Stall/Stumble fix

[mag00]

I finally got around to monkeying with the Strada. It was fairly decent until the firmware upgrade and then it was not fun to drive anymore. My Terra has been fine since the upgrade and add ons.

Strada is bone stock.

What I did is shim the throttle stop .020 and reset the ecu with the gs911 then remove the shim.

It seemed to take a little to adapt to the new setting, but today (2nd short ride) felt so much better. I think after all the adaptions it will be much better.

Idles fine.


How I did it. COLD. Took some static readings.

Throttle valve angle 1.61
Throttle valve angel shimmed 11.50

Valve position 0
Valve position shimmed .78%

idle actuator 130 both shimmed and not

After reset using .040 shim, still cold, not running.

Throttle valve angle .22
Throttle valve angel shimmed 1.5

Valve position 0
Valve position shimmed 0

Adaption value:
idle speed 1.5
Full load 85.5

Running

Idle Actuator Position 70
Throttle angle .11
-----------------------------------------------------------------------------------------
I'm going to run it a bit more, and since it is more tolerable, should't be too much trouble, except I need to rack up miles on my new Duck 62.

Then I'll plug it back in and add the new readings.

I tried running it with the .020 shim and that did not work. At first, it idled OK, but then it would not idle, so I removed the shim. Maybe if I had left the shim in it would have been fine. If the Strada starts acting up, I may try that procedure. Reset with shim and leave shimmed.

I think a new set of Brisk and a single can would perk this baby up just fine.

 

[WayneC]

Shimming the throttle is completely irrelevant to the process and the TPS is constantly recalibrated so is independent of the adaptions set and reset by the diagnostics process. With your Strada it sounds like the Dealer did not rebuild adaptions after the firmware update hence the poor running, that has also happened here

 

[mag00]

I'm going to be working on this for awhile now. I have it on the back porch so I have real easy access. Tomorrow I will hook up to the GS911 and take readings.

Then I will run it. First thing I am looking for is easier starting. This seems to have helped so far, but I need to get some miles on it to stabilize any adaptions that may happen.

On the Terra, after the update, it took 2 rides to stabilize. I am at 1 real short run and one short ride, so it may not be there yet.

The Terra does not start as good as pre software update as is the same with the Strada. However, after the first short ride, the next start up without shim, it started fine.

I don't know why I wrote .040, as it was .020 for the shim. I need to fix that in the previous post. I'm trying to document as thoroughly as I can. Eventually I'll monkey with the features of the GS911 that capture and record.

 

[WayneC]

Mag00 with all due respect, you are wasting your time with shims, the key is the adaption reset and rebuild, preferably from cold so a full table is rebuilt, you still seem to confuse adaption tables and trims.

Use the real time logging of GS911 to observe operation of the throttle, the TPS reads from bottom to wiper and wiper to top, from there it is able to determine position and angle and does so constantly. Use a power supply when doing the diags so you have stable voltages and dont flatten the battery

 

[PaulC]

With your Strada it sounds like the Dealer did not rebuild adaptions after the firmware update hence the poor running, that has also happened here

I have reset adaptions and waited for the fan to activate a few times using the GS911. While this helps with running it does not help with starting.
IMO, Both versions (or more?) of firmware have their separate issues. And when the MOSS diagnostics system was still active this issue was more apparent.
Interesting thread


Edit: Cold start issue only

 

[WayneC]

There are some 200,000 650 Rotax engine machines out there using the same throttle body, TPS and Idle Actuator, the throttle body part number changes according to which Idle Actuator is fitted and whether the pipe for the US only Tank Vent Valve is fitted.

In addition to that the TPS is used on the cars, including mine, there are probably 1,000,000 of them out there it was used widely in 3 Series model variants as well as used in motorsport. You can read more on the TPS in the Bosch Motorsport catalogues

The stumble is elsewhere in the firmware, stalling can be caused by poor adaption tables but dont combine the two together it is a mistake

 

[mag00]

Sorry to have to disagree, but my Terra runs fine with the same software, so problem has to be somewhere else.

All bikes and engines are not created equal, thus a screw on the throttle body to set the butterfly. I'm just taking it a different procedure to start with because I can. No shop would do what I am attempting. I am also documenting the findings. This is not arbitrary, this is a quantifiable test. No shop is going to spend the time doing this.

This may or may not "fix" the issue, but so far is looking very promising. My Strada is ridable again after a calibration with shim, then removing the shim. Initially I would have liked to set the adjustment screw and left it, but that is hard to get to, and installing a shim is accessible without any disassembly. I am photo documenting as well. I didn't run it with the shim, because I found it to be novel to calibrate with it and then remove it to run it, and wanted to start there with the testing.

I imagine at a shop they would have to set the screw, calibrate and leave it. That would be the procedure authorized by manufacturer.

 

[mag00]

Mag00 with all due respect, you are wasting your time with shims, the key is the adaption reset and rebuild, preferably from cold so a full table is rebuilt, you still seem to confuse adaption tables and trims.

Use the real time logging of GS911 to observe operation of the throttle, the TPS reads from bottom to wiper and wiper to top, from there it is able to determine position and angle and does so constantly. Use a power supply when doing the diags so you have stable voltages and dont flatten the battery

I am not wasting my time. I am getting real world results under real world conditions. I have left it to the engineers and manufacturer to solve this long enough. I did what I could to get the gs911 functional for the Terra, and do appreciate all the help you have provided in the project. Now that I have the tool, it is time for some good ole fashioned troubleshooting and testing to rule the day.

I do need to figure out the logging ability of the gs911, the more I use the tool, the better I get, so in due time. I can better share results too.

My first theory of initial setup of the Throttle Body is what I am going after first. There is a screw to adjust the throttle angle. I do not have the breather off, so I cannot measure the actual opening as it is set currently. The TPS does not know this either, and neither does the ECU and any tables.

I noticed on my spare TB that you can turn the screw 1/4 turn before the tps registers change. That is a significant amount of throttle opening not accounted for and since the off idle stumbles happen there, that is the first place to look for improvements. Our TPS does not have slotted holes for adjusting. Might be nice to back the tps up to a point where it notices the change instantly. Some bikes it may be setup proper.

Since I do have 2 bikes, both unique in there quirks, I have a chance to observe.

My Terra, stumbled bad with old firmware, the Strada ran pretty darn good, minor stumble, no stalling.

The Terra got the update, and it stumble the first ride and then has been fine, with only minor/very minor off idle hesitations.

The Strada got the update and started to stumble like the Terra did.

Theory, initial setup of the bike is different from bike to bike. What do I have access to to change? Throttle butterfly position and calibration. It is not in the adaptions.

So no, I do not feel I am wasting my time. Quite contrary, I think everyone else is wasting their time. but not really, because it is all information that can be parsed.

 

[mag00]

I have reset adaptions and waited for the fan to activate a few times using the GS911. While this helps with running it does not help with starting.
IMO, Both versions (or more?) of firmware have their separate issues. And when the MOSS diagnostics system was still active this issue was more apparent.
Interesting thread

Nailed it... cold start has nothing to do with nothing. It runs off of the default settings. Nothing to adapt other than sense the temp and pressure and use the pre programed settings for that condition. The IAC position during "key on" will matter. That will draw on calibration settings. So far, so good.

 

[WayneC]

Mag00, the adaption reset sorted the Strada, not playing with shims, you are looking for a mechanical solution to an electronic problem

You can use the real time capability of the GS911 to look at the operation of the TPS and it will show you position and angle

 

[mag00]

After a few miles in town, it started acting up real bad, worse than it ever was. Stalling not starting, running ragged etc. Definitely deteriorated as the adaptions took hold. Came home and reset it HOT, and it is better. Just reset all adaptions and calibrated and took off again. No special treatment. Took it for a longer ride and it stuck, with the occasional stutter off idle, so better than before still worse than before update. It's drivable at least. I will find out more tomorrow, probably take it exploring the desert.

I need to make a special tool to adjust the butterfly and go from there.

After I did the dealer firmware update I did the full reset and calibration.

Testing before running today. Throttle valve angle .22
Valve angle 0
Actuator 86

It did not start until I cracked the throttle, so the hard starting is still there. I just went the wrong direction with adjusting I guess.

After starting, Idle actuator changed with engine temp.

30º c actuator 69
50º c actuator 59
76º c actuator 53

 

[WayneC]

Mag00, a couple of questions

Just reset all adaptions and calibrated and took off again. No special treatment.

After I did the dealer firmware update I did the full reset and calibration.

What do you mean by "calibrated" ?

It did not start until I cracked the throttle, so the hard starting is still there.

What mods do you have on the Strada ?

I need to make a special tool to adjust the butterfly and go from there.
I just went the wrong direction with adjusting I guess.

After starting, Idle actuator changed with engine temp.

The BMSE knows the butterfly position from total resistance and wiper resistance measured minimum to wiper and wiper to maximum position, butterfly should be at minimum with a little free play at start with air controlled by Idle Actuator which will change as engine warms up

 

[mag00]

Mag00, a couple of questions


What do you mean by "calibrated" ? Just the options in the GS911


What mods do you have on the Strada ? Bone Stock



The BMSE knows the butterfly position from total resistance and wiper resistance measured minimum to wiper and wiper to maximum position, butterfly should be at minimum with a little free play at start with air controlled by Idle Actuator which will change as engine warms up

The ECU only knows where you tell it is at during calibration. There will be manufacturing discrepancies in the parts, so there is adjustments that can be made. What is the clearance of the throttle valve to TB wall supposed to be, when in idle position?

(1600÷2)650cc = airflow per minute. How much goes past the valve vs through the valve is changed by the AIC circuit. AIC is controlled by ECU. Initial throttle setting is by the screw opposite end of the TPS. TPS will tell ECU what angle of throttle valve is through full motion.

Idle screw can affect what the ecu tells aic, so closing the throttle valve by backing out the screw, will force the ecu to tell the aic to let more air through. Also, turning the screw in, thus increasing the air going through the throttle valve, will force the ecu to tell the aic to close up some.

That column of air flowing over the valve has different response times than if it was bypassing through the aic circuit.

Thus adjusting that screw, then resetting and calibrating will have an effect of how much air flows through either the aic circuit or over the throttle valve. This in turn can affect that initial off idle performance.

I have put off doing the testing until now. Did not want to mess up my Terra that ran great. Resetting things is easy enough now, so changing it up and trying is no big deal, with the exception of reaching the idle screw.

I have not had any trouble codes from this yet.

 

[WayneC]

Mag00, you have not answered the question re calibration, have you idled the machine until fan cuts in following the adaption reset to rebuild the tables ?

 

[mag00]

Mag00, you have not answered the question re calibration, have you idled the machine until fan cuts in following the adaption reset to rebuild the tables ?

I have done it both ways. Seems resetting when hot and then just getting on and going is the best so far. I will know more next time I ride it. Taking the Ducati Sixty2 out today. May or may not have a friend drive the Strada.

Seems our terminology may not be in sync. I use the service tool and click "calibrate" for those functions when messing with the throttle stop. Any driving after that is just adaptions, so adapting to the way I drive is prudent, and I don't sit in the driveway and let it warm up to fan coming on. Seems to work better this way. I can try it those different ways, but the fastest and best so far is driving up on the back porch, hooking up, resetting and taking off again. That will be my baseline method.

 

[isldtime]

I hate to sound like an old fart but a carb would be so nice on this bike.
I could fine tune my KTM 640 SM with washers under the air box lid and make a huge difference!

 

[WayneC]

Mag00, if you reset adaptions and then dont idle it until the fan kicks in you will give yourself problems, that is the procedure on both cars and bikes including other makes and models, on the Triumphs with Sagem/Keihin ECU's they specify idle for 22 mins, on BMW cars they specify 15 mins. If you do the rebuild when the engine is already hot then you do not get a full table rebuild, the table for the lower temperatures will be either the data that was already there or a default set of data dependent on how the engineers did the reset firmware algorithm.

When you say "any driving after that is just adaptions" what you referring to is trims, not adaptions, the adaptions are a fixed table, trims are variable, the accepted norm is that it can take a tank or two of fuel and running to see the benefit on the trims. The base adaption tables need to be in place first.

The other aspect in all this is you dont have any logging to see what is going on after you do a reset/rebuild and then be able to load it into spreadsheet or graph it. Get the bike set up so you can put the GS911 under the seat to do the real time logging. To make that easier unwrap the wiring harness section with the diags socket back to the frame on the RHS and rewrap with the diags separated so there is some free length in it to put the GS911 either across the frame near the top of the battery or next to the battery if you have a smaller Lithium battery fitted. Others here on CH can tell you how they did it

 

[mag00]

Finally I got it. Something as simple as an O2 sensor. No trouble codes tripped. Must be just bad enough to cause the issue and still run.

The one on the left is what was in and running like crap, the one on the right is a used one I had picked up awhile back and now the bike runs like it did before update. Still has a trace of stumble, but rarely, and that may get better with more miles.

 

[Mark_H]

Finally I got it. Something as simple as an O2 sensor. No trouble codes tripped. Must be just bad enough to cause the issue and still run.

The one on the left is what was in and running like crap, the one on the right is a used one I had picked up awhile back and now the bike runs like it did before update. Still has a trace of stumble, but rarely, and that may get better with more miles.

View attachment 74483

My O2 Sensor has been disconnected for the past 3k km, the bike runs like a dream now.