Runaway Alternators

[poboy]

Here's a weird one. I have catamaran with two Volvo MD7As. The alternators are off the shelf Delco Remy 63 amp, two wire units. They charge a house bank of two GC1s in parallel until target voltage is reached. At that point, two Blue Seas ACRs open and the respective start banks get charged.
A little history - I am currently at a dock with no electricity, so I have to charge the batts with just the alternators. I replaced the house bank about a month ago, topped off the charge, all was good.
This weekend, went down for the weekend. First thing, I start the engines to charge everything. I have a Victron meter to monitor the system. I note that I am getting upwards of 17 volts. I let it go about 30 minutes, but voltage never goes down. Then I smell burning rubber and find the port side has lost a belt. I pulled both alternators and took them to be checked. Port unit bad, starboard good. Reinstalled using new unit on starboard and two new belts.
Starboard is still putting out close to 18 volts at 90 amps, port is still at 17 volts at 80 amps.
Shut down and used a borrowed Honda 2000i and 15 amp charger to top the banks off. Retried the alternators and had the same result. Brought the alternators to my local shop this morning and they are working perfectly. 14.7 volts, about 65 amps.
So... What the heck is causing the internal regulator to leave the door wide open, so to speak? I changed nothing from a month ago regarding the charging system. The 120 volt, 3 stage, Charles Charger works fine. All connections are clean and tight.
I'm at a loss, any thoughts or ideas would be welcome!

Paul

[Zednotzee]

Sounds like a good one!

Probably the issue for me is that when you first changed the house bank, everything worked normally. Voltage regulators can fail wide open, which would give you an output in the 18v range, but they don't go back to a normal output on the test bench afterwards.

The only thing I can really think of at this stage is that your AC battery charger is most likely hooked directly to your batteries, while the alternators feed through the Blue-Sea relays. Can you manually isolate the batteries from one another, or bypass the relays one at a time?

I would say the chances of two alternators having the same fault at the same time is pretty remote- there's gotta be something else going on.

[poboy]

Thanks, Zednotzee. When I get back down there, I plan to hook the the alternators directly to the start batts to eliminate the house bank and the combiners. That should take most of the variables out of the equation.

Anyone else? Ideas?

[Bash]

Originally Posted by poboy

Anyone else? Ideas?

This one's got me stumped.

Come on folks, help the dude.

[westwinds]

The remote voltage sense input is sensing a voltage that is below the voltage that would cut the alternator output back to prevent over voltage. Could be a wiring problem, broken wire? SI-Series Automatic Charging Relay may also be the problem if the relays are not getting a sensing voltage that causes the relay(s) to drop and connect the batteries after the engine stats.

[RicknSue]

I'm going to assume you mean 3 wire delco alternators (the plug in two wires and the Bat+ screw on connection). In the two wires- one is the Field wire (exciter)- starts the charging, and the other is the sensor wire- the internal regulator works off whatever that sensor wire is feeding it (should be connected to a bus that actually reads battery voltage). The alternators are in 'full - field' which is , give me every thing youve got mode, and if I had to guess, reading a very low sensor voltage

[sailorchic34]

Yes, has to be in the sense point as the alternators test find at the shop. Easy thing to do is run a new sense line from each alternator to the batteries (or to the + lug of the alternator which is not ideal but works) Maybe heavy corrosion at the sense point. I'm assuming a common sense point for both alternators as both have the same issue

[poboy]

Thanks guys!! That's the input I needed. I do have an on off switch in the excite circuit (red wire on the two wire plug). Engines are just little two bangers, over 30 years old, don't want the additional load unless needed. The sense wire ( white) disappears into the wire harness, will have to investigate further. What y'all are saying makes good sense ( no punk tended ) regarding he problem, just weird that I've no problems for over three years with the same setup.
And yes, they are three wire units, not two wire!

[westwinds]

Originally Posted by westwinds

The remote voltage sense input is sensing a voltage that is below the voltage that would cut the alternator output back to prevent over voltage. Could be a wiring problem, broken wire? SI-Series Automatic Charging Relay may also be the problem if the relays are not getting a sensing voltage that causes the relay(s) to drop and connect the batteries after the engine stats.

It's not very likely, but could have the sense wires going to the house battery and if the charging relay does not connect (drop) then the engine battery would be over charged as the alternator sense wire is seeing the voltage on the house battery which would be low as it is not being charged.

[hellosailor]

poboy, there's no such thing as a "2 wire" alternator set up. When you refer to a DelcoRemy 63-amp alternator, the results I found show that is a DelcoRemy 10SI-series alternator, and those are all born as "three wire" alternators. Alternators are conventionally called "three wire" when they have a primary output wire and a separate battery voltage sense lead.

(When the sense lead is tied back directly to the output, so it is actually seeing the alternator output and not the battery voltage, that's called a "one wire" setup. That's used sometimes when you are switching battery banks, etc. and can't bother switching the battery sense lead. A kludge job.)

So what happens with any conventional 3-wire alternator, is that if the sense lead is disconnected, the regulator says "Wow, is that battery flat! I'd better go to 100% output power!" and the voltage immediately goes above 17 volts. This is also the big scam at highway repair shops, they will disconnect the sense lead for 1/2 minute and promptly blow the alternator, then show you that you need a new one. Yes, you can blow an alternator in less than a minute this way.

You also blow all the 12v lighting and electronics very quickly.

When you get back to the boat, you want to check out how those alternators were wired, and if there is any break in the sense lead(s), fix that immediately or take that alternator offline before starting the engine(s).

If there is no break in the sense leads, you have a regulator failure. Just as critical to take that alternator offline and not let it run in "full power" unregulated mode.

On a multi-battery system, the sense lead needs to be switched along with the battery switch, either using a switch with more terminals, or some other magic. So one-wire setups, or 3-wire setups where the sense lead has been tied back to the alternator/starter junction, are not uncommon.

[westwinds]

Could re-wire and have sense wire from one alternator going to the engine battery and the other alternator and its sense wire to the house battery. You could perhaps keep the Automatic Charging Relay to tie the batteries together but a manual switch would be better if the engine battery dies and you need to use the house battery for starting. Having two alternators running at the same time makes for an interesting setup, and the redundancy is a nice feature.

[poboy]

Thanks for all of the replies. I have a good handle on what I need to looking for now.
Some comments have been made regarding battery switches, so I need to add some more info on my battery setup. I used two Blue Seas Add A Battery kits to isolate the house and start banks at all times except when charging with the alternators. Each bank has its own Off-On-Combine switch and ACR. Via these switches, I can combine the start batts with each other and combine the house bank with the start bank if needed. No switching is needed for charging, though. The ACR's create a one way gate that let's voltage pass from the house to the start bank when charging, but isolates the banks from each other at all times.
I'm thinking that I need to have the sense wire tied to the house bank (I don't know what it hooked into at present).

[Maine Sail]

Originally Posted by poboy

I'm thinking that I need to have the sense wire tied to the house bank (I don't know what it hooked into at present).

If the alternator feeds the "house bank" directly then the "sense wire" must sense either the B+ terminal on the back of the alternator OR the house bank.

If the sense wire is ANYWHERE other than the B+ terminal it must sense the bank the alternator is connected directly to. If not the regulator will not see voltage increases and will just keep pumping... I have to fix this quite frequently with DIY external regulator installs. Seen more than my fair share of EXPENSIVE batteries cooked due to improper "sensing"....

In other words if your sense wire is sensing a start battery but you've moved the switch so the alt is now feeding the house bank the voltage will simply go through the roof because it won't sense any change in voltage at the start battery.

So, if the alt goes to a bank, any bank, sense there at the bank where the alt feeds. This will give you the least voltage drop at the battery post.

If the alt output can be "directed", via a battery switch, sense at the B+ or the other end of the cable, at the back of the switch, that is directly connected to the alternator B+ terminal.....

[hellosailor]

As Maine said.

Which is arguably why 1-wire systems exist. They can only do a limited amount of long-term damage (improper charging) as opposed to the amount of damage you can do immediately with a sense wire connected/not to the wrong place.

[Maine Sail]

Originally Posted by hellosailor

As Maine said.

Which is arguably why 1-wire systems exist. They can only do a limited amount of long-term damage (improper charging) as opposed to the amount of damage you can do immediately with a sense wire connected/not to the wrong place.

They may not do short term damage but often result in chronically undercharging batteries, which leads to a longer term failure...