Greetings, having an oilhead hard starting issue

[soco]

All:

Salutations! My name is Reid, based in the mid Atlantic region and I was referred here by Roger04rt. I have been trying to track down a hard coldstart issue and my research led me to find him as one of the most knowledgeable people on the various forums I inhabit (ADVrider and some others). I ended up sending him logs, and he graciously took a look at them and told me that if the problem continued, you guys were a wealth of knowledge and might be able to help me track down what I am missing. I've been working on anything with an engine since I was a wee lad, and have a lot of experience in electronics, anything mechanical, engineering, and various and sundry other technical fields. So if anyone would recommend any troubleshooting I am capable of doing everything down to de-soldering PCBs (I let my buddy do that haha!)

 

I'll try to keep this from being rambling, so let me start with the story so far. I bought the bike (2002 R1150R) after it had been sitting for a number of years. In order to ride it home, I had to keep the throttle stop lever at the first detent because it wouldn't idle. In my quest to solve this, I basically dug in and did all the maintenance one of these needs after 20 years. As I replaced parts, I realized that when the bike was cold (ambient temperature of less than 80) it would struggle to start. If it was sitting in the sun on an august day, it would be better. And if I had ridden it and it was still hot to the touch, it starts in a crank or two. This took me down the rabbit hole of attempting to replace various things that I figured might be contributing to it, and I would want to replace on an older bike I intended to keep forever. I still haven't solved my issue unfortunately. I will add a mostly chronological list under this, to keep it readable.

 

Right now, like I said the bike will struggle to start when its not hot. Once running, it runs all day without a hitch. The tank liner delaminated, so in an effort to prevent it from covering the pump intake sock I threw a bunch of acetone and screws into the tank and agitated it to get all of the large flakes out. I keep it full of gas to prevent/slow any rusting, I don't feel like having the whole thing relined until I can solve this problem. I haven't noticed any subsequent flakes when I have removed the pump housing, and it runs perfectly. Very smooth, no surging. The bike is a little noisy when revving in neutral but based on conversations, i've been told these things are in general not sewing-machine quiet. I have also noticed that if the bike sits for a few days, the fuel pump whine starts at a higher pitch, and then drops almost as if somehow it has air in it and then purges it. But the bike will still have a hard start even without this. Once it does start, it idles low until I give it a few throttle blips. Then it settles in around 1500rpm and as it heats up that rises up above 2000 which is when I pop it off the fast idle detent. (I also did other stuff like brake flushes and such but this list is only going to be related to the engine)

 

List of repairs to date:

New fuel filter, in tank lines, bosch fuel pump

charcoal cannister removal, vacuum caps on throttle body bottoms

upgraded left side chain tensioner

removal of BBS, clean. Balanced with twinmax

fast idle lever tensioning to remove slack

air filter

valve adjustment

odysee 680 kept on NOCO 10 amp tender with AGM feature

several discharge and full charge cycles to recondition battery (as per odysee)

high alternator output rectifier for new batt

new starter

injector test/rebuild by local shop

new plugs

new coil and wires

3w alternator bulb for faster initial charge

AF-XIED to see if that would help (it did not, and I have put the bike back to stock)

 

At this point there is something more nefarious going on. I was convinced it was somehow fuel related because I had to monkey around in the tank and the fuel pump sounding like it had air when sitting, but after doing everything I am beginning to think its not the case. I am concerned if it somehow voltage, because the battery was new in June and I am trying to maintain it properly (never off the tender for more than a day or two). Either way, like I said once I blip the throttle and bring the RPMs up above 1500-1700, it will idle on its own.

 

I am including 3 logs.

 

6-24: Started bike on warm day with idle lever at highest position. It started and stalled out. Second attempt it managed to stay running.

 

6-28: Started bike on hot day with idle lever at highest position. It started successfully.

 

10-22: Started bike on warmish day with idle lever just at the detent (roger told me that should be more than sufficient). Bike did its half start stumble twice, third time I managed to get it to catch and then blipped the throttle to settle it down.

 

If anyone can find a smoking gun here, or come up with something else to check, I will literally order you a pizza of your choice. I really appreciate any help you guys might be able to toss my way. I'm wondering if its something weird like resistance through the key or controls, or even the chain tensioners don't have sufficient pressure until oil circulates and the valves aren't opening right...

 

 

 

6-24 log.csv 6-28 log.csv 10-22 Log.csv

[Lowndes]

Reid,

 

Welcome to BMWST!!  LOTS of info here and people with a lot of experience, and Roger04RT is one of them (his specialty seems to be electronics and the fuel injection/ECU systems I believe).

 

Dirtrider is the universally acknowledged guru of all things BMW and I'm sure will be along directly to get you going.

 

My experience with the AF-XIED is that it takes a while for it to kick in and reach full effect, something about the ECU "averaging" the values over a period of time. Roger04RT may have explained it to you already.  I have the AF-XIED's installed on a '99 R1100S and a '99 R1100RT and they REALLY improved the starting, warm-up, low speed throttle, roll-on throttle, etc.  

 

Sounds like you got a lot done already.  Two items that I think of are the BBS passages and the OEM brake lines.  Spiegler and Galfer are two of many aftermarket sources of "braided stainless" kits (actually PTFE or Teflon tube with braided stainless reinforcement)

 

The tiny passages "to and from" the BBS (it's an air passage around the throttle plate at throttle closed position) are very susseptible to black crud build-up which restricts the idle air.  They can be cleaned with a spray can of carb cleaner, but both ends of the passage have to be CLEAN.  If you already cleaned these thoroughly, disregard this.

 

The OEM brake lines are a well known problem on bikes of this era (15-20 years old).  The DOT4 brake fluid eats them from the inside and they fail with either an "annuerism" type burst or crud settling into the calipers and locking the front wheel.  AMHIK.

 

Stay tuned. 

 

Post some pics, too.  "Without pics - It didn't happen."

[dirtrider]

3 hours ago, soco said:

All:

Salutations! My name is Reid, based in the mid Atlantic region and I was referred here by Roger04rt. I have been trying to track down a hard coldstart issue and my research led me to find him as one of the most knowledgeable people on the various forums I inhabit (ADVrider and some others). I ended up sending him logs, and he graciously took a look at them and told me that if the problem continued, you guys were a wealth of knowledge and might be able to help me track down what I am missing. I've been working on anything with an engine since I was a wee lad, and have a lot of experience in electronics, anything mechanical, engineering, and various and sundry other technical fields. So if anyone would recommend any troubleshooting I am capable of doing everything down to de-soldering PCBs (I let my buddy do that haha!)

 

I'll try to keep this from being rambling, so let me start with the story so far. I bought the bike (2002 R1150R) after it had been sitting for a number of years. In order to ride it home, I had to keep the throttle stop lever at the first detent because it wouldn't idle. In my quest to solve this, I basically dug in and did all the maintenance one of these needs after 20 years. As I replaced parts, I realized that when the bike was cold (ambient temperature of less than 80) it would struggle to start. If it was sitting in the sun on an august day, it would be better. And if I had ridden it and it was still hot to the touch, it starts in a crank or two. This took me down the rabbit hole of attempting to replace various things that I figured might be contributing to it, and I would want to replace on an older bike I intended to keep forever. I still haven't solved my issue unfortunately. I will add a mostly chronological list under this, to keep it readable.

 

Right now, like I said the bike will struggle to start when its not hot. Once running, it runs all day without a hitch. The tank liner delaminated, so in an effort to prevent it from covering the pump intake sock I threw a bunch of acetone and screws into the tank and agitated it to get all of the large flakes out. I keep it full of gas to prevent/slow any rusting, I don't feel like having the whole thing relined until I can solve this problem. I haven't noticed any subsequent flakes when I have removed the pump housing, and it runs perfectly. Very smooth, no surging. The bike is a little noisy when revving in neutral but based on conversations, i've been told these things are in general not sewing-machine quiet. I have also noticed that if the bike sits for a few days, the fuel pump whine starts at a higher pitch, and then drops almost as if somehow it has air in it and then purges it. But the bike will still have a hard start even without this. Once it does start, it idles low until I give it a few throttle blips. Then it settles in around 1500rpm and as it heats up that rises up above 2000 which is when I pop it off the fast idle detent. (I also did other stuff like brake flushes and such but this list is only going to be related to the engine)

 

List of repairs to date:

New fuel filter, in tank lines, bosch fuel pump

charcoal cannister removal, vacuum caps on throttle body bottoms

upgraded left side chain tensioner

removal of BBS, clean. Balanced with twinmax

fast idle lever tensioning to remove slack

air filter

valve adjustment

odysee 680 kept on NOCO 10 amp tender with AGM feature

several discharge and full charge cycles to recondition battery (as per odysee)

high alternator output rectifier for new batt

new starter

injector test/rebuild by local shop

new plugs

new coil and wires

3w alternator bulb for faster initial charge

AF-XIED to see if that would help (it did not, and I have put the bike back to stock)

 

At this point there is something more nefarious going on. I was convinced it was somehow fuel related because I had to monkey around in the tank and the fuel pump sounding like it had air when sitting, but after doing everything I am beginning to think its not the case. I am concerned if it somehow voltage, because the battery was new in June and I am trying to maintain it properly (never off the tender for more than a day or two). Either way, like I said once I blip the throttle and bring the RPMs up above 1500-1700, it will idle on its own.

 

I am including 3 logs.

 

6-24: Started bike on warm day with idle lever at highest position. It started and stalled out. Second attempt it managed to stay running.

 

6-28: Started bike on hot day with idle lever at highest position. It started successfully.

 

10-22: Started bike on warmish day with idle lever just at the detent (roger told me that should be more than sufficient). Bike did its half start stumble twice, third time I managed to get it to catch and then blipped the throttle to settle it down.

 

If anyone can find a smoking gun here, or come up with something else to check, I will literally order you a pizza of your choice. I really appreciate any help you guys might be able to toss my way. I'm wondering if its something weird like resistance through the key or controls, or even the chain tensioners don't have sufficient pressure until oil circulates and the valves aren't opening right...

 

 

 

6-24 log.csv 8.88 kB · 3 downloads 6-28 log.csv 2.8 kB · 1 download 10-22 Log.csv 6.67 kB · 3 downloads

Afternoon  soco

 

I will look at your logged data later today, I won't open an unknown file on my work computer. 

 

What I will probably need from you shortly  is a data log  (starting with a cold start) with the AF-XIED disconnected, o2 sensor functioning, then about a 20 mile ride (to allow Motronic to rebuild the fueling adaptives) , then allow to cool overnight, THEN you can run the start-up  data log. (after cold start allow it to run long enough to get hot but don't overheat it) 

 

Can you do that for me? 

 

Logged data with a functioning  AF-XIED skews the base data I need to see for cold starting problem discovery. 

 

You also need to give me a data log long enough to get the engine hot enough to track fueling sensor input through a usable range temperature change.  (I took a quick peek at one of your posted data logs on my phone & I only saw about 30 seconds of run time ) 

[RogerC60]

Just for future reference, do the valve adjustment first, throttle body sync after that. In your list you have them in the opposite order.

[dirtrider]

Afternoon  soco

 

Caution: DO NOT do not do another  valve adjustment, & throttle body sync before logging the data that we need. I want to see what it shows as-is so we can identify & find your starting issue.  

[soco]

1 hour ago, dirtrider said:

Afternoon  soco

 

I will look at your logged data later today, I won't open an unknown file on my work computer. 

 

What I will probably need from you shortly  is a data log  (starting with a cold start) with the AF-XIED disconnected, o2 sensor functioning, then about a 20 mile ride (to allow Motronic to rebuild the fueling adaptives) , then allow to cool overnight, THEN you can run the start-up  data log. (after cold start allow it to run long enough to get hot but don't overheat it) 

 

Can you do that for me? 

 

Logged data with a functioning  AF-XIED skews the base data I need to see for cold starting problem discovery. 

 

You also need to give me a data log long enough to get the engine hot enough to track fueling sensor input through a usable range temperature change.  (I took a quick peek at one of your posted data logs on my phone & I only saw about 30 seconds of run time ) 

 

 

Dirtrider: I am happy to do so. The AF-XIED has been disconnected from the bike since before these logs began, I will leave it that way. I have been using the GS-911 via USB to my laptop, so I'll get it working via wireless and stick my laptop in a backpack. I will try to do a ride tomorrow early if I can, and then I will do a coldstart and run it for several minutes.

 

I'll try to get it all packaged and uploaded in a few days. Edit: I won't touch anything before logging.

 

 

Lowndes: Thanks for the welcome! I intend to do the brake lines and ideally a spline lube soon. Since it doesn't start well and is sort of embarrassing to try and leave say a nice restaurant on a bike that sputters, I probably don't put more than 400 miles a year on it. I do all my riding on my DRZ. Definitely need to correct this!

 

RogerC60: That is a good point. I did an initial valve adjustment when I got it, and then did the balancing more recently. I may have taken another look at the valves since then, its hard to remember because I have spent twice the time I ride it working on it. But I don't mind re-balancing it this winter, with the twinmax its a 30 minute affair.

[dirtrider]

5 minutes ago, soco said:

 

 

Dirtrider: I am happy to do so. The AF-XIED has been disconnected from the bike since before these logs began, I will leave it that way. I have been using the GS-911 via USB to my laptop, so I'll get it working via wireless and stick my laptop in a backpack. I will try to do a ride tomorrow early if I can, and then I will do a coldstart and run it for several minutes.

 

I'll try to get it all packaged and uploaded in a few days. Edit: I won't touch anything before logging.

 

 

Lowndes: Thanks for the welcome! I intend to do the brake lines and ideally a spline lube soon. Since it doesn't start well and is sort of embarrassing to try and leave say a nice restaurant on a bike that sputters, I probably don't put more than 400 miles a year on it. I do all my riding on my DRZ. Definitely need to correct this!

 

RogerC60: That is a good point. I did an initial valve adjustment when I got it, and then did the balancing more recently. I may have taken another look at the valves since then, its hard to remember because I have spent twice the time I ride it working on it. But I don't mind re-balancing it this winter, with the twinmax its a 30 minute affair.

 

 

I am happy to do so. The AF-XIED has been disconnected from the bike since before these logs began, I will leave it that way. I have been using the GS-911 via USB to my laptop, so I'll get it working via wireless and stick my laptop in a backpack. I will try to do a ride tomorrow early if I can, and then I will do a coldstart and run it for several minutes.-- At the moment we don't need a riding data log (that 20 miles was just to re-learn the adaptive. You can just grab the data I requested  in place.   

 

I'll try to get it all packaged and uploaded in a few days. Edit: I won't touch anything before logging. -- Yes, just trap the data as is. 

 

[dirtrider]

Afternoon  soco

 

MORE: 

 

OK, I just looked at your logs (quick look).

 

Before doing what I asked for above I need you to grab me a couple of quick data logs (in place).

 

I need you to hook up your GS-911 then set to trap data__

 

With CHOKE OFF,  turn key to ON then (& smoothly as you can)  slowly open then slowly  close the twist grip 2 or 3 times. (then turn the key off).

 

Next (can be on the same log or a new log)__

 

Turn the key back on then without touching the twist grip put the choke to on. Then count to 10 (I just need a pause in the data stream) , then after the 10 count move (& hold) the choke lever to full ON (it is spring loaded so you will have to hold it there). Hold for a 10 count. 

 

(that is all I need right now) 

 

I saw some odd TPS data in one of your logs & I don't know if it is a TPS thing or if you were moving the twist grip around while trapping the data. 

 

These logs  should clarify it. 

 

You can either post the logs here or directly message the logs to me through the message function on this site.  

 

 

 

 

[dirtrider]

Afternoon  soco

 

I looked at your log files in more detail this morning & one thing I am questioning is that your AIT (Air Intake Sensor) is working correctly, or that it is even plugged in.

 

On top of your air box cover is the intake temperature sensor,  (is the pigtail harness properly plugged into that?)

 

If it is, then unplug it, & verify that the pins in the connector are not bent, opened up, or pushed back into the connector) -- Basically try to ascertain that the terminals inside the connector are making solid contact with the pins in the sensor.

 

If you have an ohmmeter see if you can get a resistance reading on the sensor with the pigtail disconnected, I also need the ambient temperature at the time of measurement.  

 

I see your intake temperature tracking step in step with the engine temperature of some or your logs. While this COULD happen it seldom if ever does as the engine starts heating right after starting  (but you didn't have long enough  run times to verify this). Plus the 2 sensors seldom if ever read exactly the same right to the 1/10 of a degree. 

 

Might be something & might not but it is something we need to verify.  

 

 

[soco]

Dirtrider:  

 

I am fairly certain that is an artifact of the short sampling period. I warmed the bike up to approximately 180 engine temp today (and run in closed loop for a while) and the intake went up slightly due to ambient heating effects, but it tracked pretty well with outdoor air.

 

I am including a throttle roll on and roll off (twice), and a fast idle test. I also did a full warmup, not touching the throttle at any point beyond fast idle lever (so as to avoid those odd TPS readings I am sure I caused).

 

Two things stick out to me:

 

1) throttle didn't go beyond 80% open even when I was maxed out. That seems...odd. But I do the motronic throttle relearn every time I have to pull the battery so maybe the computer realizes that 80% is full open?

2) my resting battery voltage is 12.2 when the light is on. I do not know what a new healthy PC680 should output with a 55w headlight bulb but I still somehow fear my battery voltages are low, despite me doing everything to keep that from happening. I even checked voltage drop across the starter cables and the shown voltage is basically the same. I think when I ohm-ed it the wire was like .01 or something. I can retest that if needed, however.

10-23 full warmup.csv 10-23 fast idle.csv 10-23 throttle roll.csv

[dirtrider]

Afternoon  soco

 

Ok, that info might help__

 

 

On the full warm-up did you get a stall then a restart? Or did you do that 0 RPM period?

 

On your full warm-up log it shows that your o2 sensor is pretty sluggish & slow responding, that might indicate a failing o2 sensor or might just be due to running in place under no load. In any case it is something we probably should understand.

 

Have you been doing a new TPS re-learn procedure after any battery disconnect or fuse #5 removal? 

 

Something to try  (or at least understand) on your poor cold start is to ride the bike as normal, THEN after the ride remove fuse #5 & leave it out until just before next cold start.

 

The just before  cold starting the next time re-install fuse #5, then do a new TPS relearn (key on/ do not start/ fully open & close twist grip 3 times).

 

Then, see how it cold starts  & cold idles    (this will remove any odd stored adaptives that could have been created by a lazy o2 sensor)  

 

You have posted up 6 log files, can you remember what ones gave you the actual cold start & stall? I can tell the stall from your log data but  can't really tell if it was intentional or the system causing it. 

 

If I was working on that motorcycle something would probably try (as a test) is to unplug the o2 sensor then ride it a few days to see if that changes the cold start stalling.  

_______________________________________________________________

 

What GS-911 do you have (a new red one or older yellow one?) The new red WiFi GS-911 can hold the logged data inside the GS-911 unit & remain connected to motorcycle  (without a computer or phone link) then set to turn itself off at a lower battery voltage state (user settable) so it can  trap automatically at key-on then go into a low battery drain state by itself after shutdown when the battery voltage drops back to nominal. (I use this a lot in trying to catch difficult to catch problems as I can view & compare problem rides vs no problem rides then find  & compare the different logs. 

 

What I'm getting at is on a couple of your logs I see the battery voltage drop to well under  9 volts at very first crank-over then on other logs not that low.  I'm trying to define IF your cold start stall ONLY happens on the starts that the battery voltage drops into or under the 8 volt range (that low voltage can mess with the Motronic's  fueling control). Not an absolute but something we need to understand and/or eliminate.   

 

 

 

 

 

  

 

[soco]

please ignore this, I accidentally sent half a post.

[soco]

28 minutes ago, dirtrider said:

Afternoon  soco

 

Ok, that info might help__

 

 

On the full warm-up did you get a stall then a restart? Or did you do that 0 RPM period?

 

On your full warm-up log it shows that your o2 sensor is pretty sluggish & slow responding, that might indicate a failing o2 sensor or might just be due to running in place under no load. In any case it is something we probably should understand.

 

Have you been doing a new TPS re-learn procedure after any battery disconnect or fuse #5 removal? 

 

Something to try  (or at least understand) on your poor cold start is to ride the bike as normal, THEN after the ride remove fuse #5 & leave it out until just before next cold start.

 

The just before  cold starting the next time re-install fuse #5, then do a new TPS relearn (key on/ do not start/ fully open & close twist grip 3 times).

 

Then, see how it cold starts  & cold idles    (this will remove any odd stored adaptives that could have been created by a lazy o2 sensor)  

 

You have posted up 6 log files, can you remember what ones gave you the actual cold start & stall? I can tell the stall from your log data but  can't really tell if it was intentional or the system causing it. 

 

If I was working on that motorcycle something would probably try (as a test) is to unplug the o2 sensor then ride it a few days to see if that changes the cold start stalling.  

_______________________________________________________________

 

What GS-911 do you have (a new red one or older yellow one?) The new red WiFi GS-911 can hold the logged data inside the GS-911 unit & remain connected to motorcycle  (without a computer or phone link) then set to turn itself off at a lower battery voltage state (user settable) so it can  trap automatically at key-on then go into a low battery drain state by itself after shutdown when the battery voltage drops back to nominal. (I use this a lot in trying to catch difficult to catch problems as I can view & compare problem rides vs no problem rides then find  & compare the different logs. 

 

What I'm getting at is on a couple of your logs I see the battery voltage drop to well under  9 volts at very first crank-over then on other logs not that low.  I'm trying to define IF your cold start stall ONLY happens on the starts that the battery voltage drops into or under the 8 volt range (that low voltage can mess with the Motronic's  fueling control). Not an absolute but something we need to understand and/or eliminate.   

 

 

 

 

 

  

 

 

 

OK, i'll try to address these as methodically as I can:

 

1) full warmup log was the bikes natural tendency to stall. it starts and coughs and half catches enough you have to stop energizing the started because the bendix would skip, but it's not actually running strongly enough to fire and it ends up dying. This is the crux of the problem I am facing. (thus my suspicion its insufficient voltage to the ECU/ignition, or somehow air in the lines that needs to purge from the injectors. But due to how gravity works, I don't understand how the fuel could disappear from the lines without leaking into the crankcase or out of the injector connections, neither of which seem to be happening)

 

2) dying O2 sensor: I have no evidence the O2 sensor is healthy, and I would be happy to change it if you think you see sufficient evidence it is a culprit.  But I won't do anything yet.

 

3) full TPS relearn after battery or fuse 5 removal: yes, I have been doing this. I fully admit that the battery was out less than 50 miles ago (I changed to GS pegs because of my hips bothering me on 3+ hour rides, and to change the pegs I also changed out the master cylinder and brake line to the ABS due to the different geometry.) I would be happy to ride it for 50-100 miles back and forth to the office if there is a particular "learning curve duration". I would also be happy to manually pull fuse 5 and relearn it, and send another warmup log if it would help.

 

4) pulling fuse 5, and doing a relearn right before another cold start: I will happily try this. I can ride the bike around tomorrow to get groceries (man, I love saddle bags!) and then pull the fuse, and do a relearn/cold start the following morning.

 

5) logs indicating stall: every stall is due to the bike not fully starting. I tend to let up after several firing events (in my uneducated terms, a few "chuggas") but the bike doesn't seem to have the wherewithal to actually stay running unless everything is hot. I will say, once the bike has actually started running (think a failed start, and then it finally runs for 20 seconds), when I shut it off it will usually start back up. That wouldn't build up sufficient heat, but it does seem to do three things:

A) if there WERE air in the fuel lines, purge it

B) build oil pressure

C) begin alternator field exicitation

6) my GS-911 is a red one. So I can figure out how to leave it in there and just try to ride the bike for a while and capture it all. I would be curious what its storage capacity is, should I dump it daily or just weekly?
 

I think at the end, I have a few things to try.

 

 

 

 

FIRST: tomorrow, do a cold start (logged) and see if it fails like usual. let it run for 20 seconds (not enough to build heat), and kill it, then log another subsequent start. If it works properly, that might indicate something.

 

SECOND: Ride the bike around after that test for some around town errands, then pull fuse five after. next time I start it at least a day after, reinsert it and relearn, then log another cold start but on a fresh motronic learn cycle.

 

THIRD: after all that, unplug the O2 sensor and put some trips and days on it, and then log yet another cold start and see what happens.

 

FOURTH: figure out how to install the GS-911 to log automatically when the bike is running. I would have to make sure it can sense a Key On Engine Off situation, so it would start logging before the alternator builds voltage but for $400 I would assume those south africans took care of that haha.

 

I do agree with you on the voltage drops, it concerns me as this battery is as new as I can get. I'd put a bigger capacity one in there if I could, or a lead acid if I thought it would resist amperage draw better. But I do not think thats the case. It bothers me that a new $150 battery seems unable to keep up, especially with a new starter from beemer boneyard (and not the cheap one either). 

 

Feel free to correct me in any of this, I cannot tell you how much I appreciate the help.

[dirtrider]

Afternoon

 

  

OK, i'll try to address these as methodically as I can:

 

1) full warmup log was the bikes natural tendency to stall. it starts and coughs and half catches enough you have to stop energizing the started because the bendix would skip, but it's not actually running strongly enough to fire and it ends up dying. This is the crux of the problem I am facing. (thus my suspicion its insufficient voltage to the ECU/ignition, or somehow air in the lines that needs to purge from the injectors. But due to how gravity works, I don't understand how the fuel could disappear from the lines without leaking into the crankcase or out of the injector connections, neither of which seem to be happening)-- It could have a dripping injector but that is the reason for the 2 second pump run at key-on, that usually purges any air out of the fuel manifold. 

 

2) dying O2 sensor: I have no evidence the O2 sensor is healthy, and I would be happy to change it if you think you see sufficient evidence it is a culprit.  But I won't do anything yet.--- Not yet as we need to find your smoking gun if possible, but riding (then cold starting) a few times with it disconnected might shine a light on that as the possible cause.  (eventually I would probably replace it.    

 

3) full TPS relearn after battery or fuse 5 removal: yes, I have been doing this. I fully admit that the battery was out less than 50 miles ago (I changed to GS pegs because of my hips bothering me on 3+ hour rides, and to change the pegs I also changed out the master cylinder and brake line to the ABS due to the different geometry.) I would be happy to ride it for 50-100 miles back and forth to the office if there is a particular "learning curve duration". I would also be happy to manually pull fuse 5 and relearn it, and send another warmup log if it would help.-- Try the fuse removal first. I use  couple of wire leads out from under the seat  with a fuse holder across the outer end with little spade terminals on the inner end  that will fit into the undereat fuse cavity terminals. That way I can remove/replace  the fuse without removing the seat. 

 

4) pulling fuse 5, and doing a relearn right before another cold start: I will happily try this. I can ride the bike around tomorrow to get groceries (man, I love saddle bags!) and then pull the fuse, and do a relearn/cold start the following morning.--- Yes, see where this takes you. 

 

5) logs indicating stall: every stall is due to the bike not fully starting. I tend to let up after several firing events (in my uneducated terms, a few "chuggas") but the bike doesn't seem to have the wherewithal to actually stay running unless everything is hot. I will say, once the bike has actually started running (think a failed start, and then it finally runs for 20 seconds), when I shut it off it will usually start back up. That wouldn't build up sufficient heat, but it does seem to do three things:-- Try starting it this way-- Put choke on, then hold twist grip about 1/8 open during cranking (try different openings).  

 

[dirtrider]

6) my GS-911 is a red one. So I can figure out how to leave it in there and just try to ride the bike for a while and capture it all. I would be curious what its storage capacity is, should I dump it daily or just weekly?-- Good question, it will hold quite a bit but also depends on how much data you set it to trap, I have that info someplace as I researched it when I bought & updated my latest RED GS-911 WiFi.  With a WiFi connection to your phone or computer you can leave the GS-911 in place then dump the logged data each day if you need to (this is what I usually do as I HATE large logs as they are a pain to look at & use  .  One problem I have had is it won't re-start logging data on it's own so I use the  key-on to wake it then use my cell phone  set a new log trap before engine starting (you might have to do this) .  If I'm making a lot of stops I just keep the GS-911 app open on my phone so a couple of clicks before starting & it is back trapping data again. If you set it to turn off  .1v above rested battery voltage it will stay on for quite a while after key-off so for short stops it should stay active (but logs get pretty darn long) . 
 

 

 

 

FIRST: tomorrow, do a cold start (logged) and see if it fails like usual. let it run for 20 seconds (not enough to build heat), and kill it, then log another subsequent start. If it works properly, that might indicate something.-- Yes the more data the better just label your data with what you did, or trapped, so we can understand what we are looking at. I usually name my data by date/time then give it an ID number. Then I make full notes with lots of info on  paper about that date/time/ID number (so I know as much as possible about what I trapped) 

 

SECOND: Ride the bike around after that test for some around town errands, then pull fuse five after. next time I start it at least a day after, reinsert it and relearn, then log another cold start but on a fresh motronic learn cycle.-- You can log this or not, I'm more interested in IF it helps then seeing the data.

 

THIRD: after all that, unplug the O2 sensor and put some trips and days on it, and then log yet another cold start and see what happens.--Same as above, more interested in IF it helps than seeing the data.

 

FOURTH: figure out how to install the GS-911 to log automatically when the bike is running. I would have to make sure it can sense a Key On Engine Off situation, so it would start logging before the alternator builds voltage but for $400 I would assume those south africans took care of that haha.-- The LED on the GS-911 will tell you if it sleeps then wakes up at key-on (mine does).  Just set the turn off voltage to be .1 volts over resting battery voltage (probably 12.7v or 12.8V on an Odyssey)

 

I do agree with you on the voltage drops, it concerns me as this battery is as new as I can get. I'd put a bigger capacity one in there if I could, or a lead acid if I thought it would resist amperage draw better. But I do not think thats the case. It bothers me that a new $150 battery seems unable to keep up, especially with a new starter from beemer boneyard (and not the cheap one either). -- Do you know how to do a (cranking) voltage drop test? If so do one on the + battery cable between battery positive post & starter positive cable  lug. Then do another between battery negative post & engine case. 

You can also try using jumper cables from another vehicle for cold starting as a test, if your start/stall goes away then we have a direction to look in. 

 

Feel free to correct me in any of this, I cannot tell you how much I appreciate the help. -- I will try to stay with you & work through this, sometimes it isn't easy to find these things especially when the obvious has already been done. It is MUCH easier when working on the motorcycle as the data stream is pretty limited on the BMW 1150 Ma 2.4 so doing it with just a data log is not easy unless there is a big smoking gun (most times there isn't an arrow saying look-here "D.A." in the data).  

[soco]

All:

 

Unfortunately, my father in law suddenly passed just after I last replied to this thread. He was working on his e36 in his garage and his heart gave out (at least he was doing something he liked). His wife found him an hour later, and my wife was 2 hours away at a wedding and her 2008 5 series wagon blew the cooling system 5 miles from the venue as she was coming home. It was...not a great situation. After dealing with supporting the two women, we had a brutally cold winter and to be honest I didn't feel like checking the bike in 20 degree weather so the project sat idle. However, they are both doing really well now, and since it finally warmed up here I went back out to check on it.

 

I found that, when pulling the motronic #5 fuse for five minutes and relearning the throttle, it seems to start perfectly. So, somehow that is the culprit. I recall dirtrider telling me that my O2 sensor looked lazy, is there a possibility that it is fouled and causing the long term fueling to be too lean as it learns? If anyone has any bright ideas on what to test next, or if I should just replace the sensor, i am open to the collective wisdom of the board.

 

Thanks everyone!

[dirtrider]

31 minutes ago, soco said:

All:

 

Unfortunately, my father in law suddenly passed just after I last replied to this thread. He was working on his e36 in his garage and his heart gave out (at least he was doing something he liked). His wife found him an hour later, and my wife was 2 hours away at a wedding and her 2008 5 series wagon blew the cooling system 5 miles from the venue as she was coming home. It was...not a great situation. After dealing with supporting the two women, we had a brutally cold winter and to be honest I didn't feel like checking the bike in 20 degree weather so the project sat idle. However, they are both doing really well now, and since it finally warmed up here I went back out to check on it.

 

I found that, when pulling the motronic #5 fuse for five minutes and relearning the throttle, it seems to start perfectly. So, somehow that is the culprit. I recall dirtrider telling me that my O2 sensor looked lazy, is there a possibility that it is fouled and causing the long term fueling to be too lean as it learns? If anyone has any bright ideas on what to test next, or if I should just replace the sensor, i am open to the collective wisdom of the board.

 

Thanks everyone!

Afternoon  soco

 

Yes, a lazy o2 could be causing skewed fueling adaptives therefore possibly causing a fueling offset at cold starting. It's probably not the TPS re-learn that helps with a removed #5 fuse but more likely a fueling adaptive issue. 

 

So next, try unplugging the o2 sensor then, remove #5 fuse for a while to re-set Motronic, then ride the bike as normal for a couple of days to see how it cold starts.

 

Does holding the throttle about 1/8 open for cold starting help the cold start? 

 

How about holding the choke up one notch past full choke  (the spring loaded position), does that help starting? 

 

You might also do a cold engine fuel flow test from the rear fuel  return line (if no flow then you might be fighting with low cold fuel pressure).

 

If nothing helps in the above then ride the bike as normal for week or so (with o2 sensor connected) -- This will give us a baseline. Then grab about 2 minutes of cold start GS-911 trapped data.

 

Then do the very same as above but this time do a 5 minute #5 fuse removal first then a TPS re-set. 

 

Then PM me the GS-911 data to  look at__

 

What is your COLD cranking voltage (measured at the battery posts) then measured again right at a fuel injector green wire? If the injector measurement is significantly lower than the battery post voltage then you might need to add a power relay for the injector power circuit  (2nd load relay relay). If both battery & injector voltages are low then possibly a battery or starter issue.  (low injector voltage lowers the cold start injector spray) 

 

 

 

 

 

 

 

 

[soco]

4 hours ago, dirtrider said:

Afternoon  soco

 

Yes, a lazy o2 could be causing skewed fueling adaptives therefore possibly causing a fueling offset at cold starting. It's probably not the TPS re-learn that helps with a removed #5 fuse but more likely a fueling adaptive issue. 

 

So next, try unplugging the o2 sensor then, remove #5 fuse for a while to re-set Motronic, then ride the bike as normal for a couple of days to see how it cold starts.

 

Does holding the throttle about 1/8 open for cold starting help the cold start? 

 

How about holding the choke up one notch past full choke  (the spring loaded position), does that help starting? 

 

You might also do a cold engine fuel flow test from the rear fuel  return line (if no flow then you might be fighting with low cold fuel pressure).

 

If nothing helps in the above then ride the bike as normal for week or so (with o2 sensor connected) -- This will give us a baseline. Then grab about 2 minutes of cold start GS-911 trapped data.

 

Then do the very same as above but this time do a 5 minute #5 fuse removal first then a TPS re-set. 

 

Then PM me the GS-911 data to  look at__

 

What is your COLD cranking voltage (measured at the battery posts) then measured again right at a fuel injector green wire? If the injector measurement is significantly lower than the battery post voltage then you might need to add a power relay for the injector power circuit  (2nd load relay relay). If both battery & injector voltages are low then possibly a battery or starter issue.  (low injector voltage lowers the cold start injector spray) 

 

Dirtrider:

 

Thanks for sticking with me through this. I generally always start the bike with the cold start held one click above the detent (i suppose one might call that second position) simply because the bike has always been hard to start, so it's become force of habit. That usually equates to 4-5° on the TPS per the GS-911. I haven't tried it with the throttle cracked in a systemic way to gauge its efficacy, since everyone online told me the cold start should be sufficient for any temperature I would ride in and I would be more likely to screw things up by using my wrist over the lever. (Plus the few times I tried didn't show that cracking it beyond the cold start lever second detent helped).

 

I think the best bet is to disconnect the O2 sensor, then ride it a few days and see what happens. Unfortunately even though it was almost 60°F today, we are supposed to have about 3" of snow on the ground tomorrow by noon. So riding logs will probably be delayed for a bit. Once the weather permits regular rides (like daily, and over 5 miles) I will start using it for regular transport and see if it is normal. After that, I can report the running conditions and grab some log data.

 

As to voltage, I will grab the data but as an unofficial test I used my NOCO 10 amp battery charger on the pigtail I installed and put it on 12v supply mode (constant 12v regardless of battery condition) and it did it's usually crappy start. Then later, I started it after pulling fuse 5 with the batter charger completely disconnected and it started marvelously. So I think the problem is more likely fueling trim being lean vs. low battery voltage. But I will definitely log some starts once I can ride it again and get some hard data.

 

By the end of this I feel like I will be tooled up and experienced enough that I am obligated to help other oilhead owners! Maybe I should start a local group or something and pay it forward...