Honda CB750K5 Restoration by Royhall

[royhall]

Am I testing through the meter yellow to earth or yellow to yellow.

Sent from my Pixel 7 Pro using Tapatalk

[K2-K6]

Each yellow to earth should give you the ac voltage on it's way to the rectifier.

NOT RUNNING with all yellows disconnected it should give you resistance from one to another of not more than 0.2 Ohm.

[royhall]

Okay thanks. I know what I'm doing now. Will have a go after work on Wednesday. Cheers.

Sent from my Pixel 7 Pro using Tapatalk

[Bryanj]

Wrong, its yellow to yellow that gives you the ac voltage, disconnected 50 to 60 vac

[royhall]

Are you sure about that Bryan, I'm definitely not an expert but that doesn't sound right. Can you elaborate a bit more.

Sent from my Pixel 7 Pro using Tapatalk

[Bryanj]

The yellows have no connection to ground, its a 3 phase alternater with the 3 phases being each pair of yellow wires
A- B
B-C
A-C
 And that is the ac voltage INTO the rectifier.
If you want to check DC OUTPUT of rectifier its red(or red/white) to green BUT  not with battery disconected or the volts can go high enough to fry the diodes in rectifier

[K2-K6]

That must be doing phase-to-phase then Bryan ? The peak reversal from alternating swop. Would that mean it's output would be something like 17 volts checked against earth ?

I think as Bryan has pointed out before, the assembly windings are usually very reliable, with any check to see that the resistance is correct and that the output is at parity for each of the three yellow, then the problem will ordinarily be located somewhere else.

Often the integrity of those first three inline connections going to the multi pin can be suspected, and need verification before moving down the line to see what arrives at reg/rec and if it comes out of there.

The manual doesn't suggest a check in voltage of three phase (just that resistance check) output, just an ammeter on DC line out of reg/rec.

[royhall]

So if I disconnect the three yellows at the rec/reg and check the ohms are below 0.2 from that point it should check the entire circuit up to that point without having to strip the bike to get at it. If it fails that test then I will have to strip it back to find the problem, if not then I can check voltage from each yellow to earth looking for around 17 volts. Is that correct, will have a go this afternoon.

[royhall]

This is Honda Bulletin on 3phase alternator on SOHC Fours
https://www.dropbox.com/s/2oqrx5cvv7byqao/3-phase%20Honda%20Charging.pdf?dl=0

Thanks for that Ash. I had a read through and I think it may have helped, not sure.

Just to clarify something. The static voltage on the new AGM battery is at 13.1 volts. If this is classed as high voltage as in the document above, would that mean that the battery wont charge until the battery voltage drops as in scenario 3? That clearly shows no output to the battery. But don't forget the system I am on is a modern combined Rec/Reg so may be a different scenario.

[BigAl (Alan)]

The alternator wiring (3 yellow wires) are not connected to the frame/earth
To test - low voltage continuity, any combination should read 0.4 ohms (yellow to yellow x 3)
If you can carry out a high voltage test (100-250 volt) - any yellow to frame/earth it would confirm that the 3 phase windings are OK
Due to age a number of Japanese classics alternators are starting to fail... Image Kawasaki alternator windings (S1'S2s KH250's)

[royhall]

The problem I have with the above is that none of my 3 meters will read accurately at that setting. I'm getting 0.8 with just touching the probes together. But that said the reading across the yellows at the rec/reg is 1.1. So 1.1 minus the 0,8 is 0.3 across all three. So if that logic is acceptable the windings and cables are okay?

Sent from my Pixel 7 Pro using Tapatalk

[taysidedragon]

The problem I have with the above is that none of my 3 meters will read accurately at that setting. I'm getting 0.8 with just touching the probes together. But that said the reading across the yellows at the rec/reg is 1.1. So 1.1 minus the 0,8 is 0.3 across all three. So if that logic is acceptable the windings and cables are okay?

Sent from my Pixel 7 Pro using Tapatalk

You have to zero the meter for resistance readings. There should be an adjuster on the side of the meter, mine is a thumb wheel.

[AshimotoK0]

The problem I have with the above is that none of my 3 meters will read accurately at that setting. I'm getting 0.8 with just touching the probes together. But that said the reading across the yellows at the rec/reg is 1.1. So 1.1 minus the 0,8 is 0.3 across all three. So if that logic is acceptable the windings and cables are okay?

Sent from my Pixel 7 Pro using Tapatalk

If you are getting the same reading on all three, then chances are they are OK. Like BigAl says though check for any resistance between the steel stator laminations and any of the three yellow wires. Any resistance at all is bad news of failure or impending failure of the winding lacquer insulation

[Oddjob]

Just read an interesting article about testing charging coils like these and his conclusion was that it was a crap shoot, if it tests bad then it's bad but even testing good means it's really only a 50/50 chance of that being true. His assertion was that the only true way to test these type of circuits was to test them dynamically, as in whilst they are operating. Due to deterioration of the lacquer the windings are coated with and the adverse effects of heat and in some cases acidic oil these coatings start to break down when in use due to expansion and loading etc.

I can't say I'm any sort of expert on the charging circuit etc but it makes sense to me and appears to be borne out by published books on the subject like David Sullivans book "Fundamental Electrical Troubleshooting"

[royhall]

With my best meter (I can find no way to zero it, problem for another day) it reads 1.1 at the chain case couplings, but as I say, the leads are 0.8 themselves. All three read the same. Bit hard to get a stator reading as the bike is fully assembled. Have checked continuity from the chain case to the plug that connects to the rec/reg and all good. The resistance reads the same from the plug as the chain case. Due to limited accuracy of the meter I will have to assume then that they are correct.

With the engine running, the voltage yellow to yellow should be about 50 volts AC, yellow to earth should be about 17 volts AC. Are these correct values. (Schools in  )

Where do I test for rec/reg output, is it the red wire in the same multiplug as the yellows at the rec/reg. When tested it was showing identical voltage to the static battery voltage. I assume then that nothing is coming out of the rec/reg. Still got to test that something is going into the rec/reg.

Heads fried now.