Battery warning light

Can somebody tell me what condition must be present for the Battery warning light to illuminate? I do not have an owners manual or a FSM, but I DO have a Haynes manual....unfortunately the Haynes does not explain about the warning lights or conditions with which they go off.
Thanks
Pat
1982 XJ750 Maxim

 

If your setup hasn't been hacked (unlikely, but possible), the battery warning system will be monitoring the voltage from a probe plugged into one of the battery's cells (the third cell, I think).

This probe has a metal tip that dips into the electrolyte. The warning system will trigger if the battery's voltage it too low, or if the electrolyte level is too low (causing the probe tip to be uncovered, so it can no-longer read the voltage).

In most XJ's these days, though, the battery light is likely to be on because an owner of the bike installed a battery that did not have a port into which the probe could be installed, and they didn't modify the battery to support it. Or because they'd lost the probe entirely.

Edit: Pretty soon you should get some follow-up posts from other forum members telling you how you can hack the warning system so that it will no-longer bug you about low electrolyte (or what it thinks is low electrolyte).

 

Whoops....never-mind. Just found it in the Haynes.....

Apparently it cycles when the electrolyte in the battery is low (or in my case when it's not even in the battery). I guess I'll have to start looking around for the probe....was not in the battery when I bought the bike.

Is that the only condition? Does it have any voltage warning signs?

 

Here's the hack I used. It's on here a bunch of places. Take about 2k Ohm of resistors, splice wires on both ends, run your sensor wire into one end, run the other end into a keyed power source, I used the taillight wire, since it was right there under the seat.

The beauty of this workaround, is that it is actually better than the original sensor, in my opinion, because if your voltage is low it will still show it, because the wire you're connected to will give it less power. Becoming more of a "voltage" light than a "battery fluid" sensor.

I have had it come on when my old battery was dying, or if I ran the lights too long without the motor running, or whatever, so you're not "Bypassing" the sensor, really, just changing what and how it measures.

 

I don't know why I'm posting this, since apparently nobody read the post I made earlier, but the original setup, with a probe in the battery, checks for both electrolyte level AND voltage.

 

Perhaps I should have been more succinct SQL....WITHOUT the probe hooked up....does the idiot light serve any other purpose? I'm guessing no...

 

Without the probe hooked up, and without hacking it like Mike suggests, no, it serves no other purpose.

Hacked (correctly), it will serve the function of a voltage monitor, but obviously not tell you about electrolyte level.

 

...except if the electrolyte is so low that the cells can't generate the proper amount of voltage. Ask me how I know this.

Oh, and for those that are interested, Chacal sells a pre-made kit for this very purpose.

 

Yeah, SQL, I know it is supposed to monitor voltage, it just never really did on mine. I should have said "better voltage sensor" not "becomes a voltage sensor" because with mine anyway, it never tripped from low voltage unless it was really really low, but now if it's just on the verge of not enough juice it will come on. But it YMMV depending on how people wire it, or the condition of their harness, etc.

 

Mike, were you using a battery that came stock with a hole for the probe, or did you have to add one? I'm thinking that, if the probe were in a cell further towards the positive terminal than stock, it might read more voltage than it expects for a "good" reading, and thereby not show low voltage until it was really low.

 

Funny thing is that my bike's been intermittently complaining about the battery. I figured it was low electrolyte because it would complain when the bike was cold-ish, even when the battery was well charged after a decent amount of riding, but it wouldn't complain when the bike was leaning on the side stand. Sure enough, finally got around to pulling the battery this evening and the electrolyte was indeed low in the sensor's cell. The funny thing, though, is that four of the cells were low (one very low), and two were fine. I'm pretty sure I filled the battery evenly originally, so I'm guessing there is some difference in the rates at which electrolyte can evaporate across the cells.

 

Yeah, I honestly don't know if my battery was original or modified, I got a new one pretty soon after I got the bike. I didn't want to have to deal with finding the special battery or modding one, that's why the bypass.

The thing is, although it worked fine before, just last night I had to get a jump to get home, wouldn't surprise me that my battery was wearing out, it is a couple or three years old, but this time I *didn't* get a warning light, so...beats me, maybe something is draining it while it sits, like my new headlight relay job, that I'm still working the kinks out of.

 

There's gotta be. Whenever I check my battery, the levels are down differently. I always fill them up as exactly evenly as possible; but when they do get low, they're never the same.