• Welcome, Guest! We hope you enjoy the excellent technical knowledge, event information and discussions that the BMW MOA forum provides. Some forum content will be hidden from you if you remain logged out. If you want to view all content, please click the 'Log in' button above and enter your BMW MOA username and password.

    If you are not an MOA member, why not take the time to join the club, so you can enjoy posting on the forum, the BMW Owners News magazine, and all of the discounts and benefits the BMW MOA offers?

Starter or Battery 2004 R1150RT

Test #2 eliminates voltage drop in the connections and cable from bike Battery(+) to the starter.
As well, it adds a huge secondary current source, yet we still see a 3.99V drop.

For that reason, I say the starter is drawing too much current.

PS: I love the way you go about collecting data.

I'm on track with you except that I know I've been undercharging the battery for a year and a half according to the info from GSAddict. So I'm going to get a proper charger later this week, cycle the battery according to the Odyssey specs and see how it does. Since I've already serviced the starter twice, once with the planetary gear cover arcing on the armature, I will likely get a new starter anyway. But I've remained curious about why the voltage dip leads to the slow, lean start.

I ran another test this morning to see if I could out why the dip causes the problem: lights bright, starter turning over quickly but slow-start and lean afterstart condition. Yesterday I separately powered the the Motronic with no improvement.

This morning I separately powered the injectors and fuel pump from a second source (not hard, just pull the relay and plug in a jumper which I recommend carrying in case you lose a relay). BINGO! With only the fuel pump and injectors powered from a second source, the bike started right up. So low injector voltage is the issue with low-battery, hard-starting.

I should have known this because last year when I noticed my alternator taking time to come on-line after starting (about a minute) I also noticed that the mixture got much richer when the battery voltage stepped up due to the alternator starting to charge. This is easy to see with an LC-1 because it's always measuring the exhaust, and the Motronic is always in Open Loop until the engine is warm so the O2 sensor doesn't adapt to the low voltage.

So my theory is this: the voltage dip is coming at a time, on first cold start, that the Motronic usually adds a lot of fuel, perhaps for as little as the first few revolutions. At that time, the voltage dip keeps the injectors from squirting enough fuel to adequately wet the TP and injection track. From then on, it is behind the eight-ball and playing catch-up with the fueling.

The Motronic does compensate for low battery voltage but this is now the second time that I've seen its compensation to be inadequate.

Still TBD, battery or starter ...


Added:
Here's how the mixture varied when the alternator came on-line slowly. This isn't real bad but it is measureable. The start plot is much worse. The step down in AFR is the alternator coming on-line at 3 minutes 20 seconds.

plug3087.jpg
 
You've nailed it again

Sorry Roger, I'm noting you have diligently tracked down the actual "cause - effect" again and the culprit is still sitting there sort of doing its thing.
The obvious next step (apart from normal and accepted maintenance routines) would be to adjust that voltage correction table.
M.S. beckons.
 
Clarification

My post to Roger is totally an inside thing - not entirely clear to others most likely. M.S. is a tunable controller product.
I concur with Rogers findings on the reason for the "lazy" first of the morning start routine on his bike. There are numerous fueling strategies common to fuel injection (prime pulse/ cranking pulse width/ enrichment versus time factor/ and warm up versus temp tables/ perhaps etc.) and I have noted the exact same symptom on my project - early on.
The thing of it is, the Motronic controller does not avail itself to change low voltage fuel PW correction which would address the issue (although a new starter or more fully charged battery is also reasonable to consider doing). I'm not questioning or condemning it, she just can't be programmed.

I was compelled to respond to your excellent post Roger, and findings as you systematically narrowed things down till the culprit reared its head.
Being on the other side of the fence so to speak, where the banquet is in full progress (M.S. is rather broad like a buffet of all kinds of tuning treats) I'm offering bread crumbs I guess.
 
So my theory is this: the voltage dip is coming at a time, on first cold start, that the Motronic usually adds a lot of fuel, perhaps for as little as the first few revolutions.

Interesting, that makes sense.
I remember not lifting the fast idle lever a while ago and tried to start. The bike sputtered and I tried to catch it with the throttle but too late.
When I cranked again the bike did not catch right away and when it did it acted just like it was overchoked. I could smell the fuel.
 
Grounds, grounds, grounds and then more grounds.

Every time I see something like this and getting more and more complicated bringing more questions than answers, I always go back to the very beginning and ask, "did you take the ground straps off and clean them?"

I have had experiences with crazy problems that got even crazier with time. Right from the introduction of electronic vehicle controls, every seminar was the same, go after the grounds first, even when they look good.

Once upon a time, I had a Detroit that the operator complained would power out. Check the engine for codes, tons of codes and none made sense. At the end of the day, one ground strap to the ECM had 2/10ths of voltage drop. I never would have imagined it could throw the ECM into fits, but it did.
 
Sorry Roger, I'm noting you have diligently tracked down the actual "cause - effect" again and the culprit is still sitting there sort of doing its thing.
The obvious next step (apart from normal and accepted maintenance routines) would be to adjust that voltage correction table.
M.S. beckons.

Steve Mullen at Nightrider.com the designer and manufacturer of the AF-XIED has spent many years improving the fuel injection of HDs. He implement a Megasquirt, here are some of the details: http://www.nightrider.com/biketech/efi_project_00.htm.

I asked him how long he thought it might take on an R1150, he thought a few to several months, faster if you could use the dyno for timing developement.

Following this thread makes my head spin. And unlike Roger's starter, it ain't slow!!:D

This was a thorny problem. I could have started throwing parts at it but that's not my M.O. It seems like there might be some mods that would make the R1150/1100 more reliable at starting. I could have lived with mine forever, I was just bothered about the, why? of it.

My post to Roger is totally an inside thing - not entirely clear to others most likely. M.S. is a tunable controller product.
I concur with Rogers findings on the reason for the "lazy" first of the morning start routine on his bike. There are numerous fueling strategies common to fuel injection (prime pulse/ cranking pulse width/ enrichment versus time factor/ and warm up versus temp tables/ perhaps etc.) and I have noted the exact same symptom on my project - early on.
The thing of it is, the Motronic controller does not avail itself to change low voltage fuel PW correction which would address the issue (although a new starter or more fully charged battery is also reasonable to consider doing). I'm not questioning or condemning it, she just can't be programmed.

I was compelled to respond to your excellent post Roger, and findings as you systematically narrowed things down till the culprit reared its head.
Being on the other side of the fence so to speak, where the banquet is in full progress (M.S. is rather broad like a buffet of all kinds of tuning treats) I'm offering bread crumbs I guess.

I think we will all be interested to hear how your Megasquirt (or Microsquirt) project progresses. It sounds like a lot of fun.

Interesting, that makes sense.
I remember not lifting the fast idle lever a while ago and tried to start. The bike sputtered and I tried to catch it with the throttle but too late.
When I cranked again the bike did not catch right away and when it did it acted just like it was overchoked. I could smell the fuel.

I left out some of the detail from this morning's test. My car battery voltage sits around 12.4V with the engine off. That is the voltage I used to power the fuel pump and injectors. The Motronic, as I started cranking was about 12.2V. The Motronic determines its fuel injector pulse width based on the voltage it sees (12.2V before start and about 13.3V after it started running.

So to recap, the Motronic had a good estimate of the fuel injector voltage when the bike started by a couple seconds later the FI voltage was about 1 Volt lower than the voltage powering the Motronic. The result, the bike started quickly, and then ran lean. This is an excellent indication of just how voltage sensitive the injectors are, and how inadequate the Motronic Dead-Time vs Battery Voltage is.

Tomorrow I will power the FI and Motronic from my car battery, with the engine running. The voltage will then be about 13.3V on both systems.

I'm starting to think about long-term solutions to the FI voltage problem. (Including getting my batter charged properly.)

RB
 
conventional lead acid battery

Interesting thread. I prefer conventional lead acid liquid cell batteries to the sealed glass mat maintenance free units even though there is more chance for leakage especially in a tip-over event. Any reason why a Yuasa 51913 battery won't work/fit a '04 R1150 RT? The 51913 dimensions are: 8.5"X7.2"X3.8". I know this battery will fit my '94 R1100RSL.

Also, there would be no charger issues associated with a conventional lead acid liquid cell battery.
 
Steve Mullen at Nightrider.com the designer and manufacturer of the AF-XIED has spent many years improving the fuel injection of HDs. He implement a Megasquirt, here are some of the details: http://www.nightrider.com/biketech/efi_project_00.htm.

I asked him how long he thought it might take on an R1150, he thought a few to several months, faster if you could use the dyno for timing developement.



This was a thorny problem. I could have started throwing parts at it but that's not my M.O. It seems like there might be some mods that would make the R1150/1100 more reliable at starting. I could have lived with mine forever, I was just bothered about the, why? of it.



I think we will all be interested to hear how your Megasquirt (or Microsquirt) project progresses. It sounds like a lot of fun.



Yes, thanks on that Roger. For those not on the same page, not to hijack thread, and briefly - I have an 83 airhead that is fuel injected with a MicroSquirt ECU for fuel and ignition control (as well a turbocharger that brought on some pretty "polarized" interest when last mentioned on the MOA forums). Daily driver, been to many rallies including this year in Salem, and reliable to the point I pull it down to change things frequently just for the fun of it.
Relative to this phenomenon you have identified, I would offer that it is quite reliably an actual short coming and does not relate to bad grounds (I do however endorse bad grounds as being a leading simple issue oft times ignored in favour of more complex and expensive repairs).
To respond to your observation of the challenge in programming an alternate ecu to an oilheads requirements, particularly the ignition mapping, I'd like to offer the following road map I'd be inclined to follow:
a) Piggy back a gen 2 or 3 MicroSquirt with the stock oilhead system and components in place.
b) Add certain sensors to allow a seamless co-existence of the second controller, such as MAP/ CLT/ O2/ TPS. The 2nd TPS might be plumbed via mechanical similar to the split throttle cable units well used and proven on BMW's. Key would be picking up the stock ecu ignition trigger signal(s), and on this chapter I can attest to many numerous ways being available to "skin the cat" as they say.
c) Tuner Studio logging via a lap top on bike. This is where it gets controversial, I've utilized one on my tuning efforts mounted to a light aluminum bracket system atop my fuel tank - one needs to avoid distraction while in motion.
d) The MS controller would be going along for the ride and logging would allow capture of the stock Motronic treatment of ignition timing versus rpm's, load and throttle states to cheat that portion of the MS tuning required.
e) An ignition map is agreeably one of the harder files to gain on your own and without some proper dyno time. All the other numerous files (VE & AFR tables/ enrichments/ corrections/ etc. etc.) are much easier to tip toe through and gain results. The Motronic Ignition Table would arguably be a "good starting point" to resume bike operation and improve from there. There is always room for improvement.
f) Logs taken under all conditions, and including cold starts, warm ups, modest cruise, transitions, heavy acel, you get the idea would provide the back drop to cheat a really good starting "MSQ" for the MicroSquirt controller to be commissioned and take over as main controller.
There's lots of hype and mystique/ concern/ fear about MegaSquirt Tuning and challenge, but in the end it is all about diligence, scientific approach, and accuracy. Lots of people out there to substantiate results, including drag strips and the Salt Flats for serious power and efficiency gains.
Cheers, Lorne.
 
Lorne, Thanks for posting this detailed info. You've got a lot of good info on Megasquirt. So that it doesn't get lost, I think it makes a lot of sense to give it its own thread. Even if it is just to describe the process of implementation and observations. I would hate to see this good stuff get lost amidst this battery and starting stuff. RB
 
Quick updates to this ongoing saga:

DY suggested looking at the grounds. Good idea, checked them by measuring high current voltage drop, they are fine. I also took out the BMW wiring CD. The grounding is certainly not obvious. In stock condition, there are four ground wires that connect to the battery and only one of them is a direct connection to the transmission. At some point I will do a thread on this. However, the Motronic, Alternator and Generator all get their grounds from the transmission/engine connection and as a result, any drop in the large cable from battery to transmission would affect measurements and operation.

Today I connected the fuel injectors to a car battery with the car running, adding a volt to the injectors during my cold start test (12.4V to 13.4V or thereabouts). With everything else connected normally, the battery voltage low as it has been. The motorcycle started quickly and ran smoothly. So the cause of my problems is low battery voltage during starting and the effect is lower voltage at the injectors that the Motronic can't/doesn't compensate enough for during starting, which is Open Loop operation.

Also, I discharged the battery for 20 minutes with the headlight, rode the motorcycle for an hour. Came back, turned on the headlights, lo and behold, the battery voltage with the headlights on but the bike not running was 12.5V for 5 minutes or so. This is 0.4V more than I have recorded in any log that I have in the past year for the same condition.

So it looks like GSAddict nailed it, the problem is that I've undercharged the battery for the past year and a half. I will discharge/charge it per detailed recommendations (that I'll post later) from Odyssey.
RB
 
Last edited:
What I know about electricity could be written on the head of a pin - but I'm impressed as all get out with the knowledge available from the people on this Forum.

:bow
 
Nice going Roger. This problem probably goes unnoticed most of the time. Your persistent troubleshooting is exemplary! I think I've managed to avoid low voltage issues by using an Odyssey approved charger on a regular basis on two of my bikes that are equipped with that battery.

They don't seem to be available in N.A. any longer but I found some in Europe and Zambia for example. If you can get your hands on one at a reasonable price the OptiMate AccuMate Pro 12V-7A-S Battery Charger is what GSAddict and I both use on our Odyssey batteries. A tad expensive but worth every penny I think.

This guy here: http://www.tecmate-int.com/accumatepro12v7a_detail.php
 
Invitation to post on complex content

Lorne, Thanks for posting this detailed info. You've got a lot of good info on Megasquirt. So that it doesn't get lost, I think it makes a lot of sense to give it its own thread. Even if it is just to describe the process of implementation and observations. I would hate to see this good stuff get lost amidst this battery and starting stuff. RB

Thanks for the invite Roger.
I enjoy reading your posts, & obviously there is a lot of really good material/ people on these forums to assist anyone looking to know more about their ride. I actually have a few topics I had previously thought of opening for general consideration or comment (as a newby to the club but well seasoned tech), but then pulled back as controversey/ angst seemed to bloom.
Kurt (Admin) on the airhead forum assured me all was well and normal - so barring an all out riot I suspect a couple of interesting sectors can be given a go.
Regards, Lorne.
 
Odyssey Charge Rates

Do an equalize charge on the Odyssey of 14.9v - 15v for at least 8-10 hours.
Then test again. You will be surprised.
I am betting that your "new" battery is not fully charged.
The Motronic does not operate well below 9.5v
The problem with AGM's is they want (and need) a higher charge voltage than 14.0v which is what Oilhead alternators typically put out.
I have seen this problem on a few bikes now.
My solution on my own machine is I modified the voltage regulator to charge @ 14.7v years ago and never had any further issues.
I love Odyssey batteries and recommend them. The catch is they must be periodically equalized (with a 5-10A charger) to maintain peak capacity in our application.
This was all confirmed by one of the engineers at Odyssey when I had this problem myself. Their site has a lot of engineering data confirming the requirements.
My machine (1150GS) always starts within 2 seconds.

I have to admit not researching the AGM charge behaviour very diligently, and find the above numbers quite surprising. I'll further reveal ignorance on the subject with a question as to the tendancy of the closed case to vent or loose medium when charge rates go over a threshold? (as flooded cells are so vulnerable to).
I had added a PC925 to my airhead a few years ago, and left the system charging it at the traditional 13.8v (albeit that system a refit that provides 900w and is flat line 13.8 throughout all rpm's). I did a cursary read on the Odyssey paperwork back then but either missed this or it didn't provide such detail.
I'm interested to follow your final accessment Roger, on your battery, and course of action. Not having problems but wish to treat the Odyssey as it should be.
I have a link on file for an external voltage regulator that can be "dialed" up/down to customs output levels, others may have seen that.
 
I have to admit not researching the AGM charge behaviour very diligently, and find the above numbers quite surprising. I'll further reveal ignorance on the subject with a question as to the tendancy of the closed case to vent or loose medium when charge rates go over a threshold? (as flooded cells are so vulnerable to).
I had added a PC925 to my airhead a few years ago, and left the system charging it at the traditional 13.8v (albeit that system a refit that provides 900w and is flat line 13.8 throughout all rpm's). I did a cursary read on the Odyssey paperwork back then but either missed this or it didn't provide such detail.
I'm interested to follow your final accessment Roger, on your battery, and course of action. Not having problems but wish to treat the Odyssey as it should be.
I have a link on file for an external voltage regulator that can be "dialed" up/down to customs output levels, others may have seen that.


I agree, the more I read about the subject of the Odyssey Battery, the more I realize that I may be short changing the life of the battery with the Jr charger I am using. I asked the person at Odyssey when I bought the battery if the float charger I had would be sufficient and he at the time of purchase, assured me it would. Sufficient vs Optimal are 2 different things.

What I do have are a couple of Jr tenders that I use on my lead acid battery(s) and recently a DieHard Gold Charger / Starter that I use for the cars. My previous Sears charger lasted about 32 yr, before I put it out to pasture. My new unit has a AGM setting and will 2 amp / 10 amp / 30amp charge or 3 sec start at a 80amp discharge. I would not use this to 30amp charge or to start a motorcycle battery with. Car or similar yes. I have put this charger on the PC680 on the 10amp and now on the 2amp setting. It will get bring the battery up to 14.68v or so. When it initial charges it will go to 15.2v but for only about 10 secs then it floats down to the 14.6v range.. But after several hours it shuts it self off and the battery settles down to 13.1v or so. Then I don't consider this a true float charger and don't think I will use this to maintain the PC680 long term. Course at the time I also did not do the needed indepth research and believed this charger would charge and a Jr would float the PC680 to a best level. It looks like I was wrong to believe that....

So now I may have (or is that need, do I really have to have?) to purchase a charger that cost's near the price of the battery to support it... :blush

This is an interesting subject, My thanks to you gentlemen for the research and sharing it with us..
 
Last edited:
This Diehard charger has medium size clamps that isn't safe to leave unattended for days on end to keep the battery floated.

With my work schedule, the bike sits for weeks if not a month at a time between rides right now. Should I float as part of a maintenance schedule and disconnect the battery between rides? That neg cable isn't the easiest thing to get to.

I want to keep the battery near or at the top of it's game, but rather would not have to partially disassemble to hook a cable up, go on a day ride and then repeat to set it back up for 3 or 4 weeks.

That is my small concern.
 
Here is what I heard from odyssey on the subject of pc680 maintenance.

Hello Roger,

Thank you for contacting EnerSys/ODYSSEY.

See attached manual reconditioning cycle for your reference. If your headlight will stay on, this would work fairly well for a way of discharging the battery.

Proper charger maintenance is critical to maximize the life and performance of the ODYSSEY battery. ODYSSEY Battery 101?

The usable energy of the ODYSSEY battery is from 11.2V (0% state of charge) OCV to 12.84V OCV (Open Circuit Voltage should be checked after a minimum of 6-8 hours rest period with no loads) or higher. When the battery goes below 10.0V OCV, you are getting into the chemical part of the battery and can cause permanent damage. At less than 8.0V the ODYSSEY limited warranty deems the battery over-discharged due to abuse or neglect. At less than 6V the battery can develop reversed cells depending on the condition of the battery. The ODYSSEY charger/maintainer can maintain the battery at a full state of charge indefinitely in or out of the application without harming the battery. If the battery becomes over-discharged (below 11.0V) then the sooner the battery can be fully charged the better and if the battery remains connected to the application during storage or extended periods of non-use. There is a link to a list of ODYSSEY approved 12V chargers located on the ODYSSEY website Product Support page for your reference. Any of the approved chargers can be used to maintain the battery indefinitely without harming the battery.

The list of approved 12V chargers is provided due to the many chargers that are programmed for the vast majority of batteries on the market that prefer low amp charging and lower float voltages that do not fully charge or maintain the ODYSSEY battery properly. The recommended charging current for an ODYSSEY battery is 40% of the 10 hour amp hour rating of the battery for cyclic applications (about 6A for a single 16Ah rated PC680), a constant float voltage of 13.5-13.8V and no constant voltages exceeding 15.0V in any kind of automatic de-sulfation/reconditioning/equalize mode. At greater than 15.0V the battery can overcharge, overheat, and/or go into thermal runaway. Maintaining the battery at less than 13.5V will bring the battery down and maintain it in an undercharged condition causing sulfation and premature failure.

For seasonal applications (non-daily use applications that set for more than 3 days in a row consistently) frequent use of an approved maintainer that meets the charge voltage requirements noted in the previous paragraph is highly recommended during the season. The preferred storage method is to fully charge the battery before storing and disconnect the battery from the application (shelf storage mode) Stored in or out of the application with no loads, the battery would not require charging for up to 2 years at 77?F or until it reaches 12.0V, whichever comes first. The self discharge rate increases significantly for temperatures above 25?C (77?F) and for every 10?C (18?F) temperature increase the storage time to recharge is decreased by half. Charge maintenance is critical to maximizing the life and performance of the battery. Freezing will not harm the battery and self discharge rates reduce significantly at colder temperatures.

It is recommended that the ODYSSEY battery be charged if it is less than 12.65V when put into use per the ODYSSEY Owner?s Manual (link provided on website Literature page for your reference). Most standard alternators/stators are not meant to be deep discharge recovery chargers and can damage the alternator/stator as well as not fully charging the battery with limited use. The Cyclic Charge Voltage range printed on the top label of the battery is the recommended voltage at the battery from the applications charging system (alternator or stator). At less than 14.1V the battery may not be getting fully charged for infrequently used applications. You can verify the battery voltage by checking the voltage at the battery at least 8 hours after application use (or off charge) and if the battery voltage is not at least 12.84V then the battery is not considered fully charged. Voltage readings taken right off charge or after use (alternator/stator charging) will be inflated and inaccurate so for a true OCV reading, you should wait at least 8 hours before checking the voltage (OCV) with 24 hours being preferred.

Any feedback is always welcome and appreciated. I hope this helps. Please contact me if you have any questions, concerns, or need further assistance.

Sincerely,
ODYSSEY Sales/Support Representative
 
Odyssey Charge Requirements

Excellent, guess I'll be checking surface voltage when I get home tonight - bikes been sitting idle for about a week now.
Thanks Roger
 
Back
Top