• 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

Odyssey Charge State

I had 12.81v by Fluke on parked bike, but clock and connections had not been disconnected yet. So, I'm happy with my charge rate and will now see how my storage charger cooperates during off season.
 
Still waiting for delivery of the charger. In the meantime ...

Below is a chart from East Penn (many thanks to Terry), a manufacturer of Gel and AGM batteries. On the chart you can see the expected battery voltage when an AGM battery is loaded at the 1-3 hour discharge rate. That would mean that my fully charged PC680 (16 aH) when loaded at a 5-16 amps should still have a 12.5 to 12.6V output. It seems pretty clear to me that my battery is not fully charged since at key-on the voltage drops from 12.9 to 12.1V or so.

With my key on I've got about a 4 amp load from the low beam and maybe a couple amps for everything else. That is within the 1-3 hour discharge range on the chart. I plan to measure my key on battery load.

While I wait for the charger and some time, I've been looking at old GS-911 log files to find the initial cold-start injector pulse length. It used to be about 3.8 mS. Lately it has been in the 4.5 mS range, up to 5.5 mS. This is due to guess what? Low battery voltage. The Motronic is trying to compensate, it's just not adding enough time. I now feel that I have triangulated this pretty well. Everything points to low voltage at the injectors due to the battery (and perhaps the starter).

It will still be a day or two before I begin reconditioning the battery.
RB

2009414153615343.jpg
 
The injectors are grounded through the ECM

The injectors see system voltage all the time and are fired when the ground circuit is completed through the ECM. Before I was so sure about voltage as the cause, I don't discount it, but a good meter on the injectors and at a high load circuit to confirm the voltage would convince me.

Ground resistance in general or resistance through the switching{(power?) I can't remember the correct term for the big transistors} transistors will reduce current to the injectors.

I couldn't make a decision about battery discharge or the right charging method based on the information seen here to date.
 
Very thorough as always Roger.

Although I tend to treat batteries like crap and expect them to last forever, you guys have got my attention.
My PC-680 is at 12.48V unloaded, and quickly drops to 11.6V when I turn on the key so even though the bike starts easily, I'm missing some capacity.
I'm going to go through the recovery process as outlined by GSAddict.
I'd also like to modify the voltage regulator.

I have a couple of questions for GSAddict:
- Odyssey recommends a limit of 15V yet you specify 17.5V, current limited to 950mA. Are you suggesting the battery is already damaged so this higher voltage is warranted but may make it worse?
- Your post #14 includes a link to info regarding the bike voltage regulator but I can't get in without a password. Am I missing something?
 
Very thorough as always Roger.

Although I tend to treat batteries like crap and expect them to last forever, you guys have got my attention.
My PC-680 is at 12.48V unloaded, and quickly drops to 11.6V when I turn on the key so even though the bike starts easily, I'm missing some capacity.
I'm going to go through the recovery process as outlined by GSAddict.
I'd also like to modify the voltage regulator.

I have a couple of questions for GSAddict:
- Odyssey recommends a limit of 15V yet you specify 17.5V, current limited to 950mA. Are you suggesting the battery is already damaged so this higher voltage is warranted but may make it worse?
- Your post #14 includes a link to info regarding the bike voltage regulator but I can't get in without a password. Am I missing something?

When doing the "recovery" charge, voltage can go as high as that because of the current limiting of 950ma. At that current thermal runaway is not possible. This voltage is required to remove sulfation from the electrodes.
 
The injectors see system voltage all the time and are fired when the ground circuit is completed through the ECM. Before I was so sure about voltage as the cause, I don't discount it, but a good meter on the injectors and at a high load circuit to confirm the voltage would convince me.

Ground resistance in general or resistance through the switching{(power?) I can't remember the correct term for the big transistors} transistors will reduce current to the injectors.

I couldn't make a decision about battery discharge or the right charging method based on the information seen here to date.

Two things I've confirmed, the loaded ground voltage drop among battery, frame and engine is a few millivolts and the voltage at the injectors is the same as the voltage measured by the Motronic.

The injectors are fairly sensitive to voltage, below is a typical chart. Either my injectors have gotten slower at low voltage with age or the Motronic doesn't adequately compensate for low voltage. The dead time at 8V is 2.5 times as long as 13V.

InjectorLatencyCompensation.png


Very thorough as always Roger.

Although I tend to treat batteries like crap and expect them to last forever, you guys have got my attention.
My PC-680 is at 12.48V unloaded, and quickly drops to 11.6V when I turn on the key so even though the bike starts easily ...

Mine starts easily too except for the first time, just somewhat slow and lean (which I see with the LC-1).
 
To recondition its AGM batteries, Odyssey recommends discharging the battery at or slower than the 5-hour rate to a final voltage of 10.02V.

The PC680 is a 16 aH battery so the 5 hr rate is about 3 amps. I used a 6 ohm resistor and the discharge started at 2 amps, averaging 1.8 amps. For a fully charged battery, it should have taken just under 9 hours to discharge. The Open Circuit Voltage at start was 12.9V which theoretically indicates a full charge.

It took about 6 hours to get to 10 volts. My "fully charged" battery (by the alternator and then BT Jr.) sourced 11 amp-hours or 70% of the claimed 16 amp-hours capacity.

(Coming at it a different way, using the AGM curves for discharge-rate vs voltage, would say the 12.1 volts I get with the headlight on means that it is about 60-65% charged.)

I started recharging it this morning with the Odyssey 12 amp charger. After that Odyssey says to discharge it again and see how many amp-hours to 10.02V, doing it up to five times or until it stops improving.
RB
 
To recondition its AGM batteries, Odyssey recommends discharging the battery at or slower than the 5-hour rate to a final voltage of 10.02V.

The PC680 is a 16 aH battery so the 5 hr rate is about 3 amps. I used a 6 ohm resistor and the discharge started at 2 amps, averaging 1.8 amps. For a fully charged battery, it should have taken just under 9 hours to discharge. The Open Circuit Voltage at start was 12.9V which theoretically indicates a full charge.

It took about 6 hours to get to 10 volts. My "fully charged" battery (by the alternator and then BT Jr.) sourced 11 amp-hours or 70% of the claimed 16 amp-hours capacity.

(Coming at it a different way, using the AGM curves for discharge-rate vs voltage, would say the 12.1 volts I get with the headlight on means that it is about 60-65% charged.)

I started recharging it this morning with the Odyssey 12 amp charger. After that Odyssey says to discharge it again and see how many amp-hours to 10.02V, doing it up to five times or until it stops improving.
RB

Been there done that a few times. Always had good success, keep us posted.
Note: I have discharged at higher rates with good results.
 
Been there done that a few times. Always had good success, keep us posted.
Note: I have discharged at higher rates with good results.

Good to know. I have a second 6 ohm that I could use ...

Amazing that I could run so long with a 2/3 charge ...
 
Very good point! I used a 50 watt, didn't attached it to a heat sink and it got to 100 degrees centigrade, and nearly seared my fingers ... ;)

Measured the filament resistance of a plain old 60 watt incandescent light bulb (17 ohms) so connect 2 in parallel (8.5 ohms) for a load, maybe? I connected one bulb across the battery and it got quite warm to the touch and glowed dimly.
 
Finished the first recharge. The Odyssey charger, like some others, charges in three steps: bulk at high current up to 14.7V, absorption at lower current at 14.7V, and float between 13.5 and 14.1V.

The bulk recharge to 14.7 volts took about 15 hours, absorption took another few overnight.

The open circuit voltage is now 12.9 volts, the same as before. However, the key-on, headlight on-voltage is now 12.6-12.7 volts (increased from 12.1 volts). I should run another discharge/recharge cycle to see if the battery capacity has been restored but I'll see how it starts first.
RB
 
or 223 ohms in this case

That and a 60 watt bulb is rated or designed to operate with 120 VAC. Bad idea. Would take awhile to discharge the bat over a 200 ohm load at the rate of 50 ma. Not sure what the filament resistance would be when applying 12 vdc to a 60 watt 120 VAC bulb. Guess if I'm bored I could find out easy enough.
 
That and a 60 watt bulb is rated or designed to operate with 120 VAC. Bad idea. Would take awhile to discharge the bat over a 200 ohm load at the rate of 50 ma. Not sure what the filament resistance would be when applying 12 vdc to a 60 watt 120 VAC bulb. Guess if I'm bored I could find out easy enough.

At the risk of starting a lightbulb thread......:stick
 
I'd get a 500A carbon pile load tester from Harbor Freight and do a load test on a fully charged battery. They're are on sale now and then (like right now) for $55. You won't have to guess the next time you have a battery or starting issue.

The load test is a test of a battery's internal resistance.

Also get a 200A shunt off eBay. Along with some cables and a multimeter with Max or Peak Hold, you can take a current draw measurement of your starter in action.

Of course another way would be to use another battery on loan to you and not a car battery that has gobs more CCA which might overshadow a starter issue.

Your alternator should be putting out 14.0 to 14.2V off idle.

Or make your own with SS welding wire. I have a commercial 500a unit but prefer this one for MC batteries.


.
 
Starting Test After PC680 Battery Cycling (GSAddict got it right and helped a lot behind the scenes)
I went out to the garage, set everything up very carefully to record data: er, er, vroom--1.1 seconds! The best start in 4-5 months. I would say normal and as expected. Success!

Setting up carefully means turning the key on but with the headlight fuse pulled, getting the gs-911 set up and configured for logging, putting a jumper in for the fuel pump relay (5 amps) and LC-1 (2 amps for the heater), starting the LC-1 logging software, putting the headlight fuse back in, and then pressing the start button. Total elapsed time with the key on before starting of about 2 minutes. The draw for the headlamp, fuel pump and LC-1 is about 13-14 amps total (plus whatever the panel lamps, gauges and ABS draw).

Fast idle lever in the detent position (1.92 degrees TPS open) and not touched from pressing the start button until the bike was warmed up and in closed loop.

By the time I pressed start, the voltage had dropped from 12.5V to 12.1 volts (that sounds worse than it is since I had a 14 amp load on it). Minimum voltage in the start log was 10V, low but not nearly as low as the 7-8V I'd seen previously.

Of note, the LC-1, which for the past several months had been resetting each time I started the bike, did not reset itself. I attribute this to better battery response.

After starting, the RPM, spark advance behaved well for the first time in a while. The idling quality was very good.

Also of note, the injection pulse times were longer than anything I've seen previously. Just before and just after start about 5 mS. But on the battery dip to 10V, a single pulse of 6.2 mS, the longest I've seen. It makes me believe that the battery induced starting voltage dips were affecting the Motronic.

Summary
At the start of this thread I was investigating slow starting on the first start of the day, and leanness in the first minute or so of idling. I found a fouled lower plug (replace all four plugs with Bosch), a weak stick coil (replaced both) and a significantly undercharged PC680 battery. Although the plugs and coils needed replacement and my bike better after replacement, it was the low voltage caused by the perpetually undercharged PC680 that was the root of the problem.

The PC680 was undercharged due to a mismatch with the R1150 alternator which only puts out 13.6 - 13.7 volts on average. That's only enough for a trickle charge of the PC680 AGM battery. I made the problem worse by keeping the battery on a Battery Tender Jr. whose float voltage of 13.1 V can lead to battery sulfation.

Low voltage at the fuel injectors causes their turn-on time to increase and thus shortens the duration of the cold start fueling, leading to lean mixtures just when they should be very rich.

The solution to my undercharged PC680 will be to add a diode to the Alternator's Voltage Regulator thereby increasing my alternator's output voltage to 14.4 to 14.7 volts. I will no longer keep my bike on a trickle charger, the PC680 doesn't need it.

Next steps are to cycle the battery again to see if it is at full capacity and to install the diode.

RB
 
Last edited:
Back
Top