What is 'Acceptable' Alt output?

[crazy steve]

I'm trying to get everything checked for the maiden flight of my resurrected SG, and the last hold-up is the alternator. I'm just not seeing the voltage levels the FSM call out...

I pre-checked all the components before using them on this bike; the regulator tested good, but the field and stator coils were a bit out of spec. Both had a bit more resistance than spec'ed (3.8 ohms on the field, and about .55 ohm on the stator coils), but other have reported these amounts and had success with running them. I have several sets, and all check pretty much the same, including a set off a bike with no known charging system issues when it last ran.

I'm getting about 13.3v at 2000 rpm, and it goes up to 13.9v at 4K. This is with everything in 'normal' operation (head, tail, running lights all on). I can't quite seem to get the 14.5v called for, even after going through all the connections again and making sure they're clean/tight. Is this 'good enough', or do I need to look for a better set of coils?

[Rasputin]

I do not know what would be the correct answer to this question, but I know mine. If it is over 14.1 it is getting in the too much category! Every bike, car, truck, and even Concrete mixers have not run over 14.1. If I had 13.5 and up I would be content.

[ManagerMike]

Everything I have ever read, been taught, experienced, and or learned in training, for over a quarter century of professional auto repair has been very consistant regarding what it takes to maintain a 12 volt (nominal) DC electrical system.
A 12 volt, lead-acid battery regardless of size or capacity, requires 13.5v to 15v charging voltage.
If you can get at least 13.5 volts at a cruising rpm fron these notoriously weak charging systems, you are indeed replenishing your battery. Expect less from these beasts below 2500 rpm.
The manual gives some unrealistic values, perhaps to discourage a potential 'do-it-yourselfer' (there aren't any here are there?) but the IDEAL charging voltage for ANY lead-acid "12 volt" (in a perfect world they are actually 12.6 volts charged, static, and unloaded) battery is 14.2.
13.5 volts is acceptable, no less. 15 volts is ok as well, no more.

[Larrym]

State of Battery Charge???

Steve,

You didn't mention the actual battery age/condition.

I had to replace my "old" sealed battery with a new "leaker" last week. (Much thanks to the site here for providing the replacement battery info...)

With lights on and running at 3K Rpm I got 14.4 VDC.

(Got the same voltage readings from the old battery under same running conditions but it wouldn't hold a charge overnight...)

Sorry but I didn't slide the inductive ammeter over the battery cable so I can't give you any ampere reading.

[cywelchjr]

My experience with these charging systems is that if you can't get at least 14.1 volts at 3k, it will NOT keep the battery charged and points to a likely regulator problem and one test for that is to ground the low side of the field coil when running at 3k RPM and if it goes up to 14.5 or better then the regulator is bad, if it doesn't it could be the rectifier or connections.

[crazy steve]

Hmmmmm....

Well,I just checked the coil resistances again, at the regulator connections so this takes into account the ones behind the fuse panel, and got 3.9 ohms and .7 ohm. Now, the coils are well above the 68 degrees quoted in the FSM for checking, probably somewhere around 120 degrees; I can hold my hand on the cover, but if it were much warmer, no. So some of the increase is undoubtably due to that.

The bad part is the more I check it, the worse it's getting... I'm now down to 13.3 at 4K.

The battery? It came with the bike, and the PO said it was 'pretty new' when I got it last fall (and looked new). Now, when I first installed it a few weeks ago (without charging it, just how it came out of the bike when I tore it down), it spun the motor no problem and operated everything I was trying to check at the time. I did run it low during my testing of the various electrical parts of the bike, but it took and held a charge no problem.

I'll try the bypass test and see what I get. I do have a couple more regulators that tested good with a meter, I may have to do a swap....

[crazy steve]

Well, still no joy.... tried the bypass and another regulator to no avail, so it's probably the coils... I'll let everything cool to room temp and check them again, I did find another set that ohms out good, but I wanted to keep a spare set for the two bikes I'm keeping.

Oh well....

[Just In Case]

Do a voltage check at the output of the regulator, using the earth that's right there. Might be better to to do a voltage drop check instead, put one lead on the the reg output and the battery +ve post and with the bike running and lights on, the reading needs to be near zero. Do the same between the reg earth pin and battery -ve post, again a near zero reading is expected. Otherwise you have a poor connection or thin wires.

You can put your meter on AC and check between the white wires too. Most meters won't be accurate due to the AC not being 50/60Hz, but you are looking for similar readings between each phase. Note that some meters can't handle the AC frequency these generators put out so if the readings don't make sense, there is no need for immediate panic. It's more a confirmation test, if the readings match each other, each winding is doing it's thing.

With headlight off, you should get 14.4V at 3-4k. With the headlight on expect 13.5ishV. None of my vehicles get 14.4V with the headlights on. Even the 3095 which (due to having 10 55W lights) has an 120amp alt and with just with the normal headlights on gets 13.6V at full revs, but 14.4V at no electrical load at just off idle.

[neJeff]

Steve, Been battling the same darn charging issues. One day is good, the next not so good. Sometimes the light won't kick on, some days it does (PO put a single throw switch on to "help" it on when it could not help itself). I have read 14.5 volts, and then 13.5 volts and have even watched it meander under 12 volts at idle. If you figure this all out let me know. I'm going to Harbor Freight today when they open and buy one of those $4.99 digital multimeters and am going to hard wire the damn thing to the battery and make a long set of leads and set the unit right in front of me at the handlebars so I can see how the charging system is performing while I am riding. I have a used regulator/rectifier coming from a friend but it won't be here for several more days...damn I hate electrical issues. Nice wiring write up by the way, printed it off and am going to refer to it often

[crazy steve]

Well, after thrashing on it some more, I've come to the conclusion that the FSM is full of BS...

The stator coils all appear to be good. After rechecking them cold, I'm getting .4 ohm readings if I leave the meter connected long enough for it to 'stabilise', although it will flicker to .5 occasionally. My meter only resolves down to .1 ohm anyway (and it's not a cheapy either; it's a Fluke). Same goes for the field coils; readings of between 3.3 and 3.5 ohms cold on all of them. I'm getting the same readings on all the coil sets I have by the way, and two sets are off bikes that were running with no known charging system issues. After I got a good connection for the regulator bypass, I saw 16+ volts for an instant, so they're working.

Regulators.... I have five regulators, all bench tested for good diodes, all passed. Three are off bikes that had no known charging system problems. Swapping in different regulators gave slightly different results, although as engine heat built up, the readings dropped a bit, which would make sense as the coils resistance would go up with heat. I am basically seeing about 12.8-13v at 2K rpm, 13.2 to 13.3 at 3K, and 13.5-13.6 at 4K. This is with the head/tail light on. Pull the fuses on the lights, and the readings jump to 13.3 at 2K, 13.5 at 3K, and 13.9 at 4K, although one regulator made it to 13.99 volts. I have yet to see 14 or over, even beyond 4K rpm.

I also went through and tested every connection to make sure they're good. I got all of them down to below .2 ohm, most at .1. My meter doesn't zero out on the ohms x 1 scale; if it measures anything, it'll show .1. The only other thing I could do would be to remove all the plug connections and hard-wire everything, but I find it hard to believe that this would be that sensitive to that little 'extra' resistance if there's even any, and that's not something I want to do.

Comments or thoughts? I think I'm seeing the best this can do, so I'll give 'er a test drive and see if it strands me with a dead battery...

[Just In Case]

Instead of trying to measure ohms, measure the voltage drop when the bikes running and lights are on. Check from the reg output to the battery post, and do the same for the earth side. This is a real world test and will show up any issues. Imagine a poor connection that has a tiny point of a good connection. This will show up ok on an ohms check, but try to run 10 amps through and there will be a voltage drop. It's like comparing thin and thick gauge wires, both will have low ohms, but the thin wire will have a higher voltage drop for the same current.

This has been the only reliable method I've had whilst tracing voltage loss, especially in starter circuits. Everything ohms out ok, but check while cranking and suddenly there's a volt between the cable and terminal. Clean up the joint and it become a few millivolts. I had the same issue with my XS key; ohmed out ok, but 0.75V loss while running. Ended up wiring the key to a relay due to the key's contacts being too burnt to sort out.

[cywelchjr]

Well, after thrashing on it some more, I've come to the conclusion that the FSM is full of BS...

The stator coils all appear to be good. After rechecking them cold, I'm getting .4 ohm readings if I leave the meter connected long enough for it to 'stabilise', although it will flicker to .5 occasionally. My meter only resolves down to .1 ohm anyway (and it's not a cheapy either; it's a Fluke). Same goes for the field coils; readings of between 3.3 and 3.5 ohms cold on all of them. I'm getting the same readings on all the coil sets I have by the way, and two sets are off bikes that were running with no known charging system issues. After I got a good connection for the regulator bypass, I saw 16+ volts for an instant, so they're working.

Regulators.... I have five regulators, all bench tested for good diodes, all passed. Three are off bikes that had no known charging system problems. Swapping in different regulators gave slightly different results, although as engine heat built up, the readings dropped a bit, which would make sense as the coils resistance would go up with heat. I am basically seeing about 12.8-13v at 2K rpm, 13.2 to 13.3 at 3K, and 13.5-13.6 at 4K. This is with the head/tail light on. Pull the fuses on the lights, and the readings jump to 13.3 at 2K, 13.5 at 3K, and 13.9 at 4K, although one regulator made it to 13.99 volts. I have yet to see 14 or over, even beyond 4K rpm.

I also went through and tested every connection to make sure they're good. I got all of them down to below .2 ohm, most at .1. My meter doesn't zero out on the ohms x 1 scale; if it measures anything, it'll show .1. The only other thing I could do would be to remove all the plug connections and hard-wire everything, but I find it hard to believe that this would be that sensitive to that little 'extra' resistance if there's even any, and that's not something I want to do.

Comments or thoughts? I think I'm seeing the best this can do, so I'll give 'er a test drive and see if it strands me with a dead battery...

I've always been able to get at least 14.2 volts with a good charging system and a good battery, but with a bad battery this charging system will NOT ever get up to full voltage. Both my 400 and 1100 get that, but ONLY with a good battery, if the battery is even marginal it will NOT make full voltage and it won't do it right away, it take about 30 seconds or so to get there.

[fredintoon]

Dammit Steve,
point the headlight at a wall and rev up.
If it gets brighter you are good to go.

[neJeff]

Steve, Last couple days I recleaned connections, grounds ect and it seemed to be charging better and light was coming on everytime. I did my "real world" voltmeter test tonight (direct to battery, live riding) and at idle pulling 12.9 (1100 rpm) and at highway speeds anywhere consistently between 14.5 to 14.9 (in the 3800 to 4000 rpm range). For some reason things seems to be working very well now. As someone else wrote already in this thread, I was was also told my an old timer you need to be in the 14 range to charge this battery effectively. My "new" regulator arrives tomorrow and I guess it will be the backup.

[crazy steve]

Just a follow-up on this.... I finally got it to show the kind of numbers everyone's talking about.... what a PITA. I got 14.6 @ 4k with the head/tail light off, and 14.2 with them on.

The problem was the connections between the regulator output and the battery. Mind you, I did clean all these and several were new (at the fuse panel), but virtually every plug-in connection had just enough resistance so that in total they dropped my output at the battery to below 14v. I went back through and meticulously re-cleaned each one (sometimes several times) and worked each one down to below .1 ohm. This included pulling the female halves out of the plugs and tightening them up for a second time.

Next one, I'm replacing these.....