Charging System Failure Modes

[RossKean]

I am in the midst of sorting out some issues with electrical system on my bike (2007 "A"). I'm fairly sure where I am headed with it but wanted a little input on some stuff. In particular, I am interested in failure modes for some of the key components. If good information is generated here, it might be a useful reference to assist others with charging issues.

R/R (Regulator/rectifier) - does this unit typically produce a "hard" fail. i.e., Does it go instantly from working properly to complete failure? Is there a short (or long) interim period where it may not produce enough amps although still work, after a fashion? Will it produce the amps but perhaps not properly rectify the output from the stator? How many people have had to replace a R/R on an FJR?

Stator - I guess the same main questions... Assuming that the magnets remain intact and in their proper places, is a hard failure (i.e. wire short or open) most likely? Do you expect to see gradual "weakening" in terms of the stator being able to produce enough amps to keep the bike (and battery) happy? I assume this could be caused by magnets weakening due to heat exposure? Other causes?

Battery - I pretty much understand the (multitude of) failure modes possible with lead-acid batteries but I thought I would include it for the sake of information for anyone who encounters this thread. (Just dealing with symptoms, not the chemistry or physical conditions for this - those are separate topics.)

AGM (Absorbent glass mat) batteries can simply degrade over time with gradually lowered capacity (soft failure). In addition to simply having lowered capacity, the battery will also have diminished ability to maintain voltage under load due to increased internal resistance. Such batteries are particularly susceptible to current delivery problems when operating at lower temperatures. All batteries will eventually succumb to this. Older AGM batteries will often exhibit a greater self-discharge rate which means they can function normally but end out "flat" after a week of inactivity.

A cell can short which may give you a low "charged" voltage. Often not a "hard" short but very much lowered resistance between adjacent cells. The battery may appear to take a charge and function normally for a time and the next day it could be showing 10V. More common with liquid electrolyte "car" batteries.

Open (total) failure. 'Nuff said; open internal connection somewhere and it is toast. Can happen with a new battery.

AGMs don't like overcharge or deep-discharge. Will also last longer in cooler temperatures.

Wiring - I think this is where I am going with this although it will depend on what I hear back from people. I read dcarver's (long) thread on the issues he had a few years ago https://www.fjrforum.com/forum//index.php/topic/146331-krzy8-gen2-charging-circuit/ . Also another thread - https://www.fjrforum.com/forum//index.php/topic/168308-how-to-check-rr-voltage-on-a-gen-ii/ . It is clear that a number of people had chronic voltage/charging issues fixed by using an auxiliary harness between the R/R and the battery. I don't dispute the results but I am wondering if anyone has any insight as to why this happened; especially after many years of "normal" operation? Has anyone identified the specific connections where high(er) resistance has caused a voltage drop on the order of 0.5 V to 1.0 V? Are there any other wiring-related problems people have discovered which degraded charging performance? (Battery connections are an obvious (common) problem and I have double checked this)

Anyway, my situation is that I am seeing less voltage at the battery than I like (13.3 at idle and 13.7 or so at 2500 rpm). I have had a couple of instances where the battery voltage had dropped to a point where the bike did not run properly and a couple of times I needed to boost it to start. Although battery condition is questionable (163,000 miles and over 9 years), it will charge properly with an external charger and will operate the bike normally with respect to starting etc. I am thinking it is at least partially charging while operating but perhaps not enough to sustain it. I am replacing the battery and will check to see if the voltage issues remain. If that is the case (and I think it will be), I will need to look into wiring harness, R/R and or stator. I want to be back to a completely healthy electrical system before I commit to a couple of long trips planned for next season.

I would appreciate other suggestions of things I should consider as well as information/experiences with R/R or stator problems. -Thanks

Edit: I found some interesting videos on checking MOSFETs, R/R and stators (as well as installation of their products) at roadstercycle.com.

https://roadstercycle.com/Roadstercycle.com%20Videos.htm

 

[RaYzerman19]

I'd guess you have to determine if your battery is sucking up charging voltage...... time for a new one don't you think? You can quick check stator windings to each other (0.13-0.19 ohms) to see if you have an open or shorted stator. My guess is it's OK since it's putting out over 13. Regulators are usually pretty reliable...... Have you got another battery you can try, to at least confirm it is or isn't the issue? I'm leaning towards battery.

 

[RossKean]

I'd guess you have to determine if your battery is sucking up charging voltage...... time for a new one don't you think? You can quick check stator windings to each other (0.13-0.19 ohms) to see if you have an open or shorted stator. My guess is it's OK since it's putting out over 13. Regulators are usually pretty reliable...... Have you got another battery you can try, to at least confirm it is or isn't the issue? I'm leaning towards battery.

I'm going to do the battery before anything else. After this much time, it certainly doesn't owe me anything. I'm pretty sure that capacity is down and self-discharge rate is higher than on a new battery. 24 hours after an external charge, it still sits at 12.9 V or so (unloaded) and should run the bike OK as long as the charging system is up to snuff. Still, a new battery is a good idea after 9+ years. The charging system works, after a fashion, but system voltage is lower than I would prefer. Once battery is eliminated from the equation, I'm on to looking at other stuff. This is why I was wondering if the R/R or stator could be "weak" rather than failed.

 

[RaYzerman19]

I have limited experience but was of the belief stators could have an individual coil fail, but regulators were either good or bad...... I'll betcha Professor Ionbeam will chime in.....

How to check a Mofset.....

 

[Bounce]

Yeah. After 9 years you're living on borrowed time. Passable for local failures where you have access to known deals/dealers and service. Not so much in BFE.

 

[PhilJet09]

Edit: I think you should get back to us after replacing the 9 year old battery.

 

[dcarver]

My thread WAS long, Ross. And once again my charging is less than before. I suspect stator but have not dug into it yet. I appreciate your desire to know the why.

 

[ionbeam]

Reader's Digest Condensed version: Pull the DC output connector off the R/R. With the engine at idle measure a minimum of 13.8 volts to a maximum of 15.8 volts. If your FJR passes this charging system test, replace your 9 year old battery.

If the voltage is still low at the new battery terminals you have voltage drops throughout your main electrical system harness. There are several connections between the R/R and the battery and even more between the battery and the main ignition switched electrical system. Every connection contributes a small voltage drop, including drops across the fuses. These are the demons that dcarver has been trying to exorcise.

R/R (Regulator/rectifier) - does this unit typically produce a "hard" fail. i.e., Does it go instantly from working properly to complete failure? Is there a short (or long) interim period where it may not produce enough amps although still work, after a fashion? Will it produce the amps but perhaps not properly rectify the output from the stator? How many people have had to replace a R/R on an FJR?

The R/R will typically produce a hard failure and will most likely do it instantly; it most likely won't be a complete failure but there will be a substantial voltage issue. A modern R/R is highly unlikely to fail unless the owner has done something to antagonize it. The rectifier section is very unlikely to fail, if it does fail, the R/R will have very low DC voltage output. If the DC regulator section fails, it is more likely to have a HIGH voltage output than a low voltage output. In old school wet cell batteries with acid/water filler caps, the first hint that the DC regulator has gone bad is low water in the battery as the over voltage boils out the water from overcharging.

There have been very few R/R issues reported here. What is harder to get information on is how many R/Rs were replaced where it was indeed the root cause of charging and electrical system issues.

Stator - I guess the same main questions... Assuming that the magnets remain intact and in their proper places, is a hard failure (i.e. wire short or open) most likely? Do you expect to see gradual "weakening" in terms of the stator being able to produce enough amps to keep the bike (and battery) happy? I assume this could be caused by magnets weakening due to heat exposure? Other causes?

The magnetic rotor is a uniform looking, cast piece, your assumption that the rotor is intact is good, the FJR's permanent magnet rotor won't weaken. The stator is much more likely to fail shorted than open. There won't be a gradual weakening, it will be good then bad. There could be a very quick period where the stator is intermittent but it will quickly become a hard failure. The vast majority of stator failures is caused by the enameling on the winding wires burning and cracking. When this happens the windings short together or the windings short to ground.

Battery - I pretty much understand the (multitude of) failure modes possible...

Done.

Wiring - I think this is where I am going with this although it will depend on what I hear back from people. I read dcarver's (long) thread on the issues...I don't dispute the results but I am wondering if anyone has any insight as to why this happened; especially after many years of "normal" operation? Has anyone identified the specific connections where high(er) resistance has caused a voltage drop on the order of 0.5 V to 1.0 V? Are there any other wiring-related problems people have discovered which degraded charging performance? (Battery connections are an obvious (common) problem and I have double checked this)

Every place there is a fuse, harness connection or stud connection there will be a small voltage drop. In the Gen II grounding system there are crimps within the harness run which may also develop voltage drops. Over time dirt and oxidation happen at every connection, once it starts it will slowly get worse. Fuses have a normal small voltage drop across the component. On my Gen I the R/R to battery voltage drop was primarily at the stud on the starter relay where the R/R DC output wires are joined together with the heavy battery cables.

Anyway, my situation is that I am seeing less voltage at the battery than I like (13.3 at idle and 13.7 or so at 2500 rpm). I have had a couple of instances where the battery voltage had dropped to a point where the bike did not run properly and a couple of times I needed to boost it to start. Although battery condition is questionable (163,000 miles and over 9 years), it will charge properly with an external charger and will operate the bike normally with respect to starting etc. I am thinking it is at least partially charging while operating but perhaps not enough to sustain it...

Testing the stator: The average DMM can't do a stator winding resistance check. Due to differences in the Gen I & Gen II charging systems the actual stator voltages may be different than the stated voltages in the tests below. What is key is that the voltages are all three equal to within 2-3 volts.

Pull the 3 wire stator connector off. With the engine OFF, set the DMM to the highest resistance scale typically 10 meg or 20 meg.

measure each pin in the connector to ground, the readings should be close to infinite resistance (no reading at all on the meter)

this verifies that none of the windings are shorted to ground

Set the DMM to AC volts, the range should be set for up to 100 VAC

Start the engine, let it idle then carefully --

measure all three pins with the red and black DMM probes going from any pin to any pin

at idle there will be roughly 20 VAC, don't sweat the actual voltage, the key is that all the voltages are equal to within 2 - 3 volts

if the voltage is wrong it will be wrong by a lot, beyond any doubt that it is different

Bring the engine RPMs up to ~5k

measure roughly 60 - 80 VAC any pin to any pin, don't sweat the actual voltage, the key is that all the voltages are equal to within 2 - 3 volts

if the voltage is wrong it will be wrong by a lot, beyond any doubt that it is different

the voltage tests verify that none of the windings are shorted together and none of the windings are open

If the stator meets these conditions it is good, no question about it

Testing the voltage regulator: (put the stator connector back on), pull off the the connector with the two DC voltage wires

Set the DMM for 20 VDC

Start the engine, let it idle

measure 13.8 to 15.8 VDC on the R/R output pins, if it passes this test the regulator is good, no question about it

Passing the stator tests and the R/R test verifies that the charging and voltage regulation system is good.

 

[RossKean]

^^^

Many thanks, Alan

Getting a new battery later in the week. If there are still issues, I will be diving into the troubleshooting sequence you have laid out in such concise detail.

 

[Harald]

I was experiencing the same voltage drop on my 07 and in the interest of expediency installed the bypass harness from roadstercycle.com. There was an immediate correction and I normally see 14.1 or 14.2 running around and no longer get into the danger zone when running the grips and vest on high in the winter.

The only time I see voltage less than 13.2 is when the bike is at idle and the fans kick on. I know it's time to investigate the fan wiring or simply replace it with larger gage wiring, but I'm not that concerned because the fans aren't usually kicking on when I have all that heated gear cranking while at idle.

 

[mcatrophy]

I was experiencing the same voltage drop on my 07 and in the interest of expediency installed the bypass harness from roadstercycle.com. There was an immediate correction and I normally see 14.1 or 14.2 running around and no longer get into the danger zone when running the grips and vest on high in the winter.
The only time I see voltage less than 13.2 is when the bike is at idle and the fans kick on. I know it's time to investigate the fan wiring or simply replace it with larger gage wiring, but I'm not that concerned because the fans aren't usually kicking on when I have all that heated gear cranking while at idle.

It won't be the fan wiring affecting the battery voltage. Any voltage drop in their wiring would result in a lower voltage on the fans, not at the battery. The only ways the fans coming on would result in lower than expected battery voiltage is either the charging system is degraded in some way or the fans are drawing more current than they should.

 

[Harald]

I was experiencing the same voltage drop on my 07 and in the interest of expediency installed the bypass harness from roadstercycle.com. There was an immediate correction and I normally see 14.1 or 14.2 running around and no longer get into the danger zone when running the grips and vest on high in the winter.

The only time I see voltage less than 13.2 is when the bike is at idle and the fans kick on. I know it's time to investigate the fan wiring or simply replace it with larger gage wiring, but I'm not that concerned because the fans aren't usually kicking on when I have all that heated gear cranking while at idle.

It won't be the fan wiring affecting the battery voltage. Any voltage drop in their wiring would result in a lower voltage on the fans, not at the battery. The only ways the fans coming on would result in lower than expected battery voiltage is either the charging system is degraded in some way or the fans are drawing more current than they should.

Hmmm. So I wonder what's normal for voltage at the battery when the bike is at idle and the fans kick on?

 

[mcatrophy]

...Hmmm. So I wonder what's normal for voltage at the battery when the bike is at idle and the fans kick on?

Sorry, can't help you there, that would be a very rare event for me. I would expect it to be below the full 14.2, though, since in my experience any additional load does drop it a little.

Anything above 12.8 does mean the battery is not being discharged, since that's around the open circuit fully charged voltage.

 

[ionbeam]

Depending on how hot the day was and how long I had been sitting on my '04 the electrical system could measure anyplace from 13.4 to 12.6(!) volts when the fans turned on at traffic lights or during a protracted sit at a stop sign. Even when my '04 was new the voltage would drop to 13.4 to 13.6 volts when the fans turned on. Even in colder weather, if the fans turned on at a traffic light the voltage would drop low enough that my heated grips controller would shut off the grips (it used ~13.0 volt reference to determine if the engine was running or not).

[Mostly useless posting below]

Mayhaps mcatrophy can add/clarify or debunk the following.

I have looked at a couple of FJRs, both Gen 1 using an oscilloscope to measure the electrical system. Both systems were 'dirty', there was a lot of electrical noise with several different types of noise present at different frequencies/bursts and modulations. Multimeters and dedicated volt meters have different methods of dealing with this noise and the user sees slightly different voltages depending on the meter type, even though the meters are looking at the same point. The fans and the horns are 'dirty' devices causing more than normal electrical system noise. It would sure be interesting to look at the electrical system with an o'scope as various equipment turns on and off and see what the real voltage implications are for the battery's charging voltage. My Datel, calibrated Fluke and calibrated Beckman meters never measured the same voltage when all were connected at the same time to the same point. This is one of the rare occasions where a quality needle movement meter may give superior results. FWIW, the scope ground was floating using the battery - terminal as the ground reference.

[/Mostly useless posting]

 

[mcatrophy]

Depending on how hot the day was and how long I had been sitting on my '04 the electrical system could measure anyplace from 13.4 to 12.6(!) volts when the fans turned on at traffic lights or during a protracted sit at a stop sign. Even when my '04 was new the voltage would drop to 13.4 to 13.6 volts when the fans turned on. Even in colder weather, if the fans turned on at a traffic light the voltage would drop low enough that my heated grips controller would shut off the grips (it used ~13.0 volt reference to determine if the engine was running or not).
[Mostly useless posting below]

Mayhaps mcatrophy can add/clarify or debunk the following.

I have looked at a couple of FJRs, both Gen 1 using an oscilloscope to measure the electrical system. Both systems were 'dirty', there was a lot of electrical noise with several different types of noise present at different frequencies/bursts and modulations. Multimeters and dedicated volt meters have different methods of dealing with this noise and the user sees slightly different voltages depending on the meter type, even though the meters are looking at the same point. The fans and the horns are 'dirty' devices causing more than normal electrical system noise. It would sure be interesting to look at the electrical system with an o'scope as various equipment turns on and off and see what the real voltage implications are for the battery's charging voltage. My Datel, calibrated Fluke and calibrated Beckman meters never measured the same voltage when all were connected at the same time to the same point. This is one of the rare occasions where a quality needle movement meter may give superior results. FWIW, the scope ground was floating using the battery - terminal as the ground reference.

[/Mostly useless posting]

Your first paragraph refers to your Gen 1 which, unless you had an after-market stator, gives noticeably lower output at tickover than Harold's Gen 2 should. I'd be very surprised if a Gen 2 dropped as low as 13.6 with just the fans on as an extra load (unless the engine idle speed was low, of course).

As for "dirty" electrical loads, yes, I agree that different meters will give different results when there is significant noise on the wiring. Again, I've had almost zero experience of my fans coming on, it's only happened when I've been in heavy traffic in high temperatures. In the possibly half a dozen times this has happened in my ten years of FJR ownership, I don't ever remember looking at my voltmeter, let alone hanging another one on the battery terminals to compare.

So I'm in no position to comment.

For what it's worth, my home-made bike's voltmeter takes instantaneous (well, something like 15 microsecond samples into the micro controller's ADC) at a rate of 1kH, it then averages every 256 readings and updates the display at that rate, so almost 4 display updates per second. It's always agreed within 0.1 volts (its display resolution) with my commercial digital multimeter, that I have checked against a known standard at my (ex) place of work.

 

[ionbeam]

For what it's worth, my home-made bike's voltmeter takes instantaneous (well, something like 15 microsecond samples into the micro controller's ADC) at a rate of 1kH, it then averages every 256 readings and updates the display at that rate, so almost 4 display updates per second. It's always agreed within 0.1 volts (its display resolution) with my commercial digital multimeter

Ohhh, digital filtering!

 

[mcatrophy]

For what it's worth, my home-made bike's voltmeter takes instantaneous (well, something like 15 microsecond samples into the micro controller's ADC) at a rate of 1kH, it then averages every 256 readings and updates the display at that rate, so almost 4 display updates per second. It's always agreed within 0.1 volts (its display resolution) with my commercial digital multimeter

:wub: Ohhh, digital filtering!

Well, sort of, just add the samples and shift some bits. Easy to implement in PIC assembler . Hardly cutting edge multi-pole elliptical stuff. But it's nicely responsive enough to follow things like indicators when at idle, without being so rapid that you can't make out the readings.
Gives me the information I want. The reality is I've rarely had any occasions where it's given me any cause for concern, only showing the voltage approaching my heated vest controller low voltage cut-off value (12.8V) when sat in Derby rush-hour traffic for too long while keeping toasty in sub-freezing temperatures.

 

[Constant Mesh]

One thing you can obviously do when the fans switch on is to note the voltage at idle and then rev the engine to 3K or so and see how much the voltage increases. At some rpm, if the charging system is working correctly, the voltage will stabilize at its nominal regulated value of 14.0 volts.

We know that the charging system output current is max at about 5K rpm. Don't know if it's a straight line function from 0 to 5K but it's an increasing function.

Probably the simplest way to get max voltage at the battery is to run an express conductor directly from the R/R's negative black conductor to the negative terminal of the battery. Who knows how many twists and turns that return circuit makes between those two points?

If you had a keen ear you could probably detect a small change in fan speed as a function of the fan's terminal voltage.

 

[Brodie]

I've been enjoying the discussion on this topic by some of the 'smartest people' we have on this forum. My knowledge has been boosted many times over by their freely sharing their thoughts and experiences through the years. Thank you gentlemen.

One observation I'd like to add to this subject is how dark the copper is on both the positive and negative heavy main wires connecting the battery to the voltage regulator. I still have the wiring harnesses I salvaged from my first '06 AE, wrecked back in 2010. It's been a donor for several components for the various custom harnesses I built for my current '08.

1. The length between battery and voltage regulator is approx. 6 1/2 feet.

2. The wire gage seems to be the metric equivalent to 12 gage.

3. There are several crimps along the length of both conductors.

The dark reddish color of the copper tells me that their design load is near the upper limit for the materials in these circuits. The wire is heating up due to the increased resistance from the oxidation on the surface of the copper. With heat comes more oxidation. What I see is a very slow thermal runaway condition happening through the years.

My advice to Ross is to go ahead and replace the battery, you got your money's worth out of the old one. Then put the Carver bypass harness in place. Install a Datel if you haven't already and call it good.

Brodie