Yanmar 3GM30 alternator wiring....

[lateral]

For anyone who is considering this route, replacing with clone etc. ADR0439 alternator
from DB electrical (Ebay) 98mm non-OEM alt pulley giving 1.5:1 ratio.

Next morning with a fresh coffee, it charged correctly.
Not certain why.

With batteries at ~70% soc. 12.48V SG 1.250+
The internally regulated clone, initial current = 65A 14.7V
Settled to 20A/14.5v over a period of 30minutes and stayed there irrespective of rpm above 1200.
About twice the what the 55A hitachi was producing.

Obviously this alternator has quite low temp compensation as well.
Moderate temperature after 1hr; I could "leave my hand on."


SG after one hr had not changed at all??

DB electrical were technically like tits on a bull. Otherwise they were good.
They knew they were selling an alternator, thats about all.

Next step will be disconnect the field earth leg and attach a ZM4 ext. reg, with temp alarm, and probably additional cooling.

As an aside, on the cheap ammeters/ bm's thread, with the new alternator the reading on Amps was really stable; does this suggest a smoother dc from
more modern rectification?

[lateral]

Update on ADR0439 alternator from DB electrical on Ebay.

I bought this alt after researching and finding only good reports from users, albeit only two.
I had reservations owing to the fan on the original delco 10si only being rated to a max of
74A & the fact it was made in China.

From the above; First test run was satisfactory----> good, to my expectation.
Yesterday, fired her up (at 12.54V SOC) and meter read 82A @14.7V./1500rpm
Increased to 2200 RPM, 106A at 17.2V
Decreased to 1500rpm, 80A at 14.7v. Shut down immediately.

Brand new 3 wire alt with, proprietry yanmar fittings, 4gau wire to house & earth.

It beats me why I think I can beat murphies law.

[becrux]

Overvoltage is a regulator problem.
Was this with the internal or did you install the ZM4?

[lateral]

No, just with the OEM internal regulator.
I will get a ZM4 this coming week and try it with that. Also fit a temperature alarm.
The alternator got hot in the 2-3mins it was going. (~60-70°C) so don't hold much hope of it operating at over 60A.

Pretty dumb of me really, expecting double the output from the same case/fan.
The pain of something cheap not performing to expectation seems way less than buying the real McCoy at premium $'s and then dueling with them.

[lateral]

Originally Posted by becrux

I used an Hitachi knockoff and switched it to P field by straightening the -brush tab and moving it from the F pole of the IR to ground pole. Removed the triple pack diodes. +brush is still connected to Lamp, so I used that for field. No soldering, no additional wires, can be easily put back.
Balmar reg provides lamp output for alarm conditions.
Tach output still works, not needed for Yanmar.

Works great, but temperature sensing for batts and alt is essential.

What performance are you getting from yours at ~70% SOC? Initially?
What does it settle to in absorbtion phase.
Are there definite drops in current when the temp compensation kicks in?
How long does it take to get back to a decent charging current?

Sorry, sounds like the inquisition, but I am toying with sending the unit back and buying one that is designed for purpose.

[becrux]

I can't answer your inquisition because I don't have an ammeter installed.
What I can tell you:
1. I never used the internal regulator. Before running the alt I switched it to P field control and ran the field from the Balmar reg.
2. The Balmar reg runs the alt at 14.6V for about an hour, but it might take 20 minutes to get there. Then it drops to 14.2V.
3. Temp alarm never sounds, probably because I dialed back the field to about 70%, meaning the max amperage is around 90 amps, but I can't tell for sure because I don't have an ammeter.

The alt cranks out as demanded by the reg. I have no complaints.

Sounds like the internal reg is toast, but it could also be the sense wire. Did you see overvoltage at the sense wire (alternator end)? If not, you have a bad connection from sense to Batt. If there ~was~ overvoltage at the alt end of the sense wire, then the internal reg is toast.

On my Yanmar 3QM30FG the wiring harness had some bad connections. I cleaned up the big plug (port side) between the switch and the alt/oil pressure/temp sensor and things got better.

Some folks power the reg or the alt from the oil pressure switch. This prevents extra loading of the engine when there is no oil pressure.

Check these things out before sending back a working alt.

[lateral]

Great info, thanks!
It is not worth sending back as its too expensive.
I will convert to ext reg. (Or toss it)

You did mention earlier that you had no ammeter, I just forgot, apologies.

As you have a good grasp of the workings of an alternator; Something that
has puzzled me.
If the sense wire is getting its voltage form the battery and the battery is in parallel with the alt ouput, say 14.7v, then the sense wire is essentially measuring the output of the alt, not the actual battery?
Cheers....again. I owe you one.

[becrux]

The idea of the sense wire going direct to the battery is to bypass all possible voltage drops between the alt and the batt - such as a battery isolator or combiner. The battery needs to 'see' 14.6 (or whatever your setpoint is) . If the alt needs to put out 15.2 volts to make this happen, that's OK, but 17 seems high. There might be a problem. If the path from the batt to the sense terminal goes through your ignition switch, you have 20 or 30 feet of voltage drop and a couple plugs and connections which could easily explain why the alt is cranking out 17+ volts in order for the sense wire to see 14.6. Also, if there are other things that run off the ignition wire, they could drop the sense voltage as well. The sense wire path has to be short and direct, and dedicated. This is true for both internal and external tegulation.
When you switch to external, you'll want to defeat the internal reg. However, i did not see any temp compensation on the ZM4, but the internal one has it. I'd think about that before installing the ZM4. You could seriously overheat that alt.

[lateral]

Originally Posted by becrux

The idea of the sense wire going direct to the battery is to bypass all possible voltage drops between the alt and the batt - such as a battery isolator or combiner. The battery needs to 'see' 14.6 (or whatever your setpoint is) . If the alt needs to put out 15.2 volts to make this happen, that's OK, but 17 seems high. There might be a problem. If the path from the batt to the sense terminal goes through your ignition switch, you have 20 or 30 feet of voltage drop and a couple plugs and connections which could easily explain why the alt is cranking out 17+ volts in order for the sense wire to see 14.6. Also, if there are other things that run off the ignition wire, they could drop the sense voltage as well. The sense wire path has to be short and direct, and dedicated. This is true for both internal and external regulation.
When you switch to external, you'll want to defeat the internal reg. However, i did not see any temp compensation on the ZM4, but the internal one has it. I'd think about that before installing the ZM4. You could seriously overheat that alt.

So, the sense wire is not measuring the battery voltage (uncharged) but is measuring the battery voltage on charge? ie gross figure.
And, should not exceed Vset, right? (Vset being max voltage regulator will allow)
Then, as you said, my reg must be toast.

My measurement of alt output is dedicated across battery bank not at alternator to earth. Amps is via a shunt to common earth bus bar. (All brand new)

Yup, I hear you on the heat. I am interested in the Mark Grasser "Smart-charge-one". It has temp cutback & temp cutoff.
And is less $ than the ZM4.
Either that, or the ZM4 and an after market temp alarm and cut back the field voltage/current manually.
Great to be able to bounce this off someone. My wife glazes over on the 2nd sentence.

[becrux]

Grasser has a good reputation. I like that the temp sensor is included.
Be sure to read the section on whether to run the two regs in parallel or not.
Good luck!

[lateral]

Originally Posted by becrux

Grasser has a good reputation. I like that the temp sensor is included.
Be sure to read the section on whether to run the two regs in parallel or not.
Good luck!

Yep, hear ya. Specially since mine has an intermttent fault.
I fired her up last night, and at idle read 15.2v / 40A and after about 30sec dropped to 14.82 / 20A stable at all rpms. (As opposed to last test which with rpm increase went to 17.2V / 106A.)
Replaced it with the Hitachi and got 14.2 / 28A with its usual drop to 10A over five minutes.

It seems likely the fault lies in the regulation components. Perhaps the full field tab is too close/loose & full fielding it at times.
What I did notice is that the Hitachi fan produces a lot more air than the clone (10si)
I will look at replacing the clones fan with a 12si fan that is more in the configuration of Hitachi. (With integral front plate to channel air more efficiently)
I'm loathe to swap fans as the hitachi has a crap pressed metal integrated pulley & it needs to be available, complete, as quick backup.)

It will be interesting to see what numbers the 10si produces with external regulation. Supposedly it has heavier stator/rotor windings appropriate to xtra output.

BTW, I've checked the output with a fluke dmm, and my budget ($25) panel power meter ain't that bad. The only downside is fitting the shunt.

[ggray]

Regarding Kevlar belts, make sure all pulleys are large enough for Kevlar.

I tried a Kevlar belt made by Gates, I think, and it became very loose after 20-30 minutes, to the point it was ruined. Tried a second thinking I just hadn't tightened enough, but same results.

Talking w/ the mfg. I learned that there is a minimum diameter (maybe 3.5 inches?) below which the belt will self destruct (my word). There is so little stretch in the Kevlar on the outer face of the belt, that all distortion of the belt in making the turn around the sheave is compression on the inner portion of the belt. Ordinary belts stretch some on the outside so are more forgiving.

Good argument for a flat serpentine belt!

[lateral]

Originally Posted by ggray

Regarding Kevlar belts, make sure all pulleys are large enough for Kevlar.

I tried a Kevlar belt made by Gates, I think, and it became very loose after 20-30 minutes, to the point it was ruined. Tried a second thinking I just hadn't tightened enough, but same results.

Talking w/ the mfg. I learned that there is a minimum diameter (maybe 3.5 inches?) below which the belt will self destruct (my word). There is so little stretch in the Kevlar on the outer face of the belt, that all distortion of the belt in making the turn around the sheave is compression on the inner portion of the belt. Ordinary belts stretch some on the outside so are more forgiving.

Good argument for a flat serpentine belt!

And, kevlar is not good in compression!

[lateral]

Ok, for any one contemplating the 10si clone for xtra output on a 3gm30, I am still in the throws of setting it up. A "Smart-charge-one" is on its way from Maine owing to a suspect regulator on the 10si. ( & for better regulation.)
I have re-drawn the circuit diagram on Marks site to what I think applies to a Delco 10si as opposed his alternators.
Attached- critique welcome.

I will report back once I have a result.

[becrux]

The adr0439 is not a 10si clone. It is an Hitachi clone.
10si has positive field control
Hitachi has negative field control

Your diagram shows the alternator remaining in negative field control. The Grasser reg can do this, but you have to switch a fuse inside it.

Or you could switch it to + field control like I did. This requires removing the triple diode pack and grounding the negative side. If you do this the lamp wire will not work as a lamp wire, but can be used as the field connection. I did this because:
1- My Balmar reg has a feed for the lamp
2- It was easy
3- No cutting or soldering wires
4- completely reversible

I didn't check whether the Grasser reg has a lamp indicator. If it does, you can do this as I did. If not, you can still make it a + field alt by cutting both rotor feeds, grounding one, and soldering a wire to the other and leading it out to the reg. (The triple diode pack stays in for the lamp feed). In this scenario it doesn't matter which one you ground, since the rotor coil is now connected only to the reg and to ground. Do what is easier.

It also doesn't matter whether you leave it as a -field or switch it to a +field. Do what is easier. If you want to do it my way, I have photos.
[Edit: Looks like the Grasser reg has no lamp output. Your diagram looks like the way to go. Don't forget to switch the fuse. ]


Ggray, thanks for the notes on kevlar. I'll keep mine for emergency spares.