Battery light flicker

[docc]

To complicate matters, it doesn't do it every time. I've not noticed it on start-up, only coming home.

 

I've been planning to add a dedicated reference wire from the battery and stop taking it from the headlamp circuit as my headlamp has dedicated harness of its own.

[luhbo]

I mentioned that I have a 2 colour LED circuit from Wayne Orwig directly attached to the instrument lighting, going from dark red over orange to a bright green. Although I have the main lights feeded through relais I definitely could see the colour of the LED changing, obviously not following any concrete scheme. Right after starting the engine it was well green, then it faded to a beginning orange, darker at higher revs, better (greener) under wet conditions, some days it was the other way round. While also thinking about switching the reference from the parking light directly to the battery I found something different to be the cause for this problem: while checking the main connectors I remembered having seen an additional separately boxed 30A fuse directly behind the aftermarket regulator on this new bike. It turned out that the reason for all decribed above was that one of its legs was totaly corroded, the green cover nearly completely molten, all just a brittle charred mess.

So, let's assume that the reg/rec system either works or not, then it might be obviously better to have the reference for any type of voltage indicator not directly at the battery but instead somewhere in the harness. In my case I could have seen earlier that something is about to go south in my harness. The LED was telling it, I was just thinking the unlikely things first. Again.

 

In short: picking up the voltage reference directly at the battery might not be the best solution.

[docc]

I like the voltmeter idea, including the LED indicators. I wish either Garmin (GPS) or Speedhut (speedometer) would understand that voltage is an important parameter to display on a motorcycle.

I decided to check some grounding under the seat since I had the battery out for a while recently while doing the clutch-and-such. Negative battery terminal took maybe an 1/8 turn to tighten. Everything took a little on both sides (I have junction blocks with additional connections on both sides.)

Loosest was the ground to the ECU case. I think that "vibration isolator" mount has come apart again. But I got the ground terminal good and tight. I'll know in the next week if the flicker quits (since it has been intermittent).

Hubert, thanks for helping me focus on the harness and connections! Hopefully, these simple tightening procedures will make the difference!

[Chuck]

To complicate matters, it doesn't do it every time. I've not noticed it on start-up, only coming home.

 

I've been planning to add a dedicated reference wire from the battery and stop taking it from the headlamp circuit as my headlamp has dedicated harness of its own.

There might be a reason that they deleted that light on later models.   

[bbolesaz]

http://www.kaiyoudai.com/bikes/duca/img/Regulator_Schematic[1].pdf

[docc]

Well, I've had the warning light give me fair warning when one of the yellow wires from the stator had broken. Enough warning to travel toward home almost fifty miles where I could be "collected."   (still fifty miles from home).

 

The only other complete charge failure I've had was a blown 30 amp fuse (from a loose positive battery terminal connection). That one, the light and poor running both appeared suddenly. Maybe two-three miles later, I was stranded.

 

Started carrying jumper cables after that.

[Guzzi2Go]

The light goes on when battery's voltage is higher than regulator's, which would typically be the case when the regulator breaks down (in a special way, have to add). Atypical case, and probably the reason why we often see it flicker, is that at idle RPM regulator's voltage is lower than battery's, especially after a vigorous ride when the battery is fully charged.

 

Alternator's voltage is merely 15V@1k RPM and it produces ~10A. If you consider typical voltage drops over diodes/regulators and the fact that you need about 10A to run the lamps, pumps, injectors, at this point, battery will be feeding the regulator, not the other way around.

 

There! I said it!   

[docc]

No more flicker. 

 

Slightly loose battery connections and, especially, the loose ECU case ground.  It's been 2,000 miles/ 3.226 km since hooking the battery back up from the clutch replacement, so I hope to remember next time to re-tighten after I've had the battery unhooked.

 

Not sure if the ECU ground could have been the main culprit.   It was certainly actually loose.

[Guzzi2Go]

The lamp is controlled by the regulator, not ECU. So no, ECU ground is not the culprit. 

 

Flicker will be back! ;-)

[docc]

The lamp is controlled by the regulator, not ECU. So no, ECU ground is not the culprit. 

 

Flicker will be back! ;-)

Haha, yes. Some things do not stay "fixed."

[Kiwi_Roy]

I wouldn't worry too much if it's charging, perhaps check the diodes in your regulator just to be sure its not just charging on only one.

 

Diode tester - Red probe to Yellow wire one then the other, Black probe to Red wire should read 0.5 to 0.6 Volts

 

The charge light is pretty useless IMHO. If you blow the light fuse it doesn't go

 

If you ever lose the tach that's a sign there's something amiss, it's fed from the same Red/Black wire after the headlight Relay.

 

A long time ago I posted the regulator schematic on here, that was before Photopirate pulled the fuse. If anyones interested in re-posting it send me an e-mail address.

[bbolesaz]

KiwiRoy, look a few posts above, I found it on google images.

[docc]

KiwiRoy, look a few posts above, I found it on google images.

Yeah, cool you found that. I couldn't get it to "display", so I've hosted it on ImgZeit:

[Guzzi2Go]

But that can't be right, can it? The Q3 requires positive GS of 2-4V to conduct, this will never be the case with the Z2 as shown. Also the SCRs seem to be wrong way round and unprotected. The Q2 seems to be more or less permanently on, driven by the alternator's negative half-period...

 

Has anyone tried simulating this circuit in e.g. SPICE?

[LowRyter]

just wanted to pass on my business with that light. I was on a campout and did about 300 miles over the weekend.

 

I noticed when I started the bike cold that the light never came on.  There after it did the normal flicker upon turning the key and starting and running.

 

I changed out that bulb on my EV a few years ago and it was heck getting it to seat.  I never got it to light when I was installing and gave up trying.  Then later it seemed to work fine.  In fact, it's bright on until it starts (as it should be?), then flickers at idle and goes out when revved.