Boxerworks.com

Thanks Barry, That looks clear - thank you for the photo it really helped - I've now checked this out as best I can....and I believe the wiring is correct - the 3 connectors at DF are on a single plug unit so hard for them to be attached incorrectly. The spade connector at the DF brush holder is fine - the voltage at the battery does not surge to 18 volts when it is disconnected....However I notice the battery is holding at 15 volts - it seems to have taken quite a charge when the voltage was so high. (New battery some months ago)
I wonder if this problem now is somehow related to the high charge in the battery ?

D+ wire regulator to diode board looks OK but I think I will have to remove the board to see it properly

You might have what is called a "surface charge" and it would bleed off very quickly with even a small load. I hope that you are putting the volt meter leads directly on the battery posts and not on the terminals that connect to the posts. Yes, they can be at a different potential.

It is essentially impossible for the battery to be part of the problem causing too high of a reading. However, it is always smart to suspect a new battery as faulty. I have seen it many dozens of times.

Do you get the same reading if you remove the negative probe and place it on a chassis ground?

The original unit has a couple of terminals that have nothing to do with the regulation. They are just put there for because it was easy. What did you do with those?

Hoof wrote:.... The spade connector at the DF brush holder is fine - the voltage at the battery does not surge to 18 volts when it is disconnected....

OK so that was a good result. We made a prediction that stopping the current flow through the rotor would stop the output surging and it did. I know that should have been blindingly obvious but with a problem like this you have to make very simple predictions and test them out. The problem has to succumb in the end to logical testing as long as you don't make assumptions however obvious they seem.

If you a certain the wiring is correct having disconnected and continuity tested every single wire and checked that the 3 pin plug is correctly wired then it points again to VR

However unlikely it may seem you can't make an assumption the new VR isn't faulty. Is there anyway you could borrow a known good regulator ? Another option which might not be so easy would be to bench test the VR. It's perfectly possible if you have a variable voltage power supply or know someone who has one.

barryh wrote:
However unlikely it may seem you can't make an assumption the new VR isn't faulty. Is there anyway you could borrow a known good regulator ? Another option which might not be so easy would be to bench test the VR. It's perfectly possible if you have a variable voltage power supply or know someone who has one.

But, it would have to be an AC power supply, correct? Most variable power supplies I run into are DC (even though they are actually rectified AC), I think you would need a single phase to test the regulator.

Major Softie wrote:
But, it would have to be an AC power supply, correct? Most variable power supplies I run into are DC (even though they are actually rectified AC), I think you would need a single phase to test the regulator.

If it's just the regulator on it's own that is being tested you can use a DC supply. To do a bench test connect D+ to the +ve terminal of a variable DC voltage bench supply and connect D- to the -ve terminal. Connect a 21 watt bulb between D- and DF
I know my wiring looks messy but most of that is just improvising with odd bits of wire with spade connectors. It's actually fairly simple. Slowly raise the voltage and monitor across D+ and D- with a digital meter. The bulb will glow brighter and brighter then go out at the switching voltage. Think of the 21watt bulb as the rotor winding it should light up (rotor passing current) until the switching voltage is reached and then go out (current ceases to pass through the rotor). If the bulb never goes out as you pass above 14 volts then the VR is faulty.
You might need to do the test a few times but don't take too long as in the case of my mechanical regulator there are some resistors at the bottom of the can which get hot.

Thanks everyone - I'll work through the tests to-day. Might take some time! My Power Supply only goes to 13.8 volts so I may need to put some extra voltage in series.....Got the new solid state regulator on at the moment - hard to believe that two of them could be faulty..but I'll check them both.

I've been thinking if anything I've done recently could be responsible - and apart from a new battery I have improved the Earth link on the voltmeter so nothing I can think of can have affected the charging.

When making repairs sometimes a new problem is created. Now you mentioned that the wiring harness to the light was replaced. Could there be a chance that something was moved or now touching something else when that harness was replaced?

A CONFESSION.................
As per the regulator test off the bike I made the check
My power supply is not variable so I checked both the regulators with a 12 and 6 volt supply (This was a 12 volt battery and a 6 volt battery in series) - they both illuminate the 21w bulb using a single battery and they both cut the current at the higher voltage - very quickly on the new solid state regulator and less so but audible on the old mechanical regulator.
SO - regulators are fine....wiring fine, connectors fine NOTHING CAN BE WRONG BUT THE CHARGING VOLTAGE IS STILL TOO HIGH!!

Now for the confession!
However I did notice that my spare battery seemed to be in a very fine state of charge ie 12 volt batteries reading nearly 16 volts! so I checked every other 12 volt battery and some dry batteries......all high by about 25%.
I moved on to a very accurate power supply giving 13.8 volts ....my meter reads 17.43 volts an error of 26.3%

Looks like my Sealey multimeter has developed a voltmeter error of 26% ....it was OK for the last decadebut will get the 'float test'

So apologies all round and thanks for all the help - my charging system is now entirely checked out and good for another 40 years or so!
I should have realized that few batteries (even new ones like mine) at rest show 15+ volts - Just thought there had been some 'overcharging' due to regulator failure....such ignorance ....Now to buy a new multimeter!

THANKS EVERYONE

Wow. I guess one good thing about my multiple 4 buck HF meters is that I can check them against each other if I need to. I'd never heard of one going bad like that.

Me either, but at least it's good news: the bike is fixed!