Easimap for Diagnosis

[AlastairA]

Hi all,

After 800 miles or so, the outlet of my exhaust system is looking very sooty - I'm suspecting it's running rich and therefore I'm thinking there may be an issue with the Lambda sensor (would this be right?).

I will have access to an MBE lead in the next couple of days and was trying to research what I would look for on the Easimap interface to check the mixture and the correct operation of the lambda probe. From what I understand, at initial startup when the engine is cold and at low throttle, the ECU runs in closed loop mode using the lambda sensor as feedback to control the fuel/air mixture - but after warm-up and at higher throttle angles it switches to open loop mode and the sensor is ignored. However - how this is displayed on the Easimap interface is not clear to me at the moment.

Any help greatly appreciated!

[Amos91]

It's the other way around. The lambda won't be used until the car is warm enough. You'll see on Easimap it says something like lambda waiting for coolant temp. 

[AlastairA]

Ok thanks - and once up to temperature what am I looking for to see if it's running rich?

[Amos91]

Log the lambda on a graph and see if it's switching voltage smoothly and regularly. 

I wouldn't be too worried about a sooty exhaust, I'd just take a spark plug out and see what the colour was to see how the engine is running. 

[John Vine]

As Olly says, it's the other way round.

At start up from cold, the lambda sensor feedback is ignored (ECU in open loop).  When the coolant reaches 60C, the ECU switches to closed loop and takes account of the lambda feedback.  At higher throttle openings or higher revs, it switches back to open loop.

Using the MBE lead and Easimap, you first need to see what the lambda sensor is doing.

If you have the MBE 9A4, you can log probably 36 different values (channels).  There's a list here from aerobod.  What you log is determined by what Easimap panels you have open.

I suggest you start by performing a logging session from a cold start, running for say 10 mins.  Then, from your logfile, display just lambda, coolant temp and engine speed.  You should see something like this:

where:

Pink is coolant temp ©
Blue is lambda
Brown is engine speed

As soon as the temp reaches 60C (about 5 mins or so), you should see regular oscillations in the lambda trace.

If your display looks something like this:

you have a faulty lambda sensor (or poor connections).

JV

 

 

 

[AlastairA]

Thanks Olly & John.

I got setup with Easimap and ran the engine from cold. I added the "Lamda" channel to the page setup, but it didn't seem to register anything. The "Oxygen A" channel did seem to be logging so I assume this is the Lambda sensor output. By the time I got this sorted out the engine was already warm, but I did capture the data below. After startup, even though the engine was warm, it seemed to run open-loop for the first minute - and then I assume the oscillation in the Oxygen A signal is it switching to closed-loop. The change in the TPS signal at the right hand end was me blipping the throttle.

Is this oscillation correct? The time scale on my graph is very different to yours John, so I assume there might be some aliasing going on in your data.

[ChrisC]

I found battery Isolation clears the fuel trims, and it takes a long time after isolation for the lambda values to fall into an acceptable range for closed loop control.  While the lambda values are ignored the base map is used and this appears to be on the rich side.  Basically every time you isolate the battery you elongate the time taken to enter closed loop control. 

[AlastairA]

Thanks Chris - very helpful and I remember you saying this on your blog. Do you recommend not using the isolation switch on a regular basis? Does the fact that it runs rich for longer cause any damage anywhere (I suppose the cat won't take kindly to the deposits)?

[John Vine]

Your log certainly looks quite different to mine (#5).  Could that be due to your ECU being a 9A4 whereas mine is a 992?  Perhaps James (aerobod) could comment if/when he sees this -- or indeed anyone who has used Easimap on both 992 and 9A4?

Looking at the 9A4 panels in Easimap,  Oxygen A (Raw) and Oxygen A (Lim) both reflect sensor voltage, so I'd say your sensor is working ok  There are also Lambda and Target Lambda panels.  Have you included those?

I don't understand the X-axis.  What time units are shown?  Also, I don't understand the delay in lambda oscillations, given that the coolant temp (RH scale) is already above 85C. Is that simply the extra time taken to drop into closed-loop (as ChrisC suggests in #7)?

The lack of oscillations when you blip the throttle is normal, as the ECU will enter open-loop at that point.  Adding the Short Term Lambda Control Status panel should show this.

JV

[AlastairA]

The Lambda and Target Lambda panels weren't showing anything John (just stayed in Demo mode). The x-axis is the sample number - but given the data was exported to Excel with 1 millisecond time intervals, its effectively time in ms.

I did a couple of cycles of firing the engine up (already warm), running fir a couple of minutes and then shutting down. The behaviour of the Oxygen A signal staying flat fir the first minute was consistent each time. 

[ChrisC]

I just use the isolator with caution, no longer every time I stop, but definitely when I park it way for the week.  I think the lambda sensor has more of a hard time due to its positioning, and the cat suffers from the pops and bangs more than being overly rich at startup. 

[John Vine]

#10  The Excel export explains a lot.  Any chance you could graph the original logfile and take a screengrab similar to #5?

AIUI, there are two parameters affecting entry to closed loop from cold.  One is coolant temp (60C min), the other is the warm-up timer.   I've no info on the second, but I believe it may be 30 or 60 secs, and I assume it comes into play even on a warm engine?

#11  ...the cat suffers from the pops and bangs...

Which is the reason I run a bypass, especially with RBs.

JV

 

[aerobod - near CYYC]

Hi Alistair, as John said the timer is normal, it is used to ensure the Lambda sensor heater is up to temperature before any reading from the sensor is considered valid. For my car it is set to a value of 60 seconds and will restart anytime the ignition is switched on.

Your raw voltage looks correct and is oscillating between 0.1 and 0.8V at about 1/3 Hz, which is reasonable for a narrow band sensor, but it is a mystery why the Lambda and Target Lambda value is only in Demo mode, perhaps try removing a couple of existing channels on the page and see if they will start registering values. It is also worth trying the Target Lambda From Map channel and see if you can add a panel for that that works.

[AlastairA]

Thank you John & James.

I re-ran Easimap again this morning, from cold. After closing many irrelevant panels, I added all the available Lambda ones from the list but only Oxygen A was returning any data (see the dashboard image below). The resulting data is also below. What is interesting is that it seems to switch to closed loop operation at a much lower coolant temperature than yours John - looks like around 43 degrees C.

Anyway - thank you all for your help. I am reassured that the Lambda sensor is working and everything looks okay. Also checked the TPS settings for idle - the TPS site seems to bounce around between 0.1 and 0.6 but not hunting.

[John Vine]

Thanks for the plot. 

Re the 43C, I'm left wondering whether the Oxygen A (Raw) panel provides the same info as the Lambda panel? 

Re Lambda Control panels, the 992 has these:

whereas the 9A4 has these:

 

One for James (aerobod), methinks!

JV

[AlastairA]

Apologies for the images being a little small and blurry - not sure how I can improve them. The panels on the left hand side are all of the ones listed under the "Lambda Control" folder for the 9A4. All of them, except for Oxygen A, were just cycling through the Demo routine. Must admit that I didn't look at any of the panels in the Short or Long Term Lambda folders.

[Jonathan Kay]

"Apologies for the images being a little small and blurry - not sure how I can improve them."

The shortcut upload automatically reduces the resolution. It's not usually a problem for photos but it can be for text and technical images.

You can avoid that by using the original upload method, starting with this tool:

If you send me one of the original images I'll do it both ways so that we can see if there's a difference.

Jonathan

[Jonathan Kay]

Images compared.

Jonathan

[aerobod - near CYYC]

Hi Alistair, Caterham may have changed the coolant temp value for closed loop operation in later cars, perhaps when they swapped from the 992 to 9A4 ECU, it can be set to any value from -30 to 130 C.

It looks as though your ECU template file that is loaded is 9A4bf49a.ec2 from 11 Feb 2020. Did you manually select this, or did Easimap auto-detect it? You may get incorrect behaviour if the file doesn't match the ECU, I would not match a newer file to the ECU than the date of the flashed file that is in the ECU. I actually use version 9A4bf54a (which seems to be the latest, from 2 Nov 2020) in my ECU due to the fact it is an unlocked ECU that I have updated the firmware in to the latest available.

[AlastairA]

Hi James,

I'm afraid that I am a novice (grasshopper level) user of Easimap. It must have auto-detected the flashed file in the ECU because I wouldn't know how to select and use a different one.

How would I check the firmware version on the ECU and how do I select a particular template file in Easimap?

Thanks,

Alastair

[aerobod - near CYYC]

Hi Alistair,

On the Menu Bar, you should be able to select <System> <Device Info> to read out what your ECU characteristics are.

To select a particular template file, click on the template file name, it will then tell you that the software will now close and allow selection of a new configuration. On clicking "OK" a file selection box will appear, it should default to <Easimap 6> <Projects> <9A4> as the directory and show all the templates available:

One other thing to try is also to ensure you are in "Advanced" mode as opposed to "Basic" mode (although this does look like the case from your screen shots), if not, click on the profile to select. Basic restricts the info available:

[AlastairA]

HI James,

Thank you again - yes, I'm using the Advanced Level profile. However, the ECU firmware is being listed as 9A4bf850 and the software automatically configures itself to use 9A4bf49a. When I click on that template file and look through the list of options,  there isn't one that corresponds exactly to the ECU firmware version.

If I do select another one, when I turn on the ignition, the software states "ECU 9A4bf850 detected. The program will close and reopen with the correct template" and it reverts to 9A4bf49a. Here is the bottom part of the screen with the template versions available to me:

The top of the list looks identical to yours. Do you have a 9A4bf850 template?

[aerobod - near CYYC]

Hi Alistair,

Ensure you have R76 software installed by checking <Help> <About> and confirming Easimap 6 vR76 from 5 Jun 2020 is installed, if not install from here: https://sbdmotorsport.co.uk/wp-content/uploads/Software/20200604A-R76-SBDPublic-6.0.zip.

The latest template release seems to be f912, the link is here: https://www.sbdmotorsport.co.uk/wp-content/uploads/Software/9A-Release-f912.zip Install it in Easimap by selecting <File> <Import Files...> and selecting the 9A-Release-f912.zip file to load.

I don't have a 9A4bf850 template, the latest I had was 9A4bf54a, until I just loaded the 9A4bf91a one

I think having the newer template beyond the version of your ECU should fix the problem (at least from a version perspective).

[AlastairA]

Hi James,

Thanks again - your help is much appreciated. Before reading your last message, I de-installed the software and reinstalled it. Once I had set the page up again with the channels of interest, connected to the car and switched the ignition on, the Lambda channels came alive! Still using the f49a template. The only thing I can think is possibly the drivers were not correctly installed first time around.

Anyway, I fired the car up and got the below results. The Oxygen A (Raw) channel was doing it what it was yesterday - oscillating nicely after warm-up. Identical to Oxygen A (Lim). The Lambda channel was also oscillating but only between 0.96 and 1.04 V as an inverse to the Oxy A. Target Lambda from Map and actual Target Lambda stays fixed at 1.00.

[John Vine]

#24  Progress!

But it's difficult to read your screengrab.  Could you possibly upload it again using the high-res icon?  Also, if you could include coolant temp, that would be useful.

JV