R500 K Emissions. HC 1,000ppm Solved again - several years later.

[anthonym]

Asking me may not be considered wise ;-) 

[CageyH]

The lambda needs to be connected to pin 7 to work as a simple o2 sensor.

 

If it is connected elsewhere, it needs to be set to wide and + controller, AFr, and the voltage/AFR conversion table needs to be configured to suit. See page 26 of the manual.

 

For me, we need confirmation of where it is connected.

[anthonym]

applying the 1,2,3 from above:

https://www.dropbox.com/s/mo4auhmxxe3fmp8/IMG_0562.m4v?dl=0

basically little or no movement in lamda afaics

[CageyH]

And if you change the lambda input to pin 7 what happens?

[CageyH]

The lambda changed to 11.7 and 11.1 as you trimmed the fuel. Which suggests that the sensor is working.

 

It would be interesting to see what the closed loop settings are for map 2.

 

Have you tried the other maps, or have the two Steve's confirmed that Map 2 is the one.

[revilla]

I personally wired Anthony's loom.

I sat at his dining table and double- and triple- checked the final loom against this wiring plan I had made. The first line on that is Lambda signal from MBE Pin 29 to Lambda Pin 7.

This was made up from cross checking an consolidating the following:

My initial analysis of the MBE loom, identifying the Pin 29 connection as needing to go to Pin 7 on the Emerald:

Dave Andrews' notes on MBE to Emerald conversion, showing MBE Pin 29 going to Emerald Pin 7. Also showing MBE Pin 24 (which is thre ground side of the Lambda signal on the MBE loom) going to Emerald Pin 30, Analogue Ground:

Carl at Emerald's notes which appear to be the same:

I guess I could have made a mistake? Unless I was having serious brain fart or I was too distracted by Anthony's porridge and cheese ... it's on Pin 7.

[anthonym]

map 3 is the mot map (the one in the latest video). I can try the other one, have to set the feedback tab first, I'll be a little time. see post #84

[revilla]

Anthony, another thought just came into my head. Do I remember late last year you changed your fuel pump? And didn't you have some problem fitting a replacement that was slightly different? I can't remember if you were running the standard fuel pressure regulator, but just wondering if you could have somehow ended up running a higher fuel pressure? That would change your fueling for the richer. Why did you change the pump? Was it a red herring in the end?

[anthonym]

This is the track map displays as map 2 (in fact on the rotary switch maps 1 and 3 are the mot map and map 2 is the non mot map

Closed loop is at 1 minute 14 secs - I lose connection for the 3 key test and hope this one for 2 is enough, loss of the usb link is  a real pita and very frequent if I move the laptop.

https://www.dropbox.com/s/vbeb3hc1bp6l0rl/IMG_0563.m4v?dl=0

it has struck me I have "reset" the TPS on only one map and maybe it has to be done for each map.

Fuel pump: yes it was a red herring in the end, I have the original pump still fitted - and a spare pump that doesn't fit without modification. What wasn't a red herring was that that experience of bad connections is what directly lead to the changing of the engine loom.

edit: it bugs me I have no idea what car my lamda sensor came off at the French scrap yard all those years ago (long before the maps were mapped).

[CageyH]

When you reset the TPS, did you apply the new settings across all the maps?

[anthonym]

no - well, not as far as I know given I don't know how to do that, short of doing it three times

[revilla]

edit: it bugs me I have no idea what car my lamda sensor came off at the French scrap yard all those years ago (long before the maps were mapped).

Don't worry about it. All Zirconia sensors are much of a muchness. So long as it has 4 wires (it's heated), the right plug on the end (it does - or we would have noticed) and a 22mm thread (if it didn't it wouldn't go in the hole!) it should be fine. If you ever wanted to swap you could go for a Rover EU2 one, or pretty much anyhting else including the cheap "generic" ones and stick a 4-pin Econoseal connector on it and it would be fine. It won't materially affect your map. I recently swapped to a Toyota one as it was marginally shorter and fitted better around my new decat pipe.

[anthonym]

thanks, nice to know, a happy 10 euro (maybe 20)

[revilla]

I've been following on my phone. Only just got a chance to look at your video. The lambda certainly looked like it responding to the fuel trim inputs, increased fuel trim sent it even richer with the lambda dropping from 11.3 to 11.1, decreased fuel trim sent it leaner with the lambda going up to 11.7. Still struggling to know why it's so far out to start with and why the ECU isn't trimming it itself. I saw that the trim was limited to +/- 10%. Anyone know if the ECU detects a lambda that is clearly beyond trim limits whether it might not even try?

[revilla]

If your lambda sensor is reading right and you're at 11.3, that's a full 30% too much fuel (14.7/11.3=1.30).

[anthonym]

Ok well that’s progress in so far as giving us sight of the excessive HC (hydrocarbon ten  times limit)  at the emissions test.

[CageyH]

There is the option on the throttle position sensor calibration screen (I think it is a drop.down box). If you have changed it, then make sure you set it across all three maps.

[CageyH]

My thoughts are that the closed loop conditions are outside of those required for the ECU to trim the fuel.

 

A quick.way to test this would be to adjust the maximum correction to 30% on the closed loop page.

 

Make sure you have the original map saved somewhere safe though. So you can put it back on to the ECU after.

[anthonym]

In case I have a version control issue I have:

imported map 3 from the original from the two steves

populated the feedback map (to 1500 two rows)

reset the tps ctrl-t

written all that back to the ecu

made this video, which shows much activity on the 2 button and not a lot on the 3 

closed loop is entered at 2:22 

video here https://www.dropbox.com/s/ob43u0vey2vy61p/IMG_0568.m4v?dl=0

 

 

 

[revilla]

Pressing the "3" button looks to be INCREASING the fuel trim so making it RICHER, that's the wrong way to go. When you pressed the "2" button and leaded it off, at one point it started cycling up and down regularly. Not familiar with the way an Emerald presents if but that looked CORRECT. You want it cycling like that, that's the correct behaviour. Can you get it settled into a cycle like that again?

That cycling shows more normal closed loop lambda control.

[revilla]

I think once you trimmed it down by -9 you brought it within range of what the ECU could control normally.

I also suspect you were still marginal on the edge of the trimmable zone there. At a rough guess, given that you were apparently at 1.3 times optimum fuel before, you would want to trim that 50 number down to 38, so a trim of -12 -ish.

The question is - why do the current map values lead to such overfuelling?

[CageyH]

The interesting thing is that the Lambda confirms it is running rich, but what has changed to make it run rich?

 

I would trim the fuel down to get it run close to 14.7, assuming that the throttle bodies etc. are correctly balanced.

[revilla]

Anybody know what kind of memory the Emerald stores its map data in?

If it's some sort of flash it will be permanent.

If it's battery backed, will it hold settings indefinitely when disconnected? Just thinking of the battery issues you've had, the car has been standing with basically no or very low battery voltage all winter. I doubt it, and I reckon a lot of other people leave their cars with the master switch killed, but might be worth asking Emerald if this could cause it to lose map data?

[anthonym]

here is the video now with 30% top end for adjustment:

(keeping me fit here!)

https://www.dropbox.com/s/2yahcj08fz8rxej/IMG_0570.m4v?dl=0

but what has changed to make it run rich? 

THAT is what started all this. The theory being find and fix that and the maps should be fine.

is this a clue or just evidence?

the oily one is no 3 from the front. These are last year's plugs now replaced.

[revilla]

Yes we're both asking the same question.

It looks like the whole closed loop / lambda control system is actually working, it's just that the mapped fuelling is so far out of kilter that it's beyond trim range. Why would not be about right as it was mapped before?

It's also odd to me that you had to go in and enable lambda control - surely at least on the MOT map that would have been enabled previously?

It does sound a bit to me as though your Emerald has somehow lost settings.