Emerald Adaptive Mapping

[Guest mcramsay]

Now without wanting to start arguments about rolling road sessions and mapping sessions which I fully understand are needed for each individual engine and set up( which I will be doing in the next few weeks either at emerald or an emerald specialist I've found in Portsmouth) I was wondering what plus points the adaptive mapping function of the emerald actually are? And is it a viable short term mapping solution prior to a proper mapping session? For those that do not know about it it uses the input from a wideband lambda to compare the actual AFR of the engine at its current load site to a per set table of values which it should be at that load site. After driving for a while it will give recommendations of alterations to the injector pulse time in each load cell to achieve the preset AFR. I'm guessing it will take quite a few attempts at driving and then applying the corrections ( each time the corrections will get smaller and smaller) until the ecu is happy with the actual AFR of the engine at each load site compared to the preset table, obviously this done not take any kind of spark Timing into consideration and I'm guessing that as you alter the AFR the spark timing will need to be adjusted also? I ask because i am removing the CAT from my exhast system, obviously this will affect the mapping of the engine, however I don't know if this will alter the AFR of the engine when using the GBSC power map? I suppose really what I'm asking is if I'm using the HSBC power map is there any bonus in letting the ecu run in adaptive mode and learn my engine and adjust as necissary? ( only short term!) or is it not worth wasting my time on something that will have no real benefit?

 

Once again I do understand the need for a full mapping session on a rolling road, I just wanted to find out how usefull ( or how much of a gimmic) the emerald adaptive function is!

[brumster]

There is nothing wrong with the self-learning, adaptive mapping of modern ECUs but holding the engine in a controlled load is difficult on the road, so there's limits to what can sometimes be done without a rolling road. If you're after the last tenths of an engines performance then a rolling road (or even better an engine dyno) is the best way to go, but for average road use or fine tuning of an engine that's already been RR'd and just needs adjustment, self-learn by wide band gets my thumbs up. There are further steps to it - I would always refine the map afterwards by hand, rather than just relying on the self-learn... and as you say, it won't correct a bad ignition map - but it sounds like this bit you already have sorted (and I'm not sure the reality of things is that the ignition map needs to be that spot on - on many a tuned engine I've seen the ignition map to be quite basic!).

 

edit: I should add, the software does a good job of making the process easy. You get a nice 'target' system on the display to assist you in holding the car at or near a load site, and there are good shortcut keys (1/2/3 and 0/-/+) to manually adjust both the fuel and ignition maps at the current load site up, down or reset to original base setting. You can use this to 'nudge' the curves one way or another on the fly. I would still say it's easier to map with two people in the car, but it's just about doable by yourself once you've had some practise. Best thing to do is find somewhere nearby with a long, steep hill

[Guest old_timbo]

I've used the fully automatic and semi autotune facility in Tuner Studio with a wide band Lamdba sensor to set-up my Megasquirt. It works well for the high RPMs large throttle opening parts of the Map where the ignition timing should be fully advanced and where you know the AFR target needs to be about 12.5/13:1 for max power. On my Pinto 36 degrees seems OK (on a RR when on a single carb and a bigger overlap cam it gave best power at 32 degrees).

Trouble is at the light throttle openings & lower rpms ignition timing makes a big difference to how well the engine runs, as well as fuel economy and pick-up. In this part of the map you need significantly more than static idle advance as the fuel burns relatively slowly (hence vacuum advance mechanisms in dizzys). Working out how much is an interactive juggling match between ignition advance, mixture and acceleration enrichment. You can run leaner mixtures and get better throttle response with more advance, but how far do you go? I don't think you can get this part of the map sorted without using a Rolling Road. The type of ECU used is not going to affect the issue.

Bottom line is that in six months of tinkering I've got it just about acceptable, but did burn out a head gasket during the learning process. My winter project was to fit a supercharger (nearly completed and the reason for fitting EFI in the first place). As a compete remap will be required and everything will be under more stress, I won't be taking any chances, so will be going straight to the RR once it runs!

[brumster]

There's a good point there - it takes a lot longer doing it yourself on the road. This isn't necessarily an issue, of course, because you're possibly not in a rush when you're not paying someone else's wages - your time is 'free', so to speak

 

I don't dick too much with the advance, to be honest. I was toying with a little electronics project to rig up a knock sensor for a level of 'safety' but it's no replacement for a good session on a RR.

[Guest mcramsay]

Thanks for the info guys, good to know! Might have a little play next weekend once the car is roadworthy again! See how I get on. Need to get a RR session booked though!