Ex-500.com - The home of the Kawasaki EX500 / Ninja 500R banner

under charging

5K views 31 replies 7 participants last post by  DannoXYZ 
#1 ·
hi guy's. since the gen 1 rebuild I seem to have a problem with the above. nothing major but thought I would run it passed you guy's in case it has been mentioned before.

just completed the bikes 2nd rebuild in my possession. back ground. 1990 gen 1 GPZ500s 74k on the clock it was getting a bit tired so needed a refurb. it is completely stock except for a Mosfet regulator. KN air filter. and bigger headlight bulb [60/55w H4] and Motad 2 into 1 exhaust, and voltage monitor light. I also run 2nd gen rear plastics. although that is just for cosmetic reasons.

now the problem. the bike runs really well with no issues. but on a run at about 5miles out the monitor turns from solid green to flashing amber. another 5miles on it starts to misfire and power goes down. if I then turn off the lights, [GPZ's have a light switch] within 400 yards the light returns to solid green and the bike runs normally again. it will continue to stay like this while ever the lights are off [last time for 150 miles] put the lights back on 5 miles down the road it does it again.

it seems as though the battery runs down while the lights are on. even though there is no obvious reason.

the battery is new [month old] shows never less than 12.75volts [12.25v while cranking] the Mosfet regulator shows a constant charge rate of 13.75v. the alternator wires read 60v ac across all 3 wires. it should be spot on. but isn't. all the connectors are clean [it's just been rebuilt]. any ideas guys.

the only thing I could think of was it's somehow under charging, but only with the lights on. is this possible. I did read somewhere the spare charging capacity of these bikes is around 100W could the bigger bulb with all the other [normal bulbs] be using more than the charge rate so draining the battery. it stops when there off.

any ideas. your bikes over there in he US are on all the time the engine is running but seem to work.
my next step would logically be to change all the bulbs for LEDS but is this required on a stock bike.

thanks in advance. Yorkie.
 
See less See more
#2 ·
Confident your install good, all your numbers are good...dunno. Curious. Unless a dodgy connection somewhere compromised by heat of the lighting circuit being enabled for a time? Perhaps repeat all electrical testing literally on the road while hot? Easy to bring a DVM along on a ride. :wink2:

Trade ya' for a Triumph 675? :grin2:

(I'm sure you've read my ongoing exploits with "the Pile".. a buds '10 Triumph 675..mercilessly hacked wiring by some PO's..think I've got the final charging fault now pinpointed and solution at hand) Sorry for veering off course.
 
#3 ·
Could be a heat related issue like ducatiman mentioned. A voltmeter mounted on the tank for testing purposes (or a full time mounted version like us VFR owners need :) )might tell you what the charging voltage looks like out on the road. If you were to stop after the 5 miles and the 10 miles, what are the charging numbers looking like, compared to the numbers you quoted above?
 
#4 ·
thanks guys. 100% confidence in the installation, 0:) every connector on the bike was cleaned with electrical cleaner and micro files on assembly [I never do half a job] good points on running voltage although I would imagine it's difficult to keep the meter probes in place while riding.
I have just ordered a digital remote gauge so should get some answers from that. also ordered a full set of LED's for the bike. just in case it is a draw voltage issue if it's a heat issue they will make no difference. of course it is possible the Mosfet is on the way out, unlikely but possible, I will get some numbers and get back when the stuff comes.
been out on it this morning 50 miles [no lights] no issues.
 
#5 ·
You can do a diode test on the MOSFET unit, but the real proof is steady charging voltage with the engine running at any revs.

Lights appear to be a part of the cause. Do you have the old 55w bulb to swap with and test again?

Alligator clips can keep the test leads in place for the Heath-Robinson voltmeter setup. :)
 
#6 ·
hi, no I don't have the old bulb it was blown and the reason I changed it. the only ones I had spare were the xenon car type. I quite liked the increased brightness. at just over the stock wattage. although the bike was fine before the refurb so maybe. will know better when the new stuff comes. will use the meter first with the present set up then change the bulbs to see any difference.

as for checking the Mosfet diodes I wouldn't know how to start and even if I did the values are unknown. as a rule the units are very reliable and either work perfectly or not at all. unlike the shunt type that can have a wide degree of unreliableness from over charging to completely gone.
 
#8 ·
hi, thanks for the vid. I did the checks as described but could not get anything like the same readings.
test 1. 427. 428 427. 426.
test 2. 0. 0. 0. 0.
test3. 1. 1.1. 1.1. 1.
test 4. 438. 438. 437. 438.
now across the tests all the readings are similar so I assume the RR is good. indeed it does throw out a constant charge of 13.54v on the bike. I also assume [through watching other vids] the values of test 2 and test 3 could be different because of the regulator embedded diodes that deal with charge reduction and my meter not being sensitive enough for exact measurements.

anyhow I got the volts digital gauge fitted and did a short [10 mile test ride] first 4 miles no issues as the lights were off. turned on the lights. charge rate constant 13.54v then started to decrease slowly down to 12.25v at which point the bike started to mis fire and the [volts light] went to amber. [as it did before] turned off the lights. and almost instantly the volts rose to 13v light went solid green. it took a further 2 miles before the volts were back to 13.54v.

I believe the issue is in the lights either a short of some kind or the circuit is pulling too much juice out of the battery and not being filled by the RR. not yet got all the LED bulbs so when I do will repeat the run. if it improves it's down to the bulbs if not I will have to delve into the light circuits. more to follow.
 
#9 · (Edited)
Battery is ONLY used to start bike. ALL power consumed after that should be from alternator only. That's because in dual-power parallel circuits, higher voltage one supplies power and lower one receives. So after starting, power sources may look like;

11.9v = battery
13.6v = RR

This means that power will ONLY flow INTO battery (higher voltage dictates flow direction). Battery voltage can never rise above RR. Most it will do is reach same voltage as RR and current stops flowing into battery. Only time power flows out of battery is if somehow RR drops its voltage below battery at later time after charging battery. In which case, it's more likely RR problem because short would manifest itself from very beginning.

Might me good idea to measure amperage draw of light circuit. Higher drain than calculated could be sign of short.

EDIT: wait, non-U.S. bikes can turn off lights. Try lights ON from very beginning. Does voltage start dropping immediately after turning on bike? Where do you have leads for meter connected?

Which RR model are you using? Does it have separate sense/reference line? Where is this connected?
 
#10 ·
hi thanks. to answer I will start at the beginning. the RR I have is a genuine Shindengen FH0-16a 6 terminal the power wires plus and negative are wired direct to the battery the 3 stator wires go direct to the plug just out of the timing case. no sense line present.

the Gammatronix battery light is connected also to the battery. but the digital volt meter is a temporary attachment so wired into the bike electrics via a separate fuse box fed from the ignition aux terminal of the fuse box. [photo below]. ignore all the wires they were there during the installation of the front and rear camera system, no longer on the bike.

yes UK models have a light switch and yes the power starts to reduce as soon as the lights are turned on.

I see the principal of the charge circuit. but understand that the alternator power output is not infinite but limited to the capacity it can feed [around 100w] via the RR, so surely once the power drain has reached this limit it will then stop charging and steal power from the battery. as the charge rate cannot be maintained above this level. to illustrate the point I once connected a 300w 12v work light to the bike while doing repairs to the garage lights. it ran for 15mins before the battery drained, despite the bike being running on a fast tick over. the battery leads also got a bit warm. so was probably not a good idea.

it has run no problem for years other than fitting the Mosfet when the shunt RR overcharged the battery and boiled it dry. it is only since the rebuild and fitting the big headlight this issue occurred.
 

Attachments

#11 · (Edited)
#12 · (Edited)
#13 ·
Is 100w really limit? Is that limit of RR or coils?/[/url]
Dunno to be honest. it's figure that's been used going back years on the forum, rule of thumb 110w [the CDI uses around 10w] so that leaves 100w spare. if you calculate the draw requirements of the bike, bulbs components ect, it comes out at around 85w so it makes sense. enough power to supply the stock system. there is no reason kawi would fit an alternator and VR capable of supplying 400w when it's needs only 85w.

I do know some members have had power issues when fitting aux equipment without reducing the bikes stock bulbs using LED's I have more tests to do before finding the answer.
 
#15 ·
I reckon these digital gauges could be a good mod for the average bike. I already know far more about what's going on in a instant readout. had the LED charge light on for a few years and been happy with it but the range of volts when it changes colour is a bit wide and it doesn't show whether the power is too high or too low just it is out of a certain range. when I have sorted this issue out. one of these will be going on both bikes permanently. not had chance to test much been raining here 3days solid. next week looks better.
 
#17 ·
Engineering always allows for some overhead, typically double or more. For example, front axle will handle double maximum possible loads encountered in most conditions. From searches here, I found that factory alternator & RR are good for 20a, or 270-watts.

Simple enough to measure actual current consumed by light circuit. Insert amp-meter in-line or use shunt and measure voltage drop. Most clamp-on amp-meters don’t have precision at less time than 50a DC loads.
 
#18 ·
ok, so I finally got to test the bike today. this is what I found.

1st test. as it was normal bulbs. but with digital gauge connected.
starting voltage. 12.75.
cranking voltage. 12.2
once started voltage rises to 12.8 then sort of clicks instantly to 13.1. out on the road [few revs] voltage 13.4. stayed like that for 4 miles. so turned on the light [all] almost instantaneously the volts dropped to 11.5v. turned them off volts click back to 13.1v then 13.4 so turned on the side lights only [no headlight] voltage dropped to 12.8v but stayed there. lights off back to 13.4.

2nd test. after changing every bulb on the bike to LED's.
starting voltage 12.75 [no change]
cranking voltage 12.2v [no change]
once started volts rise to 13.1v [some improvement] out on the road 13.4v [no change]
put on full lights. volts dropped to 12.75. [starting voltage] but dropped no lower than that for the next 5miles.
turned off the lights voltage rises to 13.4. stayed that way for the rest of the ride [20 miles]

conclusion I don't have one. it never charges more than 13.4 [sounds about right] with the LED's it still drops to a no charge scenario when there on but returns to normal when there off. at 13.4 the bike is deffo charging the battery is good. and there are no shorts on the light systems or the LED's would have not changed anything.

now completely lost the plot. my first instinct was the RR is not switching at the power drain it is supposed to but it must bee putting some charge in there as the readings are sort of constant. any ideas welcome.
 
#19 ·
Where are you connecting meter probes to measure voltage?

Let's do this test:

1. Key ON, note voltage before starting

2. Start up bike, leave lights off

3. Drive around for 5-10 minutes

4. Note vo!tage reading

5. Turn OFF bike, then key ON

6. Note battery voltage


What is battery voltage right after you turn off bike following 5-10 min ride?
 
#20 · (Edited)
no probes it's a digital gauge. connected to the battery.

1. 12.75v.
2. yes.
3. yes.
4. 13.4v.
5 yes.
6. 12.75v.

did this test twice. final reading 12.82v as soon as it was possible after the ride. same results both times. after the 2nd test I checked the gauge accuracy using a multi meter. got a reading of 12.75v on both the gauge and the meter. but of course this was after 3 or 4 mins as I had to take the seat off and fit the meter probes. note went to bike this morning [after being stood all night] reading 12.75v using multi meter. gauge reads same voltage.

edit. sometime today I will swap the RR's over from the other bike this one does not have the same issue it will ride all day with the lights on no problem. the RR is the same type but a different model.
 
#21 ·
Where does wires for your meter connect?

Final result of 12,75v leads me to suspect battery somehow. It should be in +13v range at end of ride if RR is sending out 13,4v.

Would be very easy if we can determine how many amps is flowing in headlight circuit when you turn it on.
 
#22 · (Edited)
thanks mate. will check everything out tomorrow inc lights draw in amps. but I think I know [suspect] I have a sort of answer.

the supposed genuine RR isn't. it's a fake/copy and NOT throwing out the correct sustainable charge rate. no idea how I got caught out. brain fart suppose. I took the RR off the other bike although it's a different model it looked different somehow it has all the right marks but has no casting dots on the top of the unit.

fitted the other RR to the bike. charge rate 13.85v with every light [I could turn on] charge rate 13.85v didn't ride the bike yet [still stripped down]
but it ran at fast tick over [1800] for 20 mins and still had no drop in charge rate. when the bike stopped I tried the battery voltage 13.25v.

I would guess although the other RR is showing a flat charge rate it doesn't increase as the current draw increases. and the battery suffers when the lights are off there is very little current draw so the fixed rate charges the battery just a bit.
 

Attachments

#24 · (Edited)
the supposed genuine RR isn't. it's a fake/copy and NOT throwing out the correct sustainable charge rate. no idea how I got caught out. brain fart suppose. I took the RR off the other bike although it's a different model it looked different somehow it has all the right marks but has no casting dots on the top of the unit.

fitted the other RR to the bike. charge rate 13.85v with every light [I could turn on] charge rate 13.85v didn't ride the bike yet [still stripped down] but it ran at fast tick over [1800] for 20 mins and still had no drop in charge rate. when the bike stopped I tried the battery voltage 13.25v.
Good job tracking that down!

Might not be counterfeit RR. I have seen genuine RRs fail from jump-starting from running autos or wiring-shorts. That's path I was taking with measuring current since short in lighting-circuit may cause total-draw to be higher than 20a limit of RR. In this case, limit turned out to be much, much lower than that, probably ~5a.
 
#23 ·
If I'm seeing correctly (at my age its possible I'm not) the RR in your pic is a real deal, genuine Shindengen FH010BA from an older model ZX10 Kaw. +10 year old examples still RR'ing away on both my VFR and older Ducati. As we've noted time and again, they are dead on reliable.

A VFR forum guy suffered total failure (no output) using one of the Far East copies...about 20 minutes into his 1st ride with it.
Beware the Ebay sales jargon "replacement for......" That is code for a non-original copy.

I had a charge system testing rig setup for the Triumph 675 I recently completed...while checking both RR and stator output
I took the opportunity to swap out and test all 4 FH010 in my possession on the same bike, no variation in "test subjects".
Though each is 2004-2008 vintage...each was consistent...near 14.1/14.2V output. These things are ultra dependable long life.
 
#25 ·
hi guy's thanks for your help, advice and comments. it is not often something has me stumped. but your input put me on the trail of the issue. like Sherlock Homes "when all other possibility's have been eliminated what ever is left must be true" I never for one minute suspected the RR wasn't right. I just assumed it was something else. goes to show you cannot depend on anything without checking.

your last two comments have made me feel much better about being taken in. because the RR IS a fake. albeit a very good one.
it is well made. good castings. has all the right marks and logos. even the weight is correct but side by side with a genuine Shindengen like this one.
https://www.ebay.co.uk/itm/Genine-S...303868&hash=item4d98e7364f:g:yX0AAOSwwg5ZyQuU

you can spot the small differences that prove it's not right. it doesn't have the 8 casting dots the original has and the base doesn't have the alloy cover plate it is just plain pitch. all genuine Shindengen's have these two identifying parts.

it took me in as it wasn't a cheap copy replacement but advertised as NOS [with no original packaging] you live and learn.
 
#26 ·
hi guy's thanks for your help, advice and comments. it is not often something has me stumped. but your input put me on the trail of the issue. like Sherlock Homes "when all other possibility's have been eliminated what ever is left must be true" I never for one minute suspected the RR wasn't right. I just assumed it was something else. goes to show you cannot depend on anything without checking.
That's because you forgot a couple of the essential words in ol' Sherlock's quote. I think it went more like: "when all the other possibilities have been eliminated, whatever is left, 'HOWEVER UNLIKELY' must be true". Occam's razor probably applies in this case as well....

Aside from the satisfaction of solving a knotty puzzle and actually fixing things, the thing I find interesting about diagnostics is managing the reasoning process to make sure I'm not making unwarranted assumptions ("it can't be the RR because everybody knows they are dead reliable"), or going down a useless rabbit hole due to relying on incorrect information. The annoying part is that you had a known good RR sitting right there in your garage that you could have substituted for a quick check early on. But good on you for figuring it out in the end.

So now we know that Shindengen RRs are totally reliable except every once in a while they are not. Just bad luck that the RR picked the occasion of a major refurbishing to go bad and that's probably what led you on such a merry chase because so many other things on the bike had been recently disturbed. This was a fun thread.
 
#27 ·
yeah it was a good fact based story and like all good stories had a happy ending. the bike is now fine again running as it should with the addition of new LED bulbs. did 60 miles on it today charge rate never dropped below 13.75v with all the lights on.

of course there are lessons here we can all learn from. never assume anything that you can't test first. and don't believe what your told about a product, unless you can get a guarantee what your told is true.

ironically and like a good jokes punchline. the Shindengen RR that is now working on the bike. IS the RR that was on it before it was stripped down. it had been working faultlessly for 2 years. BUT I borrowed it for the other bike while it was stripped. because the shunt one on that bike melted the RR connector when it shorted out.

rather than wait for a replacement took the one off the gen 1 to keep it on the road. the FAKE one I got was to avoid taking it back off that bike.
but had to do it anyway. now looking for a reasonably priced genuine one so I have both bikes working.
 
#28 ·
ironically and like a good jokes punchline. the Shindengen RR that is now working on the bike. IS the RR that was on it before it was stripped down. it had been working faultlessly for 2 years. BUT I borrowed it for the other bike while it was stripped. because the shunt one on that bike melted the RR connector when it shorted out.
Nice twist! Which reminds me of another applicable quote, this time Ronald Reagan referring to (I think) the Russians: "Trust, but verify".
 
#29 ·
hi guy's. I don't think this saga/repair is quite finished just yet. got the gen 1 out for a clean after it's 60 miles semi wet ride yesterday. that done. started it up to drive into the garage. and it's not charging again. WTF is going on.

no matter I'll look at it later. because my next thought. !! was it that fake RR really the reason it wasn't charging. so took 15 mins out to connect it up to the other bike. [basically a much modified gen 2] and try it on that. it's charging at 14.25v. question for self why will it run on a gen 2 but not on a gen 1. answer. it cannot as they are almost if not completely the same.

there must be a further fault on the gen 1, I didn't fix. out with the multi meter. continuity check on the out put wires, fine. alternator wires AC voltage check. 1 and 3 40.55v [at tick over] 2 and 3 nothing 0. 2 and 1 nothing. 0. reversed the probes. same result.

number 2 wire showing no voltage. go down to the alternator plug near the timing case to retest the wires as they exit the generator. hoping the stator isn't out of whack. no it's fine. 1 and 3. 40 .5v. 2 and 3. 41.1v 1 and 2 41.0v stator fine. re-connected the plug. then retest the RR input wires again all showing voltage and the bikes charging at 13.75v as it was yesterday.

so I have a fault in the no 2 wire somewhere between the alternator plug and the RR plug. not going to be easy to find without stripping the harness out. will do that tomorrow.
 
#31 ·
think. or hoping I may have this problem fixed now. pulled plugs to test for continuity all seemed fine. so tugged at the plugs while disconnected one wire came out of the plug not attached to the spade. fitted new spade end and retested seems fine now. all working as it should.

as a addon the fake RR seems to be working on the other bike for now so leaving it in place to see how it fairs. but will be ordering another genuine Mosfet ASAP, just now there doesn't seem to be any here at a fair price. either new or used.
 
#32 · (Edited)
think. or hoping I may have this problem fixed now. pulled plugs to test for continuity all seemed fine. so tugged at the plugs while disconnected one wire came out of the plug not attached to the spade. fitted new spade end and retested seems fine now. all working as it should.
Bee's knees! :grin2:

That's why connections are crimped, soldered and heat-shrink wrapped in pro-motorsports, aerospace and military applications for performance, reliability and durability. Even on small jumper, I'll do that just to save tonnes of troubleshooting time in future.

 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top