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

[rj]

You could save the jumping in and out by removing the pedal box cover or even just by moving the actuator behind the air filter where the cable attacks :-p

[anthonym]

ah well, all done now, it moved from 108 to 106 when I retightened the screws.

[anthonym]

The advice from Minister is spot on for the original setup. 
Your car may differ slightly but significantly.
The original MBE ecu had no provision to use minus numbers in the ignition map.
The Emerald has such a provision and we often use it to give improved idle stability with better resilience to ancillary loads.
For example when the cooling fan is operating and headlights are being used etc.
Emerald retain the numbers for closed and full open throttle in the ECU and unfortunately they are not saved in the map file.
A clue to what it might have been at the time of mapping can be taken from the throttle number in the idle setting which dictates the throttle number below which idle will become active.
That number is usually about 8 above whatever the number was at closed throttle.
The standard throttle pot has very little or no provision for adjustment so when the throttle is opened to allow enough airflow to support idle the pot number is usually the same if on the MBE 967 ecu with around zero degrees ignition.
We often set up idle using slightly more airflow with around minus 4 degrees of ignition. 
The Emerald ignition idle strategy adds ignition to achieve the target idle speed but does not subtract. 
As such our map can have around minus 9 degrees in the 1000 rpm zero throttle site which although it does not operate at minus 9 which would cause excessive exhaust temps it does allow for sufficient ignition to be added to achieve a slightly higher idle from cold as is the case with many original equipment road car calibrations.
Will have a look for your map files on an old pc and email asap in case that helps.

 

 

back to #32

[anthonym]

My Tuner has asked I return the TPS to where it was to start with. 0.88 I think per the op (glad I noted it there). 

I can’thelp wondering how I will know if it is running richer or leaner.

[aerobod - near CYYC]

From the screen shot you posted in the other thread, it looks like you have a wideband Lambda sensor. If the Lambda value increases, you are running leaner, if it decreases you are running richer. Under part load at below 50% revs you should see values around stoichiometric (14.7 Air to fuel ratio or Lambda 1.0), at high revs and high load you should see values between about 12.0 and 13.0 A/F (0.8 and 0.9 Lambda) with 12.5 A/F (0.85 Lambda) giving you the highest power output, assuming enough fuel is being added to prevent hot spots or any pinking.

From your screenshot I would say your TPS voltage is may be too high at the moment if you have a wideband sensor, your A/F ratio at 11.2 is very rich implying it is sensing the throttle is open more than it really is if the fuelling maps and injector values are correct, although this is somewhat mitigated by the fact the engine is only partly warmed up with a coolant temp of 53°C. Also it is indicating 0% throttle, so perhaps it is correct? I don't know the intricacies of the Emerald ECU appropriate values, but am applying values required for proper engine operation. Your battery voltage also looks a bit low at 11.7V, which can affect sensor voltage values, although the ECU should scale / compensate readings to take that into account.

I would also suggest that if you have a wideband sensor, that you check how it is calibrated, I use an external driver box for my sensor that has a calibration process that is advised to be done every year or after the initial 3 months of new sensor use, for the most accurate output.

Screen shot from the other thread

[anthonym]

Sounds like I should return the TPS back to 0.88 as per my op (this thread?)  Too rich is the exact problem, failed emissions test with ten times the limit hydrocarbons HC and so off the road until  fixed.

My lamda sensor is just the stock one in the 4th (rearmost) primary. I guess that means not wideband?

Reminded me my alternator does not activate until about 1200 rpm. Battery is on charge.

Thanks for this, I am hoping you have hit the nail on the head.

[anthonym]

 

Here’s a few more revs, 14.4 volts thankfully.

[revilla]

Hi aerobod, I replaced Anthony's engine loom for him towards the end of last year. As we had to convert a new MBE loom to Emerald I had to do quite a bit of digging into Anthony's setup. There certainly wasn't a wideband lambda controller in there. I'm pretty sure Anthony has a standard narrowband lambda sensor; I'll have to check my records to see exactly what we wired up.

I don't really know the Emerald very well, but I have to agree with you that those screen shots look odd as it does seem to think it has a wideband lambda from the info presented. Where it is getting AFR 11.2 from I'm not sure. I'm afraid I'm not really familiar with the voltages you would get out of a wideband lambda controller, I'm just wondering the the Emerald thought it was wideband and you connected a narrowband directly to it, could that result in an input voltage equivalent to what the wideband controller would give at AFR 11.2?

I notice also that it shows it running Open Loop. Again I don't know the Emerald software well enough; is that saying that it is running open loop because the lambda sensor isn't ready yet / lambda signal is unstable / implausible or is that saying that the ECU has been configured to run open loop ignoring the lambda?

[aerobod - near CYYC]

I think you may be right about the config being for wideband in the Emerald, but a narrow band is fitted. A narrow band will generate between 0.7 and 0.9V when rich and 0.1 to 0.3 V when lean with rapid switching between the two when the mixture passes through Lambda 1.0 (14.7 A/F). A wideband Bosch sensor has to have a matched amplifier which typically will have a programmable output, my LC2 system outputs 0V at Lambda 0.5 (7.4 A/F) and 5V at Lambda 1.5 (22.1 A/F) with 2.5V at Lambda 1.0. Continually showing 11.2 A/F (Lambda 0.76) would be consistent with a narrow band sensor just showing it is always rich, but the ECU may not have the right config.

I’m not sure about how Lambda is mapped to throttle or load sites in the Emerald, but in the MBE ECUs if Lambda is mapped, a 0 target value in the map indicates open loop, any other value is the target closed loop value, but if it requires too much trim from the fuel map to attain that value, it will stay open loop. It also won’t go closed loop below a given coolant value, before the sensor warm-up timer has expired or if the throttle is rapidly moved (unstable throttle value).

[oilyhands]

The lambda reading is from the stock narrow band. It isnt running closed loop because it has not been configured to do so at that RPM/load site. Closed loop isnt the default, it needs to be setup on the injection page where you determine on the map exactly where you would like it to operate. 

Oily

[EEK]

MAP is showing no vacuum - is that OK for such a low throttle opening?

[oilyhands]

There is no MAP sensor, it has throttle bodies, load sensing is via the throttle position sensor. If you are running on a plenum with tamer cams then you can use MAP and the ECU would be configured to do so. I would also be reconfiguring the data window on the bottom right to show something more useful.

Oily

[anthonym]

I have set the TPS back to the original .88 ohms (at 2k or whatever it is). I have not done the Emerald part of this. Now the picture looks like this:

 

and a video changes from warm up to open loop. I notice IACV pos is now green at 62 and does not change (since there isn't one).

So I guess now I should do the Emerald K6 TPS reset procedure as before to undo what I did. Now done 162 to 910.

 

 

[anthonym]

and now a video of the three maps, with TPS and Emerald back to where they were

and a pic (screen grabbed from the video)

 

[anthonym]

and to compare the before and after, top is before, bottom is after:

I have to guess here, but for this "Map 2" at the same revs we have different numbers, which suggests the TPS was causing unintended settings to be used. Same for the two "advance" settings. and ignition advance. For the same revs the throttle position is 13% previously and now 3% . Same speed site different load site. Of course now I wish I had an "original" image. 

Question is, does this give any evidence about the over richness being fixed? (it HC was ten times the legal limit).

[aerobod - near CYYC]

If the throttle position is physically the same now for those revs as it was before, it should be less rich as the ECU will be supplying less fuel if it is sensing a lower opening. In closed loop at a fixed RPM you should be seeing the Lambda value oscillating either side of 14.7 at a period of a second or so.

I don't think you will get your HCs down completely until it is operating closed loop at light loads. If there is any smell of fuel when running, then it is running too rich to pass an HC emissions test under those conditions.

[anthonym]

Would being at operating temperature contribute to going closed loop?

[aerobod - near CYYC]

Yes, all warm up and load parameters have to be fulfilled. Normally 50 to 60 degrees C and the engine has been operating at least 60 seconds from start (even if it is warmed up), as the sensor has to be hot enough to give a correct reading.

[aerobod - near CYYC]

In closed loop, if the Lambda value oscillates either side of 1.0 (14.7 A/F) every second or so and the fuel trims stay below 10% correction (+ or -) when checked at steady idle, 2000RPM and 3000RPM, everything should be working correctly.

[anthonym]

warmed to operating temp and it stayed closed loop

I goofled this (with thanks to one stefan)

closed loop - lambda signal is used by ecu,
open loop - lambda signal is ignored by ecu.

To summarise;

1. Cold start & warm up phase = open loop - lambda ignored,
2. Idle & part throttle operation when at running temperature = closed loop - lambda is used,
3. Full throttle = open loop - lambda is ignored,
4. Acceleration = open loop - lambda is ignored

I presume this is true generally.

think I should recheck the TPS. I do have a new one but I am afraid to fit it for fear of altering something relied on by the maps.

[anthonym]

strikes me that at idle the throttle percentage should be zero, not 3%

So I ran the engine and altered the TPS while watching the effect on the ECU on the laptop. I got the percentage down to 0, then realised the IACV had gone to minus 62, so readjusted keeping it at zero% throttle but also keeping the IACV 62 green and not red. So that seemed sensible. Engine off, check TPS ohms figure and it's .88 BUT slightly different in some very small amount, it flicks to .89 .90 momentarily but rests mainly on .88 

However, engine is still warm, fan on, lamda is still open loop. 

 

[anthonym]

is it possible I have zeroed all my maps and it is running some emergency "map"

I keep seeing "map" screens like a spreadsheet and I would have expected these to be full of numbers, not blank.

I forgot to reconnect the TPS connector and the engine started anyway.. which set me thinking.

later: ok so one has to actually "read in" the maps from the ECU, which makes sense finally. 

It looks like the ECU with lamda is supposed to go closed loop at 60 deg C plus various other parameters all met and it is not doing so. 

How does one know if the lamda oxygen sensor is working? or not...

[oilyhands]

Ther’s a lot of hokum on this thread.

You need to set the load/speed sites at which you want closed loop to operate, the default is open loop.

You also need to configure the circumstances that trigger closed loop in the configuration, you also set how much the fuelling can be trimmed.

RTM.

Oily

[anthonym]

Perhaps I should explain, I am trying to get the car to match the maps created by the two steves, not remap it. And I have no idea what I am doing, which may be apparent, I have never used this software before and right now don't have much choice. Also and what brought me to this point is I am looking for "obvious things not working" to explain the excessive HC.

I think the TPS was one. I have just "discovered" the map entry where it sets the beginning of throttle use at 162 and I now have reverted the TPS (manually with the engine running) to 161 in that regard, where as previously it was very different (listed somewhere in the other thread I think) so I hope this has fixed that.

This leaves the lamda. Apparently(as you say) it should go closed loop with the ECU when certain parameters are met, I now know what these are and that 60 deg C is the temp and it is not working. So I am wondering if the lamda sensor (bought from a French scrap yard many years ago) has failed. Clearly I have long forgotten how to recognise a failed lamda sensor as back then it seems I knew enough to get one from a scrap yard (off a Rover K of course).

I have RTM, but I don't speak Emerald K6.

[anthonym]

and here are the parameters, finally discovered after realising the maps have actively to be read from the ECU

we should have gone closed loop ages ago.

https://www.yourmechanic.com/article/symptoms-of-a-bad-or-failing-oxygen-sensor