I had a SERT installed on my '08 Ultra with stage 1 setup and it runs great. I have seen where people turnOFF the knock sensor in the ECM tuning constants function of a SERT when doing dyno runs so the sensor will not effect the timing. If the knock sensor is set toON for regular bike riding and if it detects knocking, will the retarded timing be restored as soon as the knock goes awayor does the ECM back off the timing and it stays retarded? I can't find a good definition of the ion sensing system HD uses.

 

My understanding of the Ion Knocking sensor system is that the timing is retarded only if knock is detected. It does not re-write any information to the ECM.

The closed loop function if the ECM is a narrow band setup. It only has a small band within which it can change the fuel delivery, and only does so at idle and part throttle cruising. When you grab more wick, it reverts back to the map loaded into the ECM.

As for weather or not to run the bike with the knock Sensor on or off, I ONLY turn off the knock sensor if I'm tuning the bike. You have to know when it's pinging, in order to know when you've added too much timing. You can also get knock if the bike is running too lean, but since I run wideband 02 logging stuff, I can measure that directly.

Even on the quarter mile, I leave the knock sensors on. The LAST thing I need is for detonation to ruin my motor.

Hope this helps.

RJ

 

Thanks for the input, RJ. On occasion my new bike acted like the timing was a bit retarded before I added the SERT: it ran very smoothly and very quiet but seemed to have lost some pep. I wondered if the knock sensor had told the ECM to back off the timing and that setting stayed set that way. I never saw any codes posted, however.

 

Funny you mention it, my buddy got a 2007 Street Glide. I thought it was a bit flat in spots, and had some hesitation when you goosed the throttle. I think it has to do with EPA regulations, more than actual performance tuning.

 

That is my understanding as well.

This is a common misconception. The O2 sensors are narrow-band, yes. But the way the ECU uses them, they only play a small part in the scheme of things. The way I understand it, the total range of the O2 sensors, as far as what they can authoritatively report, is ±3% (more or less) but the ECU has a range of ±25% fuel delivery with which to reign the O2 sensors back into their range.

Also, the way I understand it, pretty much everything below 40% throttle is closed-loop control. (Unless, perhaps, you get to 40% directly from 2% under load. That may call for a little "out of the box" type of "thinking" on the part of the ECU, and may well be the major reason for the fly-by-wire throttle on the '08 tourers.

 

Interesting points glens.

I didn't know the ECM would change upto 25% based on the 02 sensors. Do you know if the ECM automatically goes "rich" if the 02 sensors fail? I understand this happens with Cars, and was wondering if the Dephi units have the same failsafe.

I'm also of the understanding that the closed loop bias can be modified using the SERT. I do not have any experience with using the SERT on the 2008's so I can't say for sure. Do you happen to know if this is true?

Thanks.

RJ

 

I pretty much agree with RJ, I would very strongly suggest you retain the knock sensor in its enabled mode.

I do differ slightly, when I tune. I leave the knock sensor running. When I do a SERT data run I look at the knock sensor readings which give me an idea where things need work. I dont get too worked up where the knock sensor gives a reading below 3 over a very short time period.

WB

 

Cars tend to have wide band sensors. With the HD narrow band sensor if it fails it is already working in susch a small margin that no serious damage should be caused.

 

I haven't had any car or truck newer than a 2000 model, and frankly haven't looked even at it to see what type of O2 sensor(s) it uses. I do know for a fact that the stuff I had from the mid-'80s only had narrow-band sensors (though they were three-wire units with heaters to become usable almost immediately, not relying on the exhaust gas to bring them up to temperature as the units on our H-Ds do).

It's pretty much always been that in a fail-safe mode the ECU will ensure a rich-enough mixture to safely "limp" home. With our rather primitive Delphi units that might not be the actual case if there have been engine modifications without informing the ECU what the actual volumetric efficiency values have become, however. Not to say the pistons would melt under judicious operating conditions or anything...

I don't know what our Delphi failsafe AFRs are set to. Without some authoritative documentation, the next best way to know would be to simply yank the O2 sensors out of the loop and sniff the exhaust. I'm guessing it would be something like 13.5:1, maybe as low as 12:1. Anyone know?

 

"I'm also of the understanding that the closed loop bias can be modified using the SERT. I do not have any experience with using the SERT on the 2008's so I can't say for sure. Do you happen to know if this is true?"

Yes. My tuner maxed out the CLB tables which should bring the A/F ratio to around 14.1. Even with the richer A/Fratio, my mileage improved since I had a 103" Stage 1 download prior to the SERT.