondered how the ecu is using the 'rate-of-change' value off the MAP on the my99/00 cars?

C'mon - post some boost screens

Richard

john banks

No thanks Dowser I already had some tinkling when it was overboosting - not a bit now it is sorted. The overboost was a restrictor problem.

Boost control uses a maximum duty cycle by RPM and load, and a target. There is a map for correction values to the duty cycle if the boost is a little bit high/low or a lot high/low.

Ign map B is duplicate. Comp maps adjust for low and high det - no need to alter them on a UK car as they are quite sensible/safe already.

EvilBevel

I'm jealous John

But nice to see lots of things confirmed, like the "semi-closed loop" boost control on the MY99/00. Hence the "fluctuating" boost just before the point of the fuel cut I guess...

dowser

Thanks John
Richard

mutant_matt

So John, are you saying that there are different maps depending on the fuel in use? If so, how does it know which one to use?



Matt

john banks

It knows by the knock it sees. I don't change the low octane maps and keep them really conservative so it will still run safely on 95 RON, although I never go below 99.2 RON (with NF).

mutant_matt

So does it retard the ignition a little bit when detecting knock (whilst running on the High Octane map) and if it has to do enough of that, it just drops down to the Low Octane map then?

Can you tell which map it's running on at any given moment?

Matt

john banks

You could imply it from Delta Dash, but I don't think anyone knows the exact circumstances. You need to do quite a lot to it even before it starts retarding even a little bit. The idea is to leave the low octane maps the same as original, or in the case of the STi I think it would be better to make them more conservative. The STi and P1 maps have very little capability to add fuel and retard when compared with the UK maps. Forged pistons yes, but det is det and they don't look very safe in comparison. The low and high octane ignition maps on the STi/P1 are the same as each other - seems odd when it is mapped to 100 RON fuel!

For boost control there is a map with different errors above and below target with duty cycle compensation values to allow decent closed loop but mapped proportional only control.

[Edited by john banks - 5/18/2002 2:05:34 PM]