okay guys! i managed to get the bike home and i have now charged the battery. No problem starting the bike now but i presume i shouldn't close the book on this quite yet?

Tomorrow i'm going to get a multimeter measure the voltage at the following states:

When bike is off (what is a good value?)
When rev is 5000-6000 rpm (Good is 14-15V)
At idle (What is a good value?)

is there anything else i should measure to eliminate the most obviously possibilities?

Ignition off, anything between about 12.6v and 12.8v is good

At tick over I'd expect to see somewhere in the region of 13.5 and 14.2v

I think it's time someone wrote a sticky for this one, I'll do the words and pictures if someone like Neo will confirm I'm not talking crap....

Voltage at the battery:

Off: 12.8V or a little more indicates it's fully charged. Less than this means it's not fully charged, and when you get the low side of 12.4V after the bike has been ridden, then you might as well buy a new one because it's either not capable of accepting a decent charge, or it's not capable of holding it.

At idle: 12.5V or better, ideally not that low (I get barely more than that on mine). Any less, and it's likely that the output is little low, or something is draining the battery in a manner that it shouldn't be.

At 5K rpm: IIRC the book says 12.5 to 14.5V. Ideally, you want to be at the higher end of this scale. The lower this figure is, the less charge will be going back to the battery, so if you have had starting issues or are running a lot at low revs, you might end up with the battery having less charge in it than is ideal when you next come to start the bike. If it's quite close to 12.5 then it's a good idea to look at ways to improve it. If it's higher than 14.5V, it can also potentially be an issue depending on how much higher. If you have in excess of around 15.5V it suggests that the reg/rec unit is not dissipating enough energy, and the battery is being overcharged. In time, this will ruin the battery and it won't hold a charge for very long. Also, when they start going wrong they can occasionally toast themselves due to the excess heat generated. If it's too high, replace the reg/rec, but don't use a really cheap one as they're crap. Some people use units from a Yam R6, and there are other units you can use too.

If you aren't getting enough volts at 5K, there can be several causes. It's still possible that the battery has seen better days, however before condemning it, you can test the alternator output. The wiring on these bikes is prone to corroding at the connectors, and this causes excess resistance and reduces the charging output. The connector for the alternator is located on the right side fairing bracket, so at minimum you need to remove the right mid panel and lower fairing. There are two connectors on the bracket, you want the larger clear/white one, not the smaller black one. The connector in question has three yellow wires that come from the stator assembly in the alternator. You need to disconnect this and measure across two at a time. This is done on the wires that lead back across the engine, not the connector on the main loom. It doesn't matter what combination of the wire you measure, you just need to do all of them in pairs, in turn. Neo's usual example is to call the three wires A, B and C. You need to measure and record AB, AC and BC.

Set the meter to ohms first. Each pair of wires should have between 0.1 and 1 ohm resistance between them. No resistance means a dead short, much higher means the wiring is likely to be suspect, and infinite means one had broken. After this, set the meter to AC volts. At idle, you should get about 12-15V across each pair. At 5000rpm, the figure should be approaching 50V. This sounds like a lot, but bear in mind the job of the reg/rec is to convert this to DC and dissipate the excess, which is why it gets hot. On my bike, the wires were badly corroded, to the point where one had come adrift, and while it was gradually coming away, the extra resistance caused created sufficient heat in the connector to plastic weld the two halves together, and make it a right bugger to get apart! If your wires look manky, it is usually advised to cut them back and solder on new wires and new connectors.

If the output of the generator is good, the next port of call is the reg/rec, located on the left of rear subframe, behind the ignition box. The location of the connector for this makes it ideal for collecting road crap and corroding the pins. You can disconnect the connector and measure the three yellow wires again to see if there are any significant drops which could be caused by poor wire quality in the loom. Over time, corrosion can creep down the wires and affect the output, just like at the other end. Again, some years ago I had this on my bike and it ate through one of the output wires, leaving me with a flat battery. The connector in this case was completely destroyed, and I had to replace all the spade ends and connect them individually. You can get a kit with the necessary parts from Rick Oliver or on eBay.

If the reg/rec is getting enough juice from the yellow wires, you may find that the wires between the reg/rec and battery/earth are not performing at their best. If you reconnect the reg/rec, set the meter back to DC and measure the voltages you get at the two output pins on the connector (red and green, IIRC. The two bigger ones....). If there's a big difference between the figure here and at the battery, you issue is most likely to be poor connectivity between the reg/rec and the battery/earth. Check for corrosion on the connectors, and check to make sure the battery terminals and leads are in good health. Some people bypass the wiring loom altogether, and take the output of the reg/rec directly back to the battery using heavy duty cable and a suitable fuse to ensure that any potential 'blockages' are removed. A fuse is essential.

Off: 12.8V or a little more indicates it's fully charged. Less than this means it's not fully charged, and when you get the low side of 12.4V after the bike has been ridden, then you might as well buy a new one because it's either not capable of accepting a decent charge, or it's not capable of holding it.

Battery voltage of 12.4 after a ride does not necessarily indicate a dud battery just a suspect electrical system that needs further investigation.

At idle: 12.5V or better, ideally not that low (I get barely more than that on mine). Any less, and it's likely that the output is little low, or something is draining the battery in a manner that it shouldn't be.

Depends on the bike, NC30 I would be concerned, NC35 not sure until I did more tests as lights ON means a constant heavy load on the electrical system.

At 5K rpm: IIRC the book says 12.5 to 14.5V. Ideally, you want to be at the higher end of this scale. The lower this figure is, the less charge will be going back to the battery, so if you have had starting issues or are running a lot at low revs, you might end up with the battery having less charge in it than is ideal when you next come to start the bike. If it's quite close to 12.5 then it's a good idea to look at ways to improve it. If it's higher than 14.5V, it can also potentially be an issue depending on how much higher. If you have in excess of around 15.5V it suggests that the reg/rec unit is not dissipating enough energy, and the battery is being overcharged. In time, this will ruin the battery and it won't hold a charge for very long. Also, when they start going wrong they can occasionally toast themselves due to the excess heat generated. If it's too high, replace the reg/rec, but don't use a really cheap one as they're crap. Some people use units from a Yam R6, and there are other units you can use too.

12.5 volts at 5000 rpm is a real problem. IMHO the OEM wiring configuration will give low charge voltages a most engine rpm. OEM rect/reg maximum voltage specification is 15 volts so any reading upto this value can be assumed to be ok.

Set the meter to ohms first. Each pair of wires should have between 0.1 and 1 ohm resistance between them. No resistance means a dead short, much higher means the wiring is likely to be suspect, and infinite means one had broken.

More important is a reading of less than 2-3 ohms between each yellow wire pair when measured at the rect/reg, with the rect/reg disconnected, and there should be infinite resistance from any yellow wire to chassis.

NC30 has the rect/reg located on the left side adjacent to the battery.

A large difference between the battery voltage and the rect/reg output voltage at 5000 rpm is not a bad connection issue, simply a function of the OEM wiring configuration, which is why I recommend directly connecting the rect/reg output to the battery.

Thank you Speedy for that excellent walkthrough. You definitely should make that a sticky.

The story of my bike has taken another turn now. A couple of hours after fully charging the battery i went out with on the bike just to see how it felt. When taking off the bike felt very energetic so i felt optimistic. Super easy to start. But later on the exact same scenario played out: After 5 minutes of riding i suddenly get no response at all from the throttle and revs slowly drops. The engine actually stalled before the bike was standing still. I tried starting the bike with no luck (once again the starter motor sounded pretty weak - battery was charged just hours ago). This time it was down hill pretty much all the way home so i just rolled home. Tried starting the bike while rolling downhill with no success.

When i got home i tried starting it again, and now it did start! However, when i let the clutch out the engine stalled. Tried several times and it didn't matter how many revs i fed when slowly releasing the clutch. It stalled every time.

A lot of bizarre and unusual(?) symptoms. I'm hoping that the combination of these problems will make it easier to pin point where the fault it. What do you think? Any ideas?

Tomorrow i'll do the measuring suggested by Speedy

To me that sounds like the battery is not getting charged. Check the wiring to make sure there aren't any corroded or broken wires. Five minutes will flatten the battery - I know this from personal experience.. Before it dies, you'll very quickly notice a lack of power at mid to high revs, I presume because the ignition is demanding more juice than the battery can supply as it's not being topped up by the charging system.

Also - make sure you read through Magg's notes on my posting. He knows better than me! :-)

Enzyemer I suggest you not ride your bike until you have investigated the problem, repeated heavy discharges of the battery will damage it, if it is not already suffering. Get your multimeter and start measuring.

You also should check that the vacuum hose to the fuel tap is connected correctly and also in good condition. Poor vacuum signal equals poor fuel flow from the tap to the carbies.

magg wrote:Battery voltage of 12.4 after a ride does not necessarily indicate a dud battery just a suspect electrical system that needs further investigation.

I'm was sure the Haynes says that 12.4 or less means the battery is dead. Perhaps it actually says fully discharged, which of course is different. :-)

Depends on the bike, NC30 I would be concerned, NC35 not sure until I did more tests as lights ON means a constant heavy load on the electrical system.

I guess it wouldn't be such a bad idea to disconnect the lights and test at idle? I always presumed that the difference was down to the charging circuit being less efficient at lower revs, I seem to recall that on my TZR there was a low and high speed circuit on the charging, and that at very low engine speeds the combined output was several volts less than mid to high speeds where they were more or less constant. With the NCs being three phase I guess there is a difference. I presume the difference between the idle and mid rev range output on the AC side of things is not really relevant as such because most of the volts are lost in the reg/rec? I don't really understand why it's necessary to produce 50V to supply a 12V system, unless the reg/rec is trading volts for current using Ohm's Law?

12.5 volts at 5000 rpm is a real problem. IMHO the OEM wiring configuration will give low charge voltages a most engine rpm. OEM rect/reg maximum voltage specification is 15 volts so any reading upto this value can be assumed to be ok.

I'm not exactly over the moon with 13.5, but it's not causing me issues as far as I know. I would assume 12.5 would be teetering on the edge of discharge?

More important is a reading of less than 2-3 ohms between each yellow wire pair when measured at the rect/reg, with the rect/reg disconnected, and there should be infinite resistance from any yellow wire to chassis.

I might have to investigate that myself. It was some time ago the thing was replaced, so the insulating tape keeping the crap out could probably do with some attention.

NC30 has the rect/reg located on the left side adjacent to the battery.

I thought it looked a little different on one of the other threads. If I do get around to making a proper doc for this I'll make sure to point out the difference!

A large difference between the battery voltage and the rect/reg output voltage at 5000 rpm is not a bad connection issue, simply a function of the OEM wiring configuration, which is why I recommend directly connecting the rect/reg output to the battery.

I agree on the wiring side of things, but surely if the connectors were getting corroded that wouldn't help either? As regards the new wiring, would you take the negative back to the battery as well as the positive, or just to the nearest decent earth if they aren't necessarily the same thing? You (or was it Neo?) suggested 2mm wire or better for the AC side of things. The spades on the DC are bigger. Does that necessarily indicate a higher current and therefore chunkier wire?

Enzymer - I think the lowdown here is that more testing with your multimeter is needed (than just the battery voltage) to really get to the bottom of the issue. The checks are easy, just a process of elimination to determine which component is at fault.

Don't stress, as much as this isn't fun it's a very common problem on these bikes and when it's fixed properly it will be good for the life of the bike.

I'm was sure the Haynes says that 12.4 or less means the battery is dead. Perhaps it actually says fully discharged, which of course is different. :-)

12.4 or less is a discharged battery. Having said that, they can be discharged further but the available energy will be limited. Repeated discharged to less than 10.2 will cause long term damage. 12.6 volts or more is considered a fully charged battery.

I guess it wouldn't be such a bad idea to disconnect the lights and test at idle? I always presumed that the difference was down to the charging circuit being less efficient at lower revs, I seem to recall that on my TZR there was a low and high speed circuit on the charging, and that at very low engine speeds the combined output was several volts less than mid to high speeds where they were more or less constant. With the NCs being three phase I guess there is a difference. I presume the difference between the idle and mid rev range output on the AC side of things is not really relevant as such because most of the volts are lost in the reg/rec? I don't really understand why it's necessary to produce 50V to supply a 12V system, unless the reg/rec is trading volts for current using Ohm's Law?

I would expect an NC35 to have a battery voltage at idle sufficient to supply a small charge to the battery but obviously less than an NC30 that does not run lights ON. The 50 volts AC from the alternator is the open circuit output, you will only get a peak of 15 volts AC when the rect/reg is connected.

I'm not exactly over the moon with 13.5, but it's not causing me issues as far as I know. I would assume 12.5 would be teetering on the edge of discharge?

Any voltage over 13.8 is considered adequate to fully charge a lead acid battery, 12.5 volts is effectively a discharge state.

I agree on the wiring side of things, but surely if the connectors were getting corroded that wouldn't help either? As regards the new wiring, would you take the negative back to the battery as well as the positive, or just to the nearest decent earth if they aren't necessarily the same thing? You (or was it Neo?) suggested 2mm wire or better for the AC side of things. The spades on the DC are bigger. Does that necessarily indicate a higher current and therefore chunkier wire?

Bad connections are always a bad thing but the issue with the OEM wiring with regard to the battery charging system, is that the only chassis ground point for the loom is located as far from that battery as one could get. It needs to be at the battery. This OEM arrangement results in less than optimum charging of the battery. Simplest although not the most efficient fix is to connect the -ve output terminal of the rect/reg directly to the chassis where the battery -ve lead is bolted.