Notices
Vishnu Performance - California [Visit Site]

Conditional How-To: Lean-Run Protection

 
Thread Tools
 
Search this Thread
 
Old Jun 5, 2006 | 06:36 PM
  #1  
shiv@vishnu's Avatar
Thread Starter
Evolved Member
iTrader: (20)
 
Joined: Mar 2003
Posts: 4,941
Likes: 0
From: Danville/Blackhawk, California
Conditional How-To: Lean-Run Protection

To all those beta guys out there, here's how you set-up your Xede for lean-run protection. It's real easy to do and will pay off the first time your FPR signal line blows off or your methanol system clogs or runs out of fuel.

First step is to go to your MAF/Fuel table. Right-click and select "Edit this map". You will see this:



The conditional set-up should be disabled to start with. It should look like this:



To enable it, click on "Enable conditional features on this map." Then fill out the parameters to look like this:



How it works
The conditional feature sets the output of the map (in the case of this MAF table, Freq0out) to 0% if the following conditions are met. By setting Freq0out to 0, a fuel cut is induced. The "following conditions" are as follows...
IF Freq0in (MAFin) is greater than the threshold table at the bottom of the page AND AN1in (the SMART kit's Wideband o2 signal) is greater than 35% (1.75v or ~13.5:1) for a period longer than 500 milliseconds (1/2 sec)

As long as those three conditions are met, the fuel will be cut. The cut will be momentary and only occur under moderate to high boost. The car will run just fine a lower boost levels or vacuum. The check engine light will also kick on with a MAF circuit low code. If you don't want this code to kick on, you can set the output value to 5% instead of 0%. Still too little fuel to make any form of combustion under boost but enough MAF voltage to keep the MAF code from popping up.

To raise or lower the load threshold/curve above which the conditional goes on the alert, you can raise or lower the values in the threshold table. The numbers in the table right now should be akin to MAF values you would see at ~10%. So once boost goes above 10psi, the conditional is on alert mode. Then once AN1 goes greater (leaner) than 35% for longer than 500 milliseconds, the fuel cut will be triggered. The time period is a pretty important parameter. You want it set it as low as possible without inducing a false trigger. 500-1000 milliseconds is about right for this feature. For those who run NLS or launch control, you may have to raise the Threshold a bit (to just above what MAF value you generate at the luanch RPM) and lengthen the time duration to just above what you would expect your shift speed to be. This is because you will be essentially running without combustion during the cluch-induced fuel cut. This will be interpreted as lean by the system. In this case, you'd probably want to set your time parameter to 1000 milliseconds or so.

For those who aren't running a SMART system (ie. no wideband input), it is possible to use your factory o2 sensor as a triggering input. You will have to tap into the front o2 sensor and feed it into Pin 10 on the Xede's grey connector (AN1in). You can tap into the o2 sensor at the ECU harness. It is pin #76.

Evo 8 ECU pin-out diagram:



Due to the inferior resolution and smaller voltage span (0-1v, not 0-5v like the wideband) of the factory narrowband o2 sensor, you have to change your conditional AN1 trigger. For starters, I'd se the AN1 trigger to 17% (0.85v which would be around 13.0-13.5:1). You should be able to leave the time trigger the same.

Give it a shot!

Cheers,
Shiv
Old Jun 5, 2006 | 07:24 PM
  #2  
EvilBlueEvo8's Avatar
Evolved Member
20 Year Member
Photogenic
iTrader: (26)
 
Joined: Dec 2003
Posts: 1,054
Likes: 1
From: South Coast of MA
awesome shiv. two questions and a comment. question one: can you use another wideband (ie plx, etc) if its setup to output 0 - 5 v to AN1 with this method? question two: is there another input you can use instead of AN1 (what pin/connector would that be if so?)? and for the comment: sticky this!
Old Jun 5, 2006 | 09:20 PM
  #3  
jj_008's Avatar
Evolved Member
iTrader: (8)
 
Joined: Nov 2003
Posts: 2,064
Likes: 0
From: Salem, OR
Looks good Shiv. I still need more inputs though.
Old Jun 6, 2006 | 05:36 AM
  #4  
SuperHatch's Avatar
Evolved Member
iTrader: (23)
 
Joined: Mar 2004
Posts: 2,044
Likes: 0
From: NJ
Originally Posted by EvilBlueEvo8
awesome shiv. two questions and a comment. question one: can you use another wideband (ie plx, etc) if its setup to output 0 - 5 v to AN1 with this method? question two: is there another input you can use instead of AN1 (what pin/connector would that be if so?)? and for the comment: sticky this!
There is no reason that a different brand wideband would not work for this feature. You will have to determine the voltage output of the desired wideband (0-5V, 0-1V, etc) and make sure it does not exceed the 5V that AN1 is capable fo seeing. Hook your widebands output to the AN1 input on the Xede and set the conditional parameters just as Shiv described. Just make sure you use a voltage that is comparable to 13.5:1 afr on your wideband.

- Steve
Old Jun 6, 2006 | 09:02 AM
  #5  
Ludikraut's Avatar
Evolved Member
iTrader: (17)
 
Joined: Apr 2004
Posts: 6,224
Likes: 0
From: 41° 59' N, 87° 54' W
Originally Posted by shiv@vishnu
...For those who aren't running a SMART system (ie. no wideband input), it is possible to use your factory o2 sensor as a triggering input. You will have to tap into the front o2 sensor and feed it into Pin 10 on the Xede's grey connector (AN1in). You can tap into the o2 sensor at the ECU harness. It is pin #76...
...Due to the inferior resolution and smaller voltage span (0-1v, not 0-5v like the wideband) of the factory narrowband o2 sensor, you have to change your conditional AN1 trigger. For starters, I'd se the AN1 trigger to 17% (0.85v which would be around 13.0-13.5:1). You should be able to leave the time trigger the same.

Give it a shot!

Cheers,
Shiv
I don't see how this could work without a wideband, unless the narrowband sensor on the Evo works different than other narrowband O2 sensors.

It is my understanding that on a narrowband sensor anything other than the stoich voltage of 450mV is not a consistent reading, since it will change with the exhaust gas temperature (EGT). A higher EGT results in an increased voltage reading for any given AFR that is richer than stoich. So 850mV, for example, could mean anything from 11.5 AFR to 14.2 AFR.

References: Innovate, wiki

l8r)
Old Jun 6, 2006 | 09:12 AM
  #6  
nj1266's Avatar
Evolved Member
iTrader: (6)
 
Joined: Nov 2004
Posts: 3,254
Likes: 13
From: USA
What if I was running a wide band w/o SMART? How can I get this to work?
Old Jun 6, 2006 | 10:14 AM
  #7  
EvilBlueEvo8's Avatar
Evolved Member
20 Year Member
Photogenic
iTrader: (26)
 
Joined: Dec 2003
Posts: 1,054
Likes: 1
From: South Coast of MA
Originally Posted by nj1266
What if I was running a wide band w/o SMART? How can I get this to work?
did you read what i wrote and the answer that i got?
Old Jun 6, 2006 | 10:23 AM
  #8  
shiv@vishnu's Avatar
Thread Starter
Evolved Member
iTrader: (20)
 
Joined: Mar 2003
Posts: 4,941
Likes: 0
From: Danville/Blackhawk, California
Originally Posted by Ludikraut
I don't see how this could work without a wideband, unless the narrowband sensor on the Evo works different than other narrowband O2 sensors.

It is my understanding that on a narrowband sensor anything other than the stoich voltage of 450mV is not a consistent reading, since it will change with the exhaust gas temperature (EGT). A higher EGT results in an increased voltage reading for any given AFR that is richer than stoich. So 850mV, for example, could mean anything from 11.5 AFR to 14.2 AFR.

References: Innovate, wiki

l8r)
To use the leanrun protection with the factory norrow band, you'll need to set the fuel cut to occur if AN1in is LESS THAN the trigger voltage (~17%). Also, it is not as sensitive to EGT as you suggest. I have never seen 0.85v equate to 14.2:1 under any normal operating condition (ie. not fouled by the use of leaded gas). It's a fine triggering input if you want to catch a not-very-rich condition. I have used it in both Evo and turbo Miata applications very successfully with absolutely no false triggering.

shiv
Old Jun 6, 2006 | 11:16 AM
  #9  
nj1266's Avatar
Evolved Member
iTrader: (6)
 
Joined: Nov 2004
Posts: 3,254
Likes: 13
From: USA
Originally Posted by SuperHatch
There is no reason that a different brand wideband would not work for this feature. You will have to determine the voltage output of the desired wideband (0-5V, 0-1V, etc) and make sure it does not exceed the 5V that AN1 is capable fo seeing. Hook your widebands output to the AN1 input on the Xede and set the conditional parameters just as Shiv described. Just make sure you use a voltage that is comparable to 13.5:1 afr on your wideband.

- Steve
My Wideband is an LM-1 wth the input coming from the A/F sensor to the LM-1 box and power coming from the cig lighter. How can I extract an out put from the LM-1 box? I need a wire that comes from the LM-1 box to the AN1 pin of the Xede, correct? I have no idea how to tap into the LM-1 box. Does anyone know?
Old Jun 14, 2006 | 04:08 PM
  #10  
Jorge T's Avatar
Evolved Member
iTrader: (18)
 
Joined: May 2004
Posts: 1,494
Likes: 1
From: Houston, TX
The pic shows Output Variable is set to Freq0 out , doesn't SMART need to be set to User1?
Old Jun 14, 2006 | 04:28 PM
  #11  
shiv@vishnu's Avatar
Thread Starter
Evolved Member
iTrader: (20)
 
Joined: Mar 2003
Posts: 4,941
Likes: 0
From: Danville/Blackhawk, California
Sticky?
Old Jun 14, 2006 | 04:30 PM
  #12  
shiv@vishnu's Avatar
Thread Starter
Evolved Member
iTrader: (20)
 
Joined: Mar 2003
Posts: 4,941
Likes: 0
From: Danville/Blackhawk, California
Originally Posted by Jorge T
The pic shows Output Variable is set to Freq0 out , doesn't SMART need to be set to User1?
Yes, if you have a smart fuel table. In this model/example, it doesn't.
Old Jun 14, 2006 | 06:35 PM
  #13  
Jorge T's Avatar
Evolved Member
iTrader: (18)
 
Joined: May 2004
Posts: 1,494
Likes: 1
From: Houston, TX
Yes, we need this sticky I had to fish for it in the bottom
Old Jun 29, 2006 | 08:27 AM
  #14  
Jorge T's Avatar
Evolved Member
iTrader: (18)
 
Joined: May 2004
Posts: 1,494
Likes: 1
From: Houston, TX
Glitch?

Today after pulling out of Walgreens I punch it in 1st gear and around 6k rpm the engine cut out. Felt like conditional lean protect fuel(maf) cut. I have it set as per instructions, except for fixed value output set a 5% instead of 0.

LC-1 checks out fine, AFRs seem like always 11.8@3krpm to 11.0:1 over 6krpm in all gears except 1st where wideband trails.
Old Jun 29, 2006 | 07:25 PM
  #15  
Kronik's Avatar
Evolved Member
iTrader: (3)
 
Joined: Nov 2004
Posts: 709
Likes: 0
From: Northern, KY
Oh that's not good Jorge. I was just about to post here asking how is the beta software doing and is it ready for general release? I hope it was just a small coincidence for ya.



All times are GMT -7. The time now is 08:12 PM.