Charging voltage too high?

[jackonz]

Thanks Barry - nice to hear from someone in the County of Cheshire - I see Motorworks do an electronic regulator (Voltage regulator electronic standard output) so I'll give them a ring tomorrow. Probably a good update anyway as the old one is original and 40+ years old.

I fitted one of these to my R80/7 and it works perfectly, all plug and play.

[CafeR65]

Thanks for the input gspd, sounds like that would be worth checking while I am in the area. As it stands there is definitely no shorting between the these three wires on my bike but I might have a look at the feedthrough anyway just for my piece of mind.

I have another long shot theory that I am yet to test; since I have one of those newfangled lithium batteries with a BMS (battery management system) onboard, I wonder if the over-voltage cutout is being triggered between 14.5-15 volts which then takes the battery offline and then the regulator runs away because it has no longer has a reference voltage? As soon as I get a chance i'll fit a lead acid battery and go for another test ride to rule out the possibility.

Alternatively does anyone know a simple way I can lower my VR output and hence alternator output by approx 0.5 volts (short of replacing the VR) so that I could rule out my lithium battery overvoltage cutout theory?

[Rob Frankham]

There is a way to test an electronic voltage regulator... but... it requires a variable bench power supply capable of supplying a substantial (say 5 amps at more than 15 volts) current.

Simply connect the regulator and the bulb as shown in the diagram below. Switch on the bench supply with the voltage at - say 8 volts and slowly increase the voltage. The bulb should light and grow brighter until a voltage somewhere between 13.5 and 14.5 volts is reached, the actual critical voltage will depend on the regulator. When the critical voltage is reached, the bulb will go out. Lowering the voltage slightly will bring the bulb back on again. The on/off switching should be complete (i.e. the bulb should go out completely) and be virtually instantaneous (there may be a point where the bulb flickers slightly but that will be due to the sudden switching of the bulb affecting the stability of the supply). The critical voltage (i.e. the supply voltage where the bulb switches off) is the set voltage of the regulator and approximates the maximum voltage the functional charge circuit will attain.

It's a very simple test but it does require the use of an 'infinitely variable' bench power supply.

Regtest.jpg (26.56 KiB) Viewed 710 times

N.B. this test will not work well with a mechanical voltage regulator

Rob

[barryh]

There is a picture of my set up doing the above bench test back earlier in the thread on page 3. Rob's diagram shows how simple it really is provided you have a suitable bench power supply similar to the one shown. I wasn't doing it as a test but to adjust the set point on the original electromechanical regulator which is still working fine at 44 years old.

[gspd]

I have another long shot theory that I am yet to test; since I have one of those newfangled lithium batteries with a BMS (battery management system) onboard, I wonder if the over-voltage cutout is being triggered between 14.5-15 volts which then takes the battery offline and then the regulator runs away because it has no longer has a reference voltage? As soon as I get a chance i'll fit a lead acid battery and go for another test ride to rule out the possibility.

The only thing I know for sure about a BMS is that most bikes run well forever without one.
It's possible (but unlikely) that both the regulators you tried are faulty.
EME sells them for under $20. Might be worth trying a third one.
I've never encountered a situation where something other than a regulator (or its wiring) caused overcharging on any vehicle.
If all else fails disable the BMS and try it with a known good lead acid battery.

my D- terminal measures less than 1 ohm to ground

hmmm...not sure if this is ideal.

Question for Rob F -
I've never seen this and don't think it's even possible, but Is there theoretically an infinitely minuscule possibility that a perfect storm (corrosion, shorting, meltdown, imperfect grounding) at the diode board somehow provide an unlikely path for the black and blue circuits to bleed into each other and cause overcharging?

[Rob Frankham]

my D- terminal measures less than 1 ohm to ground

hmmm...not sure if this is ideal.

How 'less than' is 'less than'? I would like to see it a lot less than one ohm... it should be a direct short circuit... that being said, many DMs start having difficulty reading less than an ohm. Even the slightest corrosion or dirt on the probes or internal connections can affect readings that low. It's hard to say whether that is OK or not. I would be tempted to rig a temporary wire from D- to a sound ground connection on the engine to see if that makes any difference.

Question for Rob F -
I've never seen this and don't think it's even possible, but Is there theoretically an infinitely minuscule possibility that a perfect storm (corrosion, shorting, meltdown, imperfect grounding) at the diode board somehow provide an unlikely path for the black and blue circuits to bleed into each other and cause overcharging?

I'm not going to say 'absoultely not! but I can't think of any way that damage to the diode board could cause the charge circuit to overcharge. After all the 'Black circuit' goes nowhere near the diode board and the only place the two circuits come together is at the regulator so, for one to 'bleed' into the other would imply a regulator that isn't working. I suppose that, if the blue and black wires were to become short circuit in the sub-loom or the connector at the regulator, that might do it...

Rob

[CafeR65]

Sorry for my radio silence over the last few days, I had to do further research and testing before I could be sure that i'd figured this one out.

So it turns out my hunch was correct that my 'smart' lithium battery was part of the problem. I have an SSB Powersport lithium battery on my bike but I believe this issue is going to affect ANY lithium battery that has an integrated BMS (battery management system).

I have done a lot of reading on lithium BMS's over the last few days and whilst I can't find anything specific for my SSB branded battery I did find information on various other quality brand lithium batteries that specifically states there is an 'over voltage' or 'over charge' protection system built into the BMS that will disconnect the battery cells from the charging circuit if any one cell exceeds 14.6-14.8 volts (specific voltage depends on the make and model of the battery). I don't think it's a coincidence that this is the exact voltage that my charging system steadily reaches before it suddenly spikes up to 17volts.

So my theory was that my lithium battery over-voltage protection system was tripping is likely correct but what it doesn't explain is why this causes the system voltage to spike higher - my best guess is that once the battery self disconnects the lack of load on the charging system causes the alternator voltage to increase. I would have thought that the VR would continue to keep the alternator output voltage under control in this situation but clearly it doesn't.

Anyhow since I couldn't prove why this was happening all I could do was prove that my lithium battery was the cause of the problem so I fitted a lead acid battery in it's place and sure enough the problem went away. Usually the voltage spike would occur after 5-10 minutes of riding but after 30 minutes of riding with a lead acid battery fitted the highest voltage I saw was 14.7-14.8. This is slightly higher than I would expect but I beleive it is because my current VR and prior VR are both 'high output' units even though they are not advertised as such.

So moving forward from here I will likely order an adjustable VR and set it to modulare around the low 14's so that I can continue to run my fancy lithium battery without incident. I will be sure to let you all know how this goes but I am quietly confident this mystery has finally been solved.

Thank you all for your input and please pass on my findings to anybody else who is considering a battery upgrade - lithium batteries and high output voltage regulators don't mix!

[barryh]

Well done for getting to the bottom of it.