Getting the most out of lean burn (Honda)

[arcosine]

Honda CVCC ran lean 100% of then time, >18:1. it was easier back then just to rejet the carb and play with the timing advance.

[Daox]

Wow, I am surprised you were able to idle running that lean. Just goes to show Honda really can make a nice engine.

[arcosine]

How much does the ignition timing advance when it switches to lean burn mode?

[greasemonkee]

Originally Posted by arcosine

How much does the ignition timing advance when it switches to lean burn mode?

I believe it was approx 10 degrees more than stoich burn with very little advance - this was at no load.

I'll have more timing data tomorrow as I'll have to see how much the fuel controller threw the timing out. Here's the thread where I'm putting the data I've collected, which gets updated frequently. There's always something unusual that pops up about this engine. Today I confirmed this engine operates in closed loop under ALL operating conditions.

http://ecomodder.com/forum/showthrea...nfo-18050.html

[AndrzejM]

I'm just thinking, maybe it would be easier to switch ECU with Megasquirt or any other EFI computer and not fool stock ECU with additional signals to get what you want?

[Jakins]

So in the video how did you jump into lean burn so easily? (still trying to learn how to use my HX after having it for 6 years) do you just floor and let off and match up the vacuum meter at 4? Also you say lean burn stays on in 5th gear and neutral. Is that why mine idles high when I'm coasting in 5th and than when I stop the idle drops?

[greasemonkee]

Absolutely, 5th gear tables are the same as neutral so it behaves the same, if the ecu senses the car is moving it switches to the moving idle and 5th gear tables.

In the p07-A00, US federal 92-95 ecu, leanburn is reactivated by letting off the manifold pressure until you reach 7in/Hg, then it kicks in again. This is the reset procedure - otherwise it would constantly toggle between leanburn and stoich burn at the slightest throttle movement, that would be an annoyance.



As I stated before, tuning this engine with the available software on the market (ie. neptune,crome,ectune,hondata, ect) there are many obstacle to overcome that would have to be worked around aside from the cost. Just as what happens in the above paragraph, I've studied Neptune (personal favorite) and see no way to mimic that behavior. Did I mention cost? Approx $800 by the time I get another laptop to go with it. What I'm doing in spoofing the ecu only costs about $150

[slownugly]

i have a question. you said in the first post somthing about using a p07 ecu from england. what is the difference between them? do they run lean burn at different times or more frequently?

[Cobb]

What is the guage you are using to read AFR? 4 inches of murcery? I have a 2010 honda insight and myself and a few other owners find light throttle 50 mph and about 4 on the MAP sensor on the sg2 we reach 100 mpg on the MID. Wonder how this would work with the Insight 2?

[greasemonkee]

Originally Posted by slownugly

i have a question. you said in the first post somthing about using a p07 ecu from england. what is the difference between them? do they run lean burn at different times or more frequently?

There was a thread recently about the UK P07, specifically the P07-G01 ecu, which is what basjoos has in the aerocivic. Supposedly this ecu does not cut leanburn off at 2600 and stays engaged in vtec. I pondered this a while and realized running the extra rpm my defeat the purpose of staying in leanburn in the first place, but nevertheless better than the US version. Good luck finding one.

Originally Posted by Cobb

What is the guage you are using to read AFR? 4 inches of murcery? I have a 2010 honda insight and myself and a few other owners find light throttle 50 mph and about 4 on the MAP sensor on the sg2 we reach 100 mpg on the MID. Wonder how this would work with the Insight 2?

Innovate XD16 gauge and LC1 wideband controller, I just put a new sensor in.

The vacuum gauge is an autometer Z series, but am uncertain of its accuracy.



If Honda used a pressure sensor on their insight engines you could probably fudge it a bit with the fuel controller and turn the map signal down some.



Thus far I'm running leanburn about 1" of vacuum on the gauge, the Apexi meter says 15 kpa so that might as well be throttleless, it only took 7% reduction to achieve this. There's definitely a lot more useable torque and in many places where only running in 4th and <63 mph was possible I can now run the same speed or greater in 5th.

In doing this the stoichiometric burn window in getting narrow, so now engine output seems to be more fuel ratio and rpm based rather than throttle position. Accelerating through 1st in leanburn goes pretty quick and staying up with traffic from a dead stop is possible most of the time. Also the engine is producing more HP and more efficiently at high load/rpm since the air/fuel is now about 13.3 rather than 12.3.

I'm searching for a suitable air solenoid right now, the resistor jumper on the TPS signal works too. Will post values and wire specs after I wrap this up, then we'll see if it was all worth it.