Honda Grom banner

Aftermarket ecu

17K views 65 replies 9 participants last post by  Wibbly 
#1 ·
I am going to go with the 181 neo. I have done research on this and my first instinct is to get all dyno jet. I want plug n play. I want the most data I can get on the handlebar monitor/gauge. I want wb autotune, of course.

Now.. I see you guys like Aracer, and there’s options there.

So.. let’s hear what you think. My initial goal is to haul my 230lbs. uphill against the wind at 60mph at 75% throttle.
 
#43 · (Edited)
i don't believe for a second that these ecus are running closed loop throughout the whole range, what you describe would be fantastically difficult for an ecu to target, you're cherry picking a point of a very steep curve sat next to a cliff. if you give a control loop incredibly unstable feedback you aren't going to do well with your output (unless you're averaging, as it does). and given the ecu has no clue about EGTs your presumptions seem even less likely (the "target" afr could be anywhere between .78v and .95v, which could actually cover a range as rich as 10.2:1 and as lean as stoich, so how is that going to work out without EGT feedback?)

i'm familiar with what the EFEI does, i already mentioned it.


i suppose we'll find out eventually if this ecu runs full time closed loop (with a target AFR of 13.2 from the narrowband o2 sensor as you seem to suggest), or if it does EXACTLY what the standard ECU does (it IS afterall, a standard ecu running a modified project), but with different open loop mapping.


and the o2 sensor feedback is in Volts. not mV. 0.1mV is nothing. but you know that.
 
#45 · (Edited)
and the answer is bad news for you farkas.
Maybe I missed it, but the question I answered was in regard to how the narrowband can possibly "self learn".

Regarding open loop in the >90% throttle input. This is well documented, and to an extent, it does still apply fuel trims in those sections indirectly. Again, I'm going to direct you to the EFIE thread to see the logged real world data that surprised everyone when we found the >90% section was populated in an interpolation fashion from the new "learned" data from the <80% sections. Now, the speculation upon how that algorithm is designed to work with the base data in >90% area is really unknown. Others have been able to note that the algorithm for interpolation in that open loop section is only sufficient within a small percentage of fueling requirements. But from a rideability standpoint, this algorithm is entirely necessary because the switch from open to close and vice-versa can have very undesirable results. I have even read a few people have had a common point of disappointment with a specific fuel tuner because of it's abrupt change from closed to open loop at a notable RPM.

So, what Finbro/Yuminashi claims to have done is scale the >90% base map up, taking into consideration higher flow rate injectors and the 13.2 target in closed loop. This goes back to your statement about them stating "make sure you use X injector when installing ECU".

It's really a best attempt to make a one-size-fits-all ECU. is it ideal for peak performance? not at all. does it actually work? yes, for most people it's fine.
 
#49 · (Edited)
they leave it at stoich in closed loop (because of course), then in open loop they suggest fine tuning with one of their injector options, because the ecu is running without feedback. obviously they've fattened the map in open loop, but the rest is up to the injector. (They make them in small increments so you can select one that will get your open loop tuning close to 13.2:1, the target they favor). the ecu doesn't do it, the injector does.

yuminashi2 by William Glancy, on Flickr
 
#51 ·
i know exactly what you're saying, but what you are saying is NOT what is happening. because operating a control loop with unstable feedback is not feasible.


your little suggestion about tuning to 0.7v feedback followed by your graph showing how UNSTABLE the AFR can be at 0.7v only solidifies this.
 
#58 ·
..because operating a control loop with unstable feedback is not feasible.
...in your opinion.

your little suggestion about tuning to 0.7v feedback followed by your graph showing how UNSTABLE the AFR can be at 0.7v only solidifies this.
Have you actually SEEN what narrowband voltages look like in real time?

There are FOUR notable tools just within the Grom world that are able to quantifiably demonstrate fueling beyond your theory of stoich, and you still want to argue that it's, "not feasible"? These aren't coincidences and neither are the HUNDREDS of other tools in the automotive/powersport industry that do the very thing you're so adamantly denying.

Get a grip, Will. You're doing nobody favors being this dense. Nobody.
 
#53 ·
not at all. i'm just wrapping up a big winter build.


but it's important that information spread is accurate, and when it's not it should be corrected.


you just can't handle the fact that someone isn't putting you on a pedestal. you old codgers are all the same. think you're amazing because you've been around for a while.
 
#59 ·
it's pretty obvious what narrowband voltages look like... they go high, then they go low, then they go high, then they go low.

can you show us a tuner that is using the narrowband to tune to a very specific AFR that is not in the very close neighborhood to the switching point on the o2? because you tried to suggest this is what yuminashi is doing, but it's not. and it's not what the oem grom ecu is doing either.
 
#61 ·
So what's the closed loop setting on the hoedar with the stock o2? Do they have an open loop segment like all other narrowband systems? I bet it's a similar story to the others using narrowband.

They claim to use stock OR wideband. Or do I have to ask them for the story like Yuminashi.
 
#63 ·
Doood, I wish I had the time to delve into the research so deeply. Really man, thanks for your diligence. You Grom speedsters are crucial to my build.

I spent ALL day at our local sandbar. Sunny n 90. Pimp Chimp sat in the shade.

Btw, I’m getting the Aracer RC1, tell me about the iMode real time controller. Does it really do what it says it does? I think it loads 8 maps, right? Get a Monkey!!!
 

Attachments

This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top