Installed a wideband with a narrowband simulator

[parasonic]

Hey all,

Over the weekend, I installed a wideband Zeitronix ZT-2 package on my 99 Civic EX. After getting everything straightened out with it and a trip to Fry's to get a serial converter for my laptop, I played around a little with it.

Stock, my Civic was pushing for 14.7 AFR. It held this under most conditions, whether idling or crusing. At closed throttle and deceleration, the injectors were shut off, and the engine was running air. I tried setting my AFR simulated threshold to 15.6 rather than 14.7, and my gas mileage definitely increased. It is interesting to note that with narrowband data, the ECU sees only a rich/lean condition and varies the AFR continually, yielding a sinusoid.



As you can see, this system was designed for a turbo, but I don't have one of those. I am wondering what AFR I should shoot for. I am afraid to go any higher as I don't want to risk detonation. Ideas?

[JanGeo]

It should get a bit hotter with about 18 to 1 A/F ratio and I mean at 60 you run 5-10 degrees hotter coolant temps. Any leaner you will feel a lot less throttle power and may feel it skipping depending upon how good a spark you have. I would think 16 to 1 should not be a problem unless it is a really hot day. Be curious to see how much the MPG goes up with A/F ratio - could be a great way to stretch out a tank of gas to the max on a long trip.

[mrmad]

I've been considering the same WB controller for the simluated narrow band until I get the Megasquirt in my CRX. Thanks for the test! Not that I'm an expert in this area, but I wouldn't recommend going more then 16:1. The other thing I would do is change your driving habits a little. The stock ECU will stay in closed loop until about 70-80% throttle settings. If you are setting the AFR leaner then 14.7, I would either be cruising or making sure I was above 80% throttle to put the car in open loop. I would be leary about high revving, hard acceleration if the car could still be in closed loop. That is where the car could be detonating and you may not notice it.

[parasonic]

Yep, I have been careful about this. I am considering either buying a PocketPC so that I can change the AFR more readily or just going OBD I and doing Uberdata and REALLY getting into customization. I noticed that towards the beginning of second gear, the engine seemed to be "struggling" and then picked up when I gave it a little "extra" gas, such as, trying to accelerate at about 25mph at 30%-40% throttle, going by my VAFC II for TPS percentage.

I am definitely being careful with this. If all else fails, and I start running way too lean, I could unplug my O2 heater resistor that I made to simulate a narrowband sensor being directly connected to the ECU, and the ECU would throw a CEL and go back into open loop. I drove 250-300 miles yesterday on this, and it just ran it a tiny bit richer, causing me to get only about 36MPG highway.

By the way, how accurate are Scangauges on calculating MPG? Do they take the data like an integral, or do they take an average every second (or something)? I could see this being inaccurate, especially since I already have to enter in my displacement to get an "accurate" reading to begin with.

[mrmad]

When you say your mpg increased, what kind of numbers did you see?

[parasonic]

On one small errand that I ran today, 44.3 city with very light traffic and three red lights that I only had to slow down for, when I usually see only 40-41. I think that I got a 3-5% increase, but it really depends on conditions. For the trip that I am referring to, I could have just been "lucky." In certain situations, it could conceivably even do more harm than good to not have that power.

On the highway, I saw a definite increase. I was able to hold somewhere around 41MPG in 5th at 60MPH on what seemed to be about a 1% grade. That isn't bad for my car.

[parasonic]

I made another interesting observation this morning. Running my wideband O2 sensor, my ECU almost immediately goes into closed loop upon a cold start. When I had the narrowband in place, I wouldn't get closed loop until the engine had warmed to somewhere between 150 and 180degF. This means that I would have had to idle for a good long couple of minutes or have driven half to one mile to get everything warmed up.

Seeing that the only thing that has changed is that I have a new simulated input that is conceivably at a non-rejectable value very quickly and have a 20-ohm 10-watt power resistor in place of the O2 heater, I don't know why the ECU goes to closed so quickly. I am still noticing the bogging of my engine as I try to accelerate at certain RPM's. My AFR never gets above the low 16's, so I am fairly certain that I am okay, but I haven't gunned it with my current setup--I am the tiniest bit afraid. I don't know whether I should be though because my buddy with a turbo had his car started for thirty seconds once and put in the wrong map, and it was reading between 5.0 and air and was frequently around 18, and he didn't blow up his motor.

[parasonic]

Some new MPG stats today -- if you are familiar with Atlanta and its nearby area:

Woodstock, GA to Midtown Atlanta: Left at 0645, 6/4/2007, 31.9mi, 34MPH average, 37.5MPG
Midtown Atlanta to Woodstock: Left at 1620, 6/4/2007, 31.8mi, 38MPH average, 41.5MPG

It was stop-and-go for about a third of the trip to Atlanta and about a quarter on the way up. I am very impressed; my mileage is usually in the 34-36 range. Seasonal warm air and the O2 simulator have definitely made the difference.

[mrmad]

What AFR were you setting this to?

In the figures you are giving, you are showing an average mpg, and is the figure after the highest mpg?

[parasonic]

No, they were just the mileages that I was getting today. I know that I can do better than 41.5.

As for AFR, I have set it to 15.7. My motor doesn't perform quite as well, but that could be changed by either setting it back to 14.7 or possibly by getting a better spark. I'm not all that concerned about power right now, as long as I can out-accelerate most others and hit ~110MPH if need be. If I were concerned, I would have swapped in a K20Z3