Underpowered/too lean?

[rtrofimovich]

Hey all! So my journey with Buells continues! The bike is treating me well and I am having fun getting used to its riding characteristics coming from a 2010 zx6r. I've been reading all over of people pulling power wheelies with ease through 1st and 2nd. My xb pulls pretty nice, but its definitely not doing any power wheelies. The previous owner had installed a Hawk exhaust and K&N filter which is probably causing the bike to run pretty lean and probably consequently under powered. Now, the only "tuning" I've ever really messed with was installing power commander modules and race maps onto 600cc bikes, which has always had pretty good results. I've been reading tuning threads on the xb forum for a few days now and have been exposed to a lot of great information regarding wide band O2 sensors, O2 voltages, 14.7 AFR, ECM corrections, etc. Also, I've read up on burning a custom eeprom with ECMDroid. This is an option I've been looking at doing through the Buelltooth dongle, but I also read that the ECM will try to correct the values to achieve 14.7 AFR again so its basically worthless..

Question is, what's my best option to richen up the bike and retrieve some potential power gains? I know the Hawk isn't a "race"muffler, but I'm sure the higher flows of the hawk and filter can be exploited with a "proper" tune.

Sorry for adding to the number of tuning threads, but everything I read just left me wondering more what the best option would be for power gain vs. ease of application.

Thanks!

 

[njloco]

I'm not very knowledgeable with tuning, that being said, did you use the Buelltooth to take it out of closed loop ?

 

[rtrofimovich]

No, it's still reading the O2 sensor

 

[Cooter]

Putting a better tune onto your bikes ECM will NOT automatically go back to original. Seems like a lot of effort for people to do on these 1000's of threads if that was true, right?

To put it very simply:
The ECM uses the O2 sensor to adjust the fuel mix to stoic (14.7:1 AFR) in closed loop only. Closed loop is "part throttle cruise". You want that. It is the most efficient, clean burning, cool, way for the bike to run. So when you are just cruising around, theres no need to dump a bunch of gas in it.

You want more power on acceleration. That is called "open loop" operation. When you twist the throttle, the ECM goes into Open Loop (OL) and no longer reads from the O2 sensor to adjust the fuel mix to stoic. In OL, the ECM reads off the "fuel map" that was programmed at the factory using stock parts.

Once you have modded the stock engine to breath better with a better air filter and exhaust, that stock factory fuel map needs to be adjusted (enriched) to compensate for what you did. Installing a new fuel map (from a place like Buelltooth) is what makes this change. It is the best option for you and the end result of what the Power Commanders try to do, but without all the restrictions they have.

If you are unsure of the process that is laid out very simply on the Buelltooth site, don't forget you have the option of IDSpeed.com

Even a tired XB9 will power wheelie in 1st:eagerness:

 

[rtrofimovich]

Perfect, that's exactly what I was unsure about! There was a lot of information I read through that didn't make 100% sense to me and that's essentially the biggest question I had. Thanks for that info Cooter!!

 

[d_adams]

Side note, you have no idea what it's actually doing and you're taking someone else's word for it that it's running at the correct air/fuel ratio *UNLESS* you go to the trouble of installing a wideband sensor (or 2, my preferred method) and have either the capability of logging the info/data for playback and adjustment or a live display showing it. Want it to run right? Wideband or a 4 (or 5) gas analyzer. Pretty much everything else is a wild a s s guess.

And the option of IDSpeed.com is actually www.idspd.com

 

[rtrofimovich]

I agree. But if there is a performance gain and the bike still runs well, then I would assume the AFR should be better than before, correct?

 

[marcel]

@d_adams, Is a wideband O2 sensor also better if you are logging a ride in ECMdroid to adjust the map? and If so, is a Wideband sensor plug and play? and if that is so, where do I find one? (Buell XB9s 2003)

 

[Cooter]

Yes, tremendously better. A NB sensor only knows rich/lean. A WB sensor knows how much rich/lean. Since you have the O2 disabled (OL operation only during datalogging) so the ECM won't self-adjust, the datalog you put into MLV will be much more detailed with a WB O2 sensor.
MLV will know how much to adjust that part of the map. Not just up or down.

The WB sensor screws into the same hole the NB came from and the controller should fit anywhere under the seat. You can use the WB sensor for normal operation after datalogging because the controller should have a NB emulator built in.

The internet

(all of these answers are very simplified FYI)

 

[gunter]

I disagree. The ECM does not care about the real deviation from stoich in CL, it cares about what it thinks the deviation is. No one knows, how this is computed from the NB O2 voltage, and if the ECM indicates a 10% offset, it's not said, that the mixture is 10% off.

 

[rchuff]

^^^ Spoken from the tuning God himself!! :up::up:

 

[Joe7Bros]

^^^ Spoken from the tuning God himself!! :up::up:

rchuff, just noticed you're in Willow Grove; I'm on Pine Rd in Philly; Montco (Huntingdon Valley) is the other side of the street

 

[rchuff]

^^^Grew up in Huntingdon Valley on pine Rd.

 

[Joe7Bros]

We'll have to arrange a play date! I'm near Pine and Alburger

 

[Cooter]

Hey Gunter, thank you for dropping in!
I’m not sure what you’re trying to explain?
I wasn’t suggesting the WB sensor can tell the stock ECM how much to adjust (I know the ECM can only read a NB signal in operation, the reason for the NB emulator from a WB sensor). I was explaining data logging with a WB sensor is better because the data you get to input in MLV is more accurate. That MLV would know how much to adjust the map it spits out vs. only knowing toadjust (and not by how much).

Do I have that right?

 

[rchuff]

It sounds like if you real don't know what you are talking about maybe you shouldn't be talking about it?

 

[Cooter]

rchuff: You talking to me?

 

[gunter]

Hey Gunter, thank you for dropping in!
I was explaining data logging with a WB sensor is better because the data you get to input in MLV is more accurate.

With 6-7 data queries per second you are missing 50 - 90% of all ignition cycles. Taking this into account, swapping an qualitative analysis with a quantitative one will not really increase accuracy.

The second point is that many people forget the goal of tuning the ECM: it's not to get a "good" map, the goal is to stop AFV from dangling around. AFV is based on EGO correction. EGO correction works perfectly well in the ECM, and a WB O2 will show an average lambda of 1.0 in closed loop, even with a bad map. But EGO correction is the key. A wildly floating EGO correction will lead to a wildly floating AFV, so limiting EGO correction is the main task to get a stable AFV. EGO correction is what the ECM assumes the map is off stoichiometry, however it is computed. The ECM does not really care about the mixture, except that it shall not get (permanently) lean or rich, but only from a qualitative point of view.

 

[Cooter]

I see the point you are making and it's valid for sure. But not the part I was trying to explain to the OP.
Of course, trying to tune CL with a WB won't do much. It's only looking for stoic and a NB does that fine, in CL.
I'm talking about datalogging OL operation with a WB to get the OL map closer, faster.*

You're talking the most important part. Getting the OL map perfect is the goal, yes. If your EGO correction doesn't have to correct, you'll be at perfect fuel inputs instantaneously (and where it matters). The CL map will always search for stoic (as it should) and clean, safe, running but at cruise conditions only. As long as it's not "always lean" or "always rich" you're good.


* To your first point, You are saying that more data points is better than less-but-better data? I didn't know logging WB vs. NB slowed that down! The more you know.....

 

[gunter]

You are fully right, if the ECM is running in OL only, there are no corrective actions applied, hence no data available except a lean - rich transition from the NB O2 sensor.

My impression was the question aimed at if it's preferable to force the ECM in OL, next tune the maps using a WB O2, then go back to CL operations afterwards. I do not see a real advantage in such a procedure. There's a reason, WB O2 logging was added to EcmSpy, because we ran all these tests a decade ago, when trying to find out an easy procedure to adjust the maps. In the end, using the dynamic maps lead to faster results than anything else and tuning could be done in (halfways) real life rides.

> To your first point, You are saying that more data points is better than less-but-better data?

Well, it depends, IMO. The WB O2 controller is spitting out data much faster than the ECM, so they somehow must be correlated with each other. The ECM knows exactly when to evaluate NB O2 voltage to get the information it requires. Running the engine at 50 Hz will provide 40 ms between each ignition. I do not remember how many data packages are send during this period, but let it be only 2 - which one indicates the exhaust gas oxygen level correctly? This is just the "inter-ignition" gap. The ECM needs about 130 ms to emit one data packet, good for 3 ignition cycles. Which of these three is shown in the ECM data, and to which do the WB O2 data apply? To make things even worse, riding is a highly volatile process, which then adds more uncertaincies as the operating point is constantly changing. Sure, this all can be coped with on an EC dyno, but this is not what most people are willing to pay. Therefore, more digits do not necessarily mean higher accuracy, especially when the time frames are differing. It's just swapping one problem with another.