Emerald v's 14cux catalyst fuel maps

[blitzracing]

One for the Emerald boys. The 14CUX constantly swings the mixture between about 15:1 and 12:1 in closed loop to keep the catalyst at its optimum for burning off the Carbon monoxide, hydrocarbons and other nastys. These upper and lower limits are programmed to do this, and not simply a function of the constant correction feedback from the lambda probes. Some of this constant swing, is at least partly responsible for the tendency for the cars to snatch at low RPM and low airflows, when the mixture is at its weakest. Firstly, am I right in assuming the Emerald has this "mixture swing" function already mapped into it, and can you alter the the upper and lower mixture limits? I should think this is somewhat more tricky than simply setting the mixture for the maximum power point.

Mark

[thenick]

I've not heard of it doing this... On closed loop or adaptive mode though, you can set it to trim to a different target AFM at any of the load/speed sites which may give the desired effect if set up correctly. I'd give Dave a ring at Emerald personally as he's going to give the best answer! (www.emeraldm3d.com)

[dnb]

I would guess one of two options

1. It doesn't bother with the mixture swing at all, and the lambda corrections from closed loop feedback are assumed to be sufficient.

Remember that strictly speaking the Emerald is not intended for use on the road, like most other aftermarket ECUs, so won't have to demonstrate compliance with anything like "type approval" style emissions compliance tests.

2. It does do the mixture swing when appropriate, but the parameters are not user adjustable. Again, this is perfectly satisfactory since catalyst performance is quite well understood so the "hard wired" numbers will probably work OK.

My vote is for 1, because it's fairly simple to implement and it works tolerably well. Dave at Emerald is the only one who'll know for sure, so it's worth asking the question of him.

[eliot]

Not heard of swinging between 12 and 15. I thought that the cat had to be held at 14.7AFR (Lambda 1) for optimum burn off - hence why narrow band lambda exist and why they can only [reliably] tell you when you are at 14.7.

[Pupp]

I think it must be the first option that DNB identifies...I've seen no indication of mixture cycling on my AFR gauge (it could be seen on the 14cux), nor have a I seen anything mentioned in the documentation

[trackcar]

Why would you cycle a cat mixture down to 12:1 ? that's a full power mix (richer than you would normally target for full power, most turbo cars would be happy flat out at 12:1) and will reduce exhaust gas temps and makes the cat less efficient .. most bizarre.

[dnb]

The cycling is indeed a known thing for catted cars. At first I thought it was just a feature of closed loop lambda control on old narrowband sensors, then I read up a bit about how cats work and they do seem to require this cycling behaviour for optimal emissions performance.

However I have never seen excursions as far as 12:1 on any car I've looked at. I suspect a typo here!!

[trackcar]

Indeed .. I'm not doubting that the mixture may be intentionally cycled within certain limits (though how you control that accurately on a narrowband sensor with a delay time feedback loop I have no idea) but was amazed at the richness it cycles to .. as DNB says are you sure the numbers are correct ?

[blitzracing]

I only have an LED bar graph on the car, with no numbers on it, only lean, correct and lean. I did cross reference a generic lambda sensor voltage curve against the relevant LED's to get some idea of the readings. On full load open loop, I get the richest LED under "correct", that I reckoned was around 12:1 and would make sense for maximum power. Just because a sensor is narrow band, it does not mean the voltages stop changing at each end of the scale, the voltage changes are just much smaller. It does not take a lot of electronics to compensate for this by boosting the signal change at each end outside its linear response around 14.7:1 mid point, so you can still read a moderately wider range overall. Wide band sensors are nice because the response is linear over a much wider range, that makes things simpler.

May be of interest:

http://www.pelicanparts.com/techarticles/mult_air_...

The reason for the slightly strange question, I was thinking if you could reduce the mixture swing parameters on the 14CUX, it would help make the cars a lot more drivable at low speed, albeit at the expense of emissions slightly, but not going as far as using the non cat map, that is definitely a bad idea..

Mark