Topic: Charging Problem

[scottly]

What you need to do now is disconnect the three yellow wires from stator, usually at plug on electric panel BUT make sure white and green are connected, and measure AC voltage on each of the 3 pairs with engine running at about 3000 rpm

WRONG!!! If the field is not magnetized, there can be NO AC produced by the stator!

[scottly]

• Voltage “at engine”: 13,28
• White to +ve terminal: open

What do you mean by Voltage "at engine"?
What do you mean by White to +ve:open?
Did you try unplugging the white wire from the regulator and jumping it to battery + and checking for magnetism? The key doesn't even need to be on for this test.

[The Lone Builder]

Having completed all the above tests, I started up the motor and at 2500 rpm the battery voltage was 13,5 rising to 14,5 at 300rpm.

Is it possible that heat is the problem and something is breaking down when the engine gets hot? I have thought that the engine gets very hot recently - spit sizzles on the cases and even an hour or two after stopping the engine is still too hot to touch.

I thought to take it for a ride and see how it performs when I return. What would be the most likely thing to fail, and how would I test it?

[scottly]

The regulator is the most likely suspect.

[The Lone Builder]

• Voltage “at engine”: 13,28
• White to +ve terminal: open

What do you mean by Voltage "at engine"? Voltage between battery +ve and engine casing/frame etc.
What do you mean by White to +ve:open? There is no connection between the two - infinite resistance
Did you try unplugging the white wire from the regulator and jumping it to battery + and checking for magnetism? The key doesn't even need to be on for this test. Yes, and same result; the crash bar was magnetised nut not the alternator cover.

[The Lone Builder]

The regulator is the most likely suspect.

K. Not field coil or stator?

[scottly]

Only the heads of the screws that hold the field coil inside the alternator cover get magnetized, not the aluminum cover itself.

[The Lone Builder]

What you need to do now is disconnect the three yellow wires from stator, usually at plug on electric panel BUT make sure white and green are connected, and measure AC voltage on each of the 3 pairs with engine running at about 3000 rpm

WRONG!!! If the field is not magnetized, there can be NO AC produced by the stator!

Correct me if I misunderstand this, but leaving green and white connected will energise the field coil and so there will still be output. But should I be measuring voltage or current?

[The Lone Builder]

Only the heads of the screws that hold the field coil inside the alternator cover get magnetized, not the aluminum cover itself.

K. I’ll check that shortly, but I don’t think they were. Only the section of the crash bar directly opposite the cover is magnetised; the rest not!

[scottly]

The regulator is the most likely suspect.

K. Not field coil or stator?

When you measured 4.6 ohms from the white wire (disconnected from the reg) to battery -, it checks the field coil and it's associated wiring. If the problem re-occurs, check this first; if still good, that would point back to the reg, if no magnetism at the field.

[scottly]

Only the heads of the screws that hold the field coil inside the alternator cover get magnetized, not the aluminum cover itself.

K. I’ll check that shortly, but I don’t think they were. Only the section of the crash bar directly opposite the cover is magnetised; the rest not!

Once again, if the field coil is not magnetized, there can be NO output from the alternator.

[HondaMan]

Lone:
Does your voltmeter have a 10 Amp setting for measuring current? If it does, I might suggest this test:
1. Set the meter to 10 AMPs mode.
2. Disconnect the WHT wire from the regulator and hook the meter (+) lead to the terminal where the WHT wire was.
3. Hook the WHT wire to the other meter wire.
4. Turn on the bike's key: this hookup will show how much current is going to the field coil (engine need not be running). In the 'normal' CB750 field coil this is:
    (12 volts / 7.2 ohms) = 1.6 amps (or so).
Since you have the field coil from the 750A engine, I think they are 6.0 ohms instead, so the amperage will be a closer to 2 amps.

This test will show if the (cold) regulator is putting out current to the field coil to make it excite the alternator.
If possible, repeating this test with everything warmed up could be revealing: if the lower-ohms field coil is overheating the transistor that feeds it, and that transistor is the PNP type, it could be slowly turning off the power to the field coil when hot. That's why I asked about the circuit design.

[bryanj]

The yellows need to be disconnected from the loom and you will have 3 pairs of 2(a=b, a-c, b-c) and each one should be about 60 VOLTS ACas i remember it.
Yes you need the field connected to do this but not the yellow .
This will tell you if all 3 of the windings is working correctly.

I have been troubleshooting these for nearly 50 years and in all that time i have seen 1 rectifier with a broken wire, 1 regulator with a "sticky" point contact and 1 physicaly damaged stator winding, both the last two were on two of my own bikes.
Every other problem was bad connections or broken wires

[scottly]

2. Disconnect the WHT wire from the regulator and hook the meter (+) lead to the terminal where the WHT wire was.
3. Hook the WHT wire to the other meter wire.
4. Turn on the bike's key: this hookup will show how much current is going to the field coil (engine need not be running).

Mark, he measured the field current: 0.2 amp.

[The Lone Builder]

I just measured that current with the engine warm/hot.

With the key off, I was getting 0,007A. With the key on the meter registered an open line!

On the run earlier, using head and tail lights.

Starting voltage 12,8 rising to 14+ at 3000

At a stop the voltage was 12,93 rising to 13,4 at 3000 and up to 6000 at 4000
 
At the end it was 12,57 rising to 12,71 at 2000 and 12,9 at 3500

So do I now need to trace the white wire?

[scottly]

Don't bother trying to measure the current. What voltage do you measure at the reg White terminal with the key on, engine not running, when hot? Did you try the magnetism test cold, then hot?

[bryanj]

If you have good voltage at start the drops off when warm it is either electronics failing when hot or a connection failing when hot.
First thing is try a normal regulator

[scottly]

If you have good voltage at start the drops off when warm it is either electronics failing when hot or a connection failing when hot.
First thing is try a normal regulator

I agree, especially the part about electronics (the transistor in the reg), which is more likely than a connector in this case. Like Hondaman mentioned, the lower resistance of the auto field coil may have taxed the transistor in your "hybrid" regulator.

[grcamna2]

When 'Lone' determines which component is the problem,could he use a regulator from a CB750A ?

[The Lone Builder]

A thought.

The bike is running very hot, i think. Possible reasons for this are:

• Too lean a mixture
• Ignition too advanced

Any others?
When I rebuilt the carbs, I moved the needle from slot 3 to the 2nd slot, which would make the engine run leaner and hotter. Now the plugs are very clean; not burning but very clean.

So if I undid this change would the result be lower temps, and hence maybe elimination?

I will check the advance also, and maybe adjust the air screw. Someone please confirm that to make the mix richer, I turn the screw in, so from its current (1 turn out) setting to ⅞ turn out.

I’ve made contact with the regulator chap and am waiting on a response.

[bryanj]

Original Honda reg from an A model is same as the rest of the fours, its the generator coils that are altered to give a higher output as revs on the A were expected to be lower than the manual bikes

[HondaMan]

A thought.

The bike is running very hot, i think. Possible reasons for this are:

• Too lean a mixture
• Ignition too advanced

Any others?
When I rebuilt the carbs, I moved the needle from slot 3 to the 2nd slot, which would make the engine run leaner and hotter. Now the plugs are very clean; not burning but very clean.

So if I undid this change would the result be lower temps, and hence maybe elimination?

I will check the advance also, and maybe adjust the air screw. Someone please confirm that to make the mix richer, I turn the screw in, so from its current (1 turn out) setting to ⅞ turn out.

I’ve made contact with the regulator chap and am waiting on a response.

The 2nd notch is pretty lean. The OEM notch used on the 750 made after 1971 was the 4th notch because the increased restriction of the HM341 pipes stalled the airflow slightly in the 3500+ RPM ranges. Before that they used the middle notch with the HM300 more-open pipes. Each notch is a little bit over 1% change. You can also 'adjust' the mixture somewhat by changing fuel: going from Regular to Midgrade is also about a 1% richer change, for example.

(Here comes the fight...) Turning the airscrew inward leans out the mixture below 2000 RPM. Normal setting with brand-new carbs is 1 turn out: after years of seating and reseating those screws this becomes more like 7/8 turn out for the same mixture because the soft, tapered screw seat gets widened by the force of the screw's wedged tip. The full operating range of (a new carb's) screw is 1/2 to 1-1/2 turns, so you can surmise that today, this is more like 3/8 to 1-3/8 turns. Beyond the 1-1/2 turns there is no further enrichening because it is then limited by the fuel feed rate of the #40 pilot jet, regardless of the air entry, and that hole size doesn't change.

About the spark advance: if it is WAY advanced it can increase the heating, but it will also knock (ping) if more than 5 degrees further-than-stock advance is used. This is due more to cam timing than almost anything else. You can test against this with fuel octane adjustments in run-time: if switching to premium gas makes the engine run cooler, then the advance is too far. The pipes will then burn hotter, and you will quickly see some discoloring starting at the bends of the headers - the pipes will start to turn 'golden' right there. This is due to the slower-burning fuel not being used up yet as it exits the engine, so it still burns on its way to the muffler(s) above 4000 RPM.

[rotortiller]

I have been on the planet for 69 years and  I have yet to witness a charging problem corrected through carb jetting or ignition timing. If it was and you did manage to mask a heat sensitive alternator issue, it will be a temporary fix. Correcting a lean fuel system is still good though. If it was a no charge case then correcting a heat sensitive circuit is wise too.

[The Lone Builder]

 
An update.

After buying a new multimeter, the current through the white wire at the regulator is 2,5A with a voltage of 12,6. Bearing in mind the resistance of the coil, 2,9-3,0A would be “right”. It didn’t make sense that the current was 0,04 and the battery was still getting some charge.

I refitted a voltage monitor (single LED that changes colour with voltage) and had a short ride yesterday. This said the battery is being overcharged at anything over 2500 rpm. I checked this with my new voltmeter and this is indeed the case. If I keep revs to between 2500 and 3000 with the headlight on, the voltage remains in the 14V area, which should be OK.

So as several of you suggested, it looks like the regulator is the problem. I have bought one on eBay that I should get sometime this weekend, so fingers crossed.

The only thing now is, when I ride, If that little light flashes red, (>15,2V) I get all worried that I’ going to fry my battery; beforehand, no light, no worry! 

[The Lone Builder]

I have been on the planet for 69 years and  I have yet to witness a charging problem corrected through carb jetting or ignition timing. If it was and you did manage to mask a heat sensitive alternator issue, it will be a temporary fix. Correcting a lean fuel system is still good though. If it was a no charge case then correcting a heat sensitive circuit is wise too.

I borrowed a timing light from AutoZone, and the timing on ⅔ was quite advanced. I fixed that and it responds somewhat better now. Will see how the temperature feels later.