Honda D Series Forum banner

1 - 20 of 64 Posts

·
Registered
CRX ECODiesel Ram
Joined
·
1,036 Posts
Discussion Starter #1
I upgraded to the PLX wideband for faster response times, and the newer bosch sensor. I decided to play around with full time close loop since I am running the Hondata s300 and the PLX seems to have a fast enough response time to do it.

I had noticed my prior sensor would still show a lot of afr "wander" in closed loop, but it was an acceptable range most of the time. Continuous tweaking of the tune was not improving much.

I set my closed loop parameters to cover 99% of my driving, and went for a test drive. In part throttle drive (low load) and cruising it was rock solid in the 14.4-15 range, while under heavy throttle it was about 14.0.

Obviously this still needs a lot of work, however my initial results are surprisingly nice, I was not expecting such stable AFR and stock like drivability. I have been trying to tweak the high load afr to be richer, and since I have NEVER tuned like this its definitely something different.

Eventually when i have installed my cam, transmission and a new header I will dyno tune as well, but for now on the street this has been pretty neat.
 

·
Registered
Joined
·
5,880 Posts
What do you mean by heavy load and why would you do/want this?
you even quoted him saying THROTTLE not load.


Big difference. more throttle means the ecu is adding fuel and pulling some timing at certain points.


Other than being a little lean under high throttle, seems good.
 

·
Registered
CRX ECODiesel Ram
Joined
·
1,036 Posts
Discussion Starter #4
mattlison pretty well covered the load/throttle statement.

right now in normal commuting type driving it works so much better due to a faster wideband reading. the hondata was always a step behind because the wideband was giving it data that was not "up to date" so it always seemed to over compensate by swinging wide in both directions. For wide open throttle this was less an issue, but with such great response why not figure out how to set target lambda and let the system handle the fine tuning response, especially with the climate considerations i mention next...

The reason I would like to make this work is the variation in humidity and temperature in the area I live. Daily a temperature swing of 40 degrees F is not unusual, and the annual swing can exceed 90 degrees. Humidity also ranges widely. Having the closed loop IAT, barometric and other adjustments seems like it would be a good idea since what works great at 90% humidity and 110 degrees is not always perfect for 20 degrees and 30% humidity.

I am playing with the parameters to get to the point where I understand what each one does, I don't THINK I am THAT stupid but some of the stuff I am reading on the Hondata site doesn't full make sense to me, and that is more likely an aspect of lack of experience on my part. But I also know enough to know I don't know enough...and that the Kpro and newer ecus are even more bad ass than I realized since it seems they do all this natively...
 

·
Registered
Joined
·
504 Posts
I can see a lot of reasons why a person would do this. My next question is how well it works? I would want to be a lot richer under heavy throttle. Foot to the floor I would want it to be low to mid 13's. A lot of newer cars run closed loop 100% of the time, but the ecu's are very smart. You just give it the parameters and it learns your style of driving. My audi will "self" tune itself to a certain point with turbo upgrades. The computer knows it wasn't a to be at X a/f at x boost pressure so when it's not there, it adds or subtracts as necessary. After a little bit of driving, it then knows around how much fuel is needed there so it only has to work a little bit. Also, the use of a knock sensor is great too as it will pull timing based on this as well.
 

·
Registered
ej8
Joined
·
6,365 Posts
Not much new.

I did this with my neptune setup and it runs damn good.
The high vac areas (17 to 22) crusing are in the 17:1
Not a hint of dent and rides smooth.

When you get the wide band in there you cam start tuning it like a lean burning car.
 

·
Registered
CRX ECODiesel Ram
Joined
·
1,036 Posts
Discussion Starter #8
The settings for closed loop afr are pretty limited. I'm just going to keep richening the load based target afr until i see enough change...
 

·
Premium Member
88 RT4WD wagon
Joined
·
227 Posts
it works for me on s300, keep your tables as close as possible then only allow it to deviate a small percentage. I allow 5% lean and 15% rich. Its fun to go full throttle watch it miss by a little bit then lock on my target lambda(which if youre using it this way then e10 or ethanol free should give you the same lambda)

just have to hope your sensor doesnt fail....
 

·
Registered
ej8
Joined
·
6,365 Posts
Does neptune allow cell based target afr in closed loop?
You set it up by defining how much it can add or subtract fuel.
Then you pick the AFR you want it to target.
then you pick when it switch open to close.

It doesnt do it by cell, But if the car is tune to your likely and then you adjust the afr it should stay pretty close to what you want.

and they would say. "its all in the tuning."
 

·
Registered
CRX ECODiesel Ram
Joined
·
1,036 Posts
Discussion Starter #12
Sounds like Neptune has it very similar to the Hondata then. I would LOVE to see cell based targets, or at least columns, but I can live with this, basically it looks like setting the target afr is going to be based on MAP reading, and tweaking the differing levels to allow for decent interpretation is where I am heading.
 

·
Registered
CRX ECODiesel Ram
Joined
·
1,036 Posts
Discussion Starter #13
it works for me on s300, keep your tables as close as possible then only allow it to deviate a small percentage. I allow 5% lean and 15% rich. Its fun to go full throttle watch it miss by a little bit then lock on my target lambda(which if youre using it this way then e10 or ethanol free should give you the same lambda)

just have to hope your sensor doesnt fail....
I am using a brand new Bosch lsu 4.9 so it should have plenty of life in it.
By any chance can you send me your closed loop and closed loop advanced settings in a screen shot? I know I am not doing anything new but I am curious to see if I am on a path that will get what I want, and it sounds like you mostly have it.
 

·
Registered
CRX ECODiesel Ram
Joined
·
1,036 Posts
Discussion Starter #15
thank you, not too far off of what I was and am using. I am going to continue to drive and test seeing how changes in your direction assist me, yesterday I had it so that my WOT afr was in the mid to low 13s, however my cruising was also sitting high 13s to low 14 rather than that leaner spot I want.

i had a ten percent range for adjustment, but adding more range that will allow it to richen up more is definitely not a bad idea to try.

I will be curious to see how much more development Hondata does in this direction, I suspect not much but who knows...
 

·
Premium Member
88 RT4WD wagon
Joined
·
227 Posts
honestly if i turn it off its still really close to my fuel tables. It seems to work well though. my thoughts were to only allow so much on the leaning because even if the sensor fails i should be fairly safe. and since i usually retune in the summer it can use a little extra fuel in the winter.
 

·
Registered
94 Integra
Joined
·
1,225 Posts
One thing to note about the OEM narrowband code, it's supposed to oscillate for improved emissions. Oscillations for a 3-way cat aren't long enough to reduce MPG's. My H23A AFR's swing like your narrowband did, yet gets >30mpg all around and has seen a max of 379mi/10gal on a long trip in a very heavy Integra.

Also the voltage scale involved also factor into it. A tiny AFR change could change narrowband O2 voltage by half a volt at certain ranges. Hondata/NepTune/eCtune all have very poor narrowband conversions, mostly due to how semi-linear (low resolution) their narrowband voltage-to-afr charts are.


Audi/VW 2.0T gasser loves to run at 0 psi/vac...
But at what drive pressure? A restrictive OEM Turbo system at 0psi may consume less oxygen than a D15 at -5psi due to scavenging. This is the same reason a ramhorn on the same engine/turbo can make 75whp+ over a cast log manifold, or how header A can make 10whp more than header B.
 

·
Premium Member
88 RT4WD wagon
Joined
·
227 Posts
if you're using a narrowband to do this......ehhh you really just cant. Id would even stick to PLX, the MTX-L, or AEM UEGO for the fastest reads and make sure your ground for it is on the thermostat housing for the least voltage drop possible. put your sensor where they recommend too. the stock location is about 8 inches from the head and most say 18-30 for recommended.

you definitely could only do this with a wideband and you want your whole scale to be as accurate as possible for obvious reasons.

I'll see if i can make a log or a video of what the o2 gauge does sometime soon.
 

·
Registered
CRX ECODiesel Ram
Joined
·
1,036 Posts
Discussion Starter #19
I have not had a narrowband in my car for close to 2 years. The sensor cost of the wideband is cheap enough it makes no point anymore, and I already invested in the Hondata anyway.
I am going to play with the ground location and see if that changes anything as well. I have to say that I am very pleased with the PLX DM6 and Bosch lsu 4.9.

I am still seeing some leaner conditions under full throttle, though it is getting closer, I suspect that is more to do with my tune than anything, and since I am putting a cam in next week I am not in a rush to fix it until then...where I was seeing 14-14.2 I am now seeing 13.5-14, with occasional blips higher.
I like the lean conditions I am seeing when cruising though, despite driving liek a complete asshole the past few days where the gas pedal is an off/on switch mostly I am seeing better than normal economy. Where it was rock solid 14.7 in cruising I am now closer to 16, and performance/response is acceptable as well.
 

·
Premium Member
88 RT4WD wagon
Joined
·
227 Posts
i had my last one cruising at about 17.5-18 but needed gobs of timing to stay smooth. squeezed out 40mpg from a rt4wd wagon though, then the motor mount bolt decided to be a quitter and dropped into the timing belt.
 
1 - 20 of 64 Posts
Top