2007 Street and a single backfire

[ticktock]

have a 2007 Street Glide with a Arlen Ness Big Sucker and a Rinehart True Dual Exhaust system. It has been dyno tuned with a SERT twice using 159PZ003-A0.MT6 and 159PZ004-A1.MT6 respectively by two different tuners. I have also ran other calibrations even the ones delivered on the CD. Here is the problem.

When I am crusing along at two RPM positions, I have a single backfire when I left off of the throttle and go straight to zero throttle position. The RPM's are around 2600 and 2800. I have logged the data and cannot see any abnormalities that the o2 integrators aren't correcting. The backfire does not have each and every time, rather when the map load goes down below 15. I see in the data during this time that the injectors switch off for a few frames and the spark advance goes to 12 in both cylinders. The interesting this is that there is no 12 at 15 kPa in the spark advance tables anywhere in the rpm range or even in that column at all.

In the AFR table I have 14.6 at 27 - 85 and 13.5 at 20 kPa. It gets worse if the the afr is 12.5 at 20 kPa.

I have looked over the step tests that the tuner did and through all of the RPM ranges it appears to be pretty close to 13.0 where he tuned it.

When the bike gets hotter it tends to have this issue more. Every once in a while if I downshift from 5th to 4th gear I may get a pop as the rpms pass through 2500-2600 rpms.

I have replaced the TP sensor and that did not do any good and I don't want to continue to throw parts at it.

I have also heard rumors that the issue may ben when the fuel injectors kick back on again but don't know what the work around is.

I know that the backfire may be caused by a lean condition. For the life of me I can't find one. I have been through taking the exhaust on and off checking all areas for intake leaks etc. I have also had two harley shops do the same and neither of them could find anything either. In addition I have smoothed the map between 0,2,5,7 positions where the bike was not tuned. And of course I have adjusted the decel enleanment table up and down the range either adding or removing deceleration fuel.

Any ideas?

[Mike]

Injector on/off time would be my best guess. Race tuner does not give us access to the table or controller for this. I does vary from map to map and sometimes a map that has the right stratigy for decel pop or backfires can be used as a base form which to cut and paste the other tables into. Our problem is that we've not been able to identify whcih maps these may be. Another possibility could be leaking injectors leaving residual fuel in the intake causing the backfire when fuel from throttle transistions is added. Last, poor intake seals could also be a possibility as the added air may not be fully corrected for by the IAC motor, again during transistions.
Mike

[ticktock]

Thanks for the reply!

I have had the injectors checked for leaking as well as the intake seals. I looked at the injectors myself by holding the throttle open and just turning the ignition switch on (multiple times) to see if there was any spray - there was none that I could detect. I suppose either of these could be some microscopic pinhole type of condition but I doubt that it would just manifest itself in two rpm ranges only. But with that said, I will leave that open as still a possibility. The backfire also switches cylinders - as soon as I think I have it worked out of one it moves to the other!

Regarding the strategy for decel pop - are you speaking of spark advance or the ve tables?


I did notice that when I put 13.5 in the 20 kPa it makes it harder to occur. I was thinking that perhaps I take it out of closed loop at 27 kPa and also use 13.5 in that area as well. Is this a strategy that sounds like it would work? Are there any downfalls with this approach?

[ticktock]

Well I tried the 13.5 and 13.8 in the 27 kPa area and that did not seem to make any difference. When you get time can you explain adaptive fuel to me and how it actually works with the ECM?

[Mike]

The injector off times are governed by a table, or other method, that is not accessable via SERT. The AFV is a shift in the VE tables that allows a certain amount of "memory" to occure based on corrections required by the O2's. I.E., if the O2's are constantly shifting to the rich side the majority of time, then the computer will shift the entire VE table that direction by up to 10%. This shift will then be saved thereby reducing the amount the O2's have to shift on a constant basis and when the ignition is cycled. O2 corrections only occure in real time and do not permanently change VE values and so these corrections take a certain amount of time to "catch up" each time the key is cycled. This may not be an entirely accurate description of the actual method used but is the basic philosophy. This is why the AFV should be reset to 100% prior to tuning, otherwise the tuning values will be incorrect by the amount of AFV shift should it be reset at some later time. This can only be done via digital technician, at this time.
Mike

[ticktock]

That is excellent information and really helps to fill in a lot of gaps I had in understanding the impact to open loop.

Is AFV constantly learning? Say that I a had mine reset right now, would it relearn from scratch again?

Also, in my particular instance I was thinking of switching the baffles to performance baffles. BUB says that these are a little more restrictive and have a bit more backpressure that will give the bike more torque. I was thinking that if I swapped them perhaps it would disrupt the combination that is causing the backfire. Do you think this would help? As I indicated earlier I have had the bike tuned and I was wondering if the ECM could handle the baffle switch without having to have another tune done - do you have any thoughts on this?

Again, thanks. It is nice to have a place where someone is knowledgable. I can't get this kind of information from my dealer.

[Mike]

Thank you for the kind words. You could do the baffle change as a temporary test but retuning should be done if permanently changed. I do not know what affect this may have on your decel pop, it may get worse but at least you could infer a cause and affect.

The AFV is constantly being adjusted but over a much longer time vs the O2's. It will readjust if reset to 100%. In theory, if the tune is correct the O2 and AFV shifts would be minimal.
Mike

[ticktock]

OK, I give up! I changed the baffles to the performance baffles and after the engine got hot it started again with the decel pop in the same places. It was better than before (prbly since I can't hear it because of the new rumble) but it is still there. I am pretty sure that I had an issue with the stock exhaust but that was so long ago I can't remember for sure. Do you have any suggestions at all that I could check? I am pretty sure that it is not an exhaust or tuning problem. Any possibility that the ECM could be defective and/or just not calculating properly?

To date I have:
1.) Mapped with 159pz003 - Custom Dyno Tuned twice
2.) ReMapped with 159pz004 - Custom Dyno Tuned
3.) Replaced the TPS sensor
4.) Crank Position Sensor was replaced
5.) Installed Rinehart Performance baffles
6.) Tested with the calibration files as delivered on the CD
7.) Checked for intake leaks by spraying water on the throttle body etc
8.) Checked for injector leaks by cycling the ignition switch several times and looking for fuel/spray inside the throttle body.
9.) Checked for exhaust leaks
10.) Disassembled and reassembled the entire exhaust system and replaced the gaskets
11.) Sealed the mufflers with high temp copper silicon.

I'm out of ideas and am not sure if this is a tuning problem or something else. Any help would be MUCH appreciated - I'm even willing to entertain ideas that are "out of the box"!

As a note - the bike runs fine otherwise...

[Mike]

As discussed earlier, it may be the injectors coming back on after shutting off completely from closed throttle. Unfortunately, we do not have a means to change this part of the ECM programming. This single pop has been seen on other bikes and we've not been successful in eleminating it. I believe your are correct that it has likely always been there but the stock mufflers do a good job masking it. I doubt that it is an ECM problem. just the nature of the programming. Wish I could be of more help. Best advise I could give would be to try different SERT maps (from the CD). If you find one that seems to solve the issue, then cut and paste your VE, AFR and timing tables into that cal. Because of the time involved this is not something we could do for our customers but may be something for an individual to try on their own.
Mike

[ticktock]

What would be safe ones to try on my setup?

[ticktock]

Also, do you feel like the fuel injectors are controlled within the calibration file itself in an area where we can't see or edit? If so, should I be copying every single table from my current calibration file into another one?

[Mike]

The injector off times are dictated by the calibration but the table that does so (within the cal) is not accessable. This is why I'm recommending trying different cals because not all seem to use the same values. I.E., some cals turn the injectors off over a wider range than others and some do not turn them back off from closed to open to closed again while others do. So there is some difference in the programming of these tables or strategies from cal to cal. Just for test purposes, any cal for your motor size would be ok for short tests. I would only paste over the VE, AFR and maybe timing tables (timing may play a part in the decel pop so we might try this later). Also tuning constants should be checked to be sure they match your original cal and corrected if needed.
Mike

[ticktock]

I don't mean to be dense about this, but if I took 159AE003-A1.MT6 which is not for touring but for shorty duals with a race exhaust would that be something to try?

[Mike]

Could be, the VE table (when pasted over) will correct for the base map differences. Another thought came to mind last night, you may try adjusting your Accel enrichment and Decel Enleanment tables, it may be a balance of those two as the solution. This along with setting the 20% AFR values either higher or lower (we've heard both have solved this issue but we are tending towards minimizing that column from 1250rpm and up). So now it's a balance of three tables, you can see why it's been difficult to resolve. No one set of values works for all bikes.
Mike

[ticktock]

I have worked with the accel and decel tables, but neither seem to be the solution. The accel never really comes into play according to my data recordings when I let off of the throttle. I do see some decel fuel going in and have worked with that a bit but I don't see/feel anything unless I go to one extreme or the other and when that happens I get a pop on either end. Regarding the AFR in 20, I currently have set to 13.5 (up from 12.5) and that does help some but I was worried that if I increased that too much then I would not have any bottom end performance. When you say minimizing, what exactly are you referring to - I am thinking that my approach my not be aggressive enough?

[Mike]

I agree, you may need to get more aggressive. By minimum, I mean the min. values the table will allow to be entered. If going to the higher AFR's appears to be helping, I would continue in that direction. As I indicated earlier, we've had reports from some that going to higher AFR values in the 20 column only has helped with the pop, while others have told us that reducing them helps. Do not go above 14.5 as a 14.6 value will cause it to go into closed loop. I would hve no problem going to 14.5 from 1250 rpm and up as a test, then back down from there until the pop returns. You should not see any change in low rpm performance as even small throttle/load changes go to 40+ typically.
Mike

[ticktock]

Do you know what the difference would be between calibrations
159pz004-a1.mt6 and 159qd001-a1.mt6 other than one is marked for international (qd)? The tables all appear to be exactly the same.

Would an international model be more stringent on pollution etc or less than our epa?

[ticktock]

I finally tried your advice and loaded 159ae004-01.mt6 with all of my table data from 159pz004-01 and the bike never popped or hesitated at all. This calibration seems to work for me at the moment (only humidity will tell). Is this safe to continue running or could I risk engine damage by running a calibration that was not originally built for touring models? I copied every single table over.

[Mike]

If you copied over the VE tables from your known good map, then you should be fine to run.
Mike

[ticktock]

Thanks Mike.

One last question and I think I am done! At the moment I have copied all of the tables over including timing, step table etc. So in a nutshell the 159ae003 map looks identical to 159pz004 on the surface (with my modified/tuned VE values).

I find that this is working fine but was wondering which is better - to run the 159ae003 (shorty duals) map with the timing that came with it and just add my VE values or am I better off with the timing from the 159pz004 (touring) map? At the end of the day - does it matter for rideability, fuel economy, smoothness etc?

[Mike]

I would be inclined to use the touring timing tables. Typically, the higher the timing values your motor can stand without a lot of detonation, the better your fuel milage. Lazier timing tends to be smoother but also elevates exhaust pipe temperatures.
Mike