The stud of the rect/reg is not usually connected to the rect/reg -ve output terminal, which is why you need to splice a lead into the existing rect/reg -ve lead and connect it to chassis, as the simplest solution. The method I employed has the same effect, I just preferred taking the output of the rect/reg directly to the battery, no extra connections to go bad.

The issue I have with the OEM configuration is that there is one earth and then there is another earth. The rect/reg output is connected to one earth, the green loom wire, and the battery -ve terminal is connected to another earth, the chassis. These two earths then join at a point at the front of the engine. The result is that the full output of the rect/reg does not appear across the battery because in simple terms the two earth systems are competing for the rect/reg output. The result is a lower battery charging voltage that diminishes with increasing rpm, as was the case with my bike. I have seen similar results with a bike from other manufacturer.

The explanation is complicated and best demonstrated by measuring the battery voltage with the engine running at idle and then at 5000 rpm. Link the rect/reg -ve output terminal to the chassis and repeat the measurements again. Would be interested in your results

I assume the fuse you are referring to is the 30 A one included in the method I employed. Fuses only protect the source, in this case my battery, from a short circuit load, in this case the a possibility of a faulty rect/reg. It limits the current flowing out of the battery, not into it.

The same principal applies with the OEM wiring system. The OEM 30 A main fuse stops a fault in the bike's electrical system, possibly a failed rect/reg, from causing a current larger than 30 amps to flow from the battery.

Providing the wiring from the alternator to the rect/reg is in good condition there is no need to change it.

Have I answered your questions, if not let me know.

Seems a sensible answer to me. I just wondered if any of the potential voltage loss might occur before the juice got to the reg/rec. I guess that considering a healthy one should kick out around 18V anyway and the reg/rec is designed to dissipate anything significantly higher than the necessary charging level, it makes sense that losses after the reg/rec would be more important to negate than ones before.

magg wrote:The stud of the rect/reg is not usually connected to the rect/reg -ve output terminal, which is why you need to splice a lead into the existing rect/reg -ve lead and connect it to chassis, as the simplest solution. The method I employed has the same effect, I just preferred taking the output of the rect/reg directly to the battery, no extra connections to go bad.

The issue I have with the OEM configuration is that there is one earth and then there is another earth. The rect/reg output is connected to one earth, the green loom wire, and the battery -ve terminal is connected to another earth, the chassis. These two earths then join at a point at the front of the engine. The result is that the full output of the rect/reg does not appear across the battery because in simple terms the two earth systems are competing for the rect/reg output. The result is a lower battery charging voltage that diminishes with increasing rpm, as was the case with my bike. I have seen similar results with a bike from other manufacturer.

The explanation is complicated and best demonstrated by measuring the battery voltage with the engine running at idle and then at 5000 rpm. Link the rect/reg -ve output terminal to the chassis and repeat the measurements again. Would be interested in your results

I assume the fuse you are referring to is the 30 A one included in the method I employed. Fuses only protect the source, in this case my battery, from a short circuit load, in this case the a possibility of a faulty rect/reg. It limits the current flowing out of the battery, not into it.

The same principal applies with the OEM wiring system. The OEM 30 A main fuse stops a fault in the bike's electrical system, possibly a failed rect/reg, from causing a current larger than 30 amps to flow from the battery.

Providing the wiring from the alternator to the rect/reg is in good condition there is no need to change it.

Have I answered your questions, if not let me know.

great info there i assumed there was an earth from the actual reg via the chassis, i will too add an earth.

On a side note the fuse is to protect the cable to stop it melting when over current is applied

The OEM 30 A main fuse is to stop a faulty rect/reg from damaging the battery, and by association, the earth wire that runs the length of the loom to the chassis connection at the front of the engine. I have my doubts that the earth loom wire would be saved by the fuse, as 30 amps is a large current to pass down the loom earth wire and this is another problem I see with the OEM configuration.

By linking the rect/reg -ve output terminal directly to the chassis, the loom earth wire is bypassed and the 30 A fuse should better protect the battery in the event of a rect/reg failure.

The only loss of alternator output, prior to the rect/reg, is usually bad connections at the interface between the alternator/loom and again between the loom/rect/reg.

How is my version of War & Peace going, LOL.

A pity honda didn't have someone working for them in the early 90's with this sort of electrical insight magg!

Another chapter of War & Peace.

This wiring configuration would appear to be typical of motorcycle electrical systems. I think it follows from the design brief given for the loom design/manufacturer. The chassis is the earth return for the starter motor with a short heavy link between the chassis and battery. The loom earth wire is the return for the other electrical units, including the rect/reg. There are few, if any, electrical units that use the chassis as the earth return. The only downfall with this design IMO is that the loom earth connection to the chassis is so far away from the battery.

In the tables below are the battery voltages I recorded when the wiring was changed to connect the rect/reg directly to the battery for a bike with 35/60 watt headlights. The battery voltages for a bike with 55/60 watt headlights would be different as indicated in the tables.

Idle:

No lights 14.0 volts
Low beam 13.3 volts ( expect 13.0 volts with 55/60 watt headlight)
High Beam 12.7 volts

RPM to achieve 14.6 volts:

No lights 2000
Low Beam 2500 (expect 3000 with 55/60 watt headlights)
High beam 3500

What size/capacity wire would you recommend for connecting the reg/rec directly to the battery? Would 2mm be sufficient, or would something a little heavier be more suitable?

http://www.cablesandconnectors.com/wiregauge.html

Looking at this table id go for 2.5 -3mm wire personally.

I used the same stuff that goes in earthed household appliance wiring (my dads a builder and his shed is a treasure trove for shit like that) and it works really well.

No idea what thickness it is though.......