Preventing Voltage Spike from Alternator Cut-Off

[expozen]

I sort of get why disconnecting the battery from a running alternator will damage the diodes. Because a voltage spike is generated on the discharge side of the diodes that will damage them. Much like a solenoid valve can generate water hammer in a piping system.

Couldn't that same voltage spike damage other equipment? And if such a voltage spike can damage equipment, how come there doesn't seem to be much concern about disconnecting the battery (load) from other types of running charge sources, like a solar charge controller or shore charger?

If there was a circuit like something below, would the battery isolator prevent such voltage spikes. Or would it too generate a voltage spike when the charge circuit is disconnected by the BMS, also damaging the MOSFET's (and maybe the other charge sources)?

How come a transient voltage suppressor doesn't seem to exist as a solution?

[john61ct]

Just don't remove the load / sink, prevent the spikes.

A cheap lead batt will do the job if you don't already have a hardwired bank.

[Shenandoah52]

Originally Posted by expozen

Couldn't that same voltage spike damage other equipment?

I had a problem years ago where the engine on my sportfisher kept blowing up the ignition coil. It wasn't until I blew the 3rd coil that I figured out what the problem was. It was a voltage spike.

The original helm station had an amp meter gauge but not a volt meter gauge and I never saw the amp meter gauge go critical. After installing a volt gauge I was out running and the engine hiccuped like it did just before the coil would let go and looking at the voltmeter it showed 24 volts going thru my 12 volt system. It was these volt spikes destroying my coils.

What I didn't know was that the main wiring harness connector the factory had installed used a hard rubber for the connector plug, which is a huge no-no, instead of the proper grade of plastic. The bad connections at that plug caused the voltage regulator to sporadically lose the ability to regulate the voltage coming from the alternator.

So yea, voltage spikes can damage other items in the system that depend on a certain voltage to work. In the case of those vintage ignition coils they used a ballast resister to lower the 12 volts to 8 - 8.5 volts for them to function properly. Pushing 24 volts through the ballast resister raised the coil voltage to the point they would overheat, burst, and dump their oil over the intake manifold.

[expozen]

Originally Posted by john61ct

Just don't remove the load / sink, prevent the spikes.

A cheap lead batt will do the job if you don't already have a hardwired bank.

I'm not sure what you mean by this, exactly... Is that not what I have shown in the wiring diagram? There is a lead starter battery connected to the alternator output. Do you mean do something different? And what about voltage spikes from other charge sources? All those go to the lead battery too?

[Viking Sailor]

Sterling Power Alternator Protection Device for 12 volts.


https://www.sterling-power-usa.com/S...iondevice.aspx

[john61ct]

No, then you're all set and the issue is moot, taken care of.

[john61ct]

Sterling has excellent products, but several are solutions to problems rarely encountered in real life.

[Wotname]

Originally Posted by expozen

I sort of get why disconnecting the battery from a running alternator will damage the diodes. Because a voltage spike is generated on the discharge side of the diodes that will damage them. Much like a solenoid valve can generate water hammer in a piping system.

Not quite right.
The spike is caused when the current is suddenly reduced to zero and the magnetic field collapses cutting though the stator coils of the alternator. Remember voltage is created in a conductor when the conductor is moved though a magnetic field OR when a magnetic field moves though a conductor. The faster the speed (ether rising or falling), the higher the voltage. So when the alternator is generating a output current and this is suddenly reduced to zero, the magnetic field around the stator coils collapses and induces a very high voltage in the stator windings. This is what kills the diodes.

Originally Posted by expozen

Couldn't that same voltage spike damage other equipment? And if such a voltage spike can damage equipment, how come there doesn't seem to be much concern about disconnecting the battery (load) from other types of running charge sources, like a solar charge controller or shore charger?

The voltage spike is contained inside the alternator (where the stator is) and can't affect anything downstream of the open circuit. Of course if there are other coils of conductors downstream, a spike is develop in them as the current creases.

[expozen]

Originally Posted by Wotname

Not quite right.
The spike is caused when the current is suddenly reduced to zero and the magnetic field collapses cutting though the stator coils of the alternator. Remember voltage is created in a conductor when the conductor is moved though a magnetic field OR when a magnetic field moves though a conductor. The faster the speed (ether rising or falling), the higher the voltage. So when the alternator is generating a output current and this is suddenly reduced to zero, the magnetic field around the stator coils collapses and induces a very high voltage in the stator windings. This is what kills the diodes.

Oh... that makes more sense. Which is why the voltage spike seems to be specifically an alternator issue and not a general problem with disconnecting from charge sources.

Originally Posted by Wotname

The voltage spike is contained inside the alternator (where the stator is) and can't affect anything downstream of the open circuit. Of course if there are other coils of conductors downstream, a spike is develop in them as the current creases.

So, would the example I provided be a legitimate way of avoiding the voltage spike from an alternator? I imagine that because the current can still go to the lead starter battery, by way of the battery isolator, there wouldn't be a drop in current that would generate a voltage spike.

[smac999]

your diagram should be fine. if the charging bus is shut off, alt current will still go to the start battery. and the alt shouldn't blow up.

the better way is to have the BMS shut off the regulator, instead of the alt output. but then in this case. the alt will stop charging the engine battery as well. which could be bad if you need to motor a ways to get back to a dock to fix the issue

[Wotname]

Originally Posted by expozen

Oh... that makes more sense. Which is why the voltage spike seems to be specifically an alternator issue and not a general problem with disconnecting from charge sources.



So, would the example I provided be a legitimate way of avoiding the voltage spike from an alternator? I imagine that because the current can still go to the lead starter battery, by way of the battery isolator, there wouldn't be a drop in current that would generate a voltage spike.

Time doesn't allow for a long answer right now (maybe later?).

Short answer - I would have to make assumptions of exactly where you will be disconnecting the load of the working alternator. The spike will occur everywhere between the stator coils and the disconnection point.

Note, if the alternator is not providing a output current, then no spike will occur so if the regulator is used to stop the field current, there will be no output current to interrupt and thus no spike.

Likewise if the alternator is disconnected at it's output terminal while it is stationary and then run up, it will be safe as there is no output current to disrupt.

Finally, if the output current is not interrupted - no spike.

[Lake-Effect]

Interesting discussion.


I'm currently playing with a surplus gas engine that I'm going to use to drive an alternator, as a 12v charger/generator. I've been pondering whether i need some sort of battery sense to control the field current, or to act as a start interlock (no ignition til there's a battery connected), or to simply include a smaller 12v battery into the system so it's never without a battery load,

[john61ct]

Originally Posted by expozen

Which is why the voltage spike seems to be specifically an alternator issue and not a general problem with disconnecting from charge sources.

Actually, that is only true in practice, because amps current rates are generally lower with other source types.

And cutting the "upstream" power to the source is usually very straightforward.

But in theory, if you had 200A flowing in your House circuit, and just suddenly isolated the only battery, the resulting load dump would creat surge / spikes that could very well damage sensitive electronic load devices.

Cranking the starter can also be an issue, hence SI "starter isolation" feature on ACRs.

Some people even use a complete separate battery for their electronics to isolate problems.

[expozen]

Originally Posted by smac999

the better way is to have the BMS shut off the regulator, instead of the alt output.

You mean wire it like shown below?

[expozen]

Originally Posted by john61ct

But in theory, if you had 200A flowing in your House circuit, and just suddenly isolated the only battery, the resulting load dump would creat surge / spikes that could very well damage sensitive electronic load devices.

So, it seems like there is two different phenomenon at work here:

There's the potential voltage spike generated by the collapse of the magnetic field caused by the ceasing of current flowing through the alternator.

And there's a voltage spike generated by just ceasing of current in general, much like a water hammer effect.

Originally Posted by john61ct

Actually, that is only true in practice, because amps current rates are generally lower with other source types.

I'm curious to know if the shore charger and solar were on the same circuit, and they were disconnected while outputting their rated current, would they be damaged from the theoretical voltage spike? Or would that be something that dependent on the implementation of the chargers? Or am I just overthinking it?

Originally Posted by john61ct

And cutting the "upstream" power to the source is usually very straightforward.

So you're saying that the most appropriate way to wire it would be how I've shown it below?