Alternator diodes

[womble1964]

There seems to be a myth amongst boaters that disconnecting the output of an alternator while it is producing charge will "blow" the rectification diodes.

I can see no reason for this to happen.

Could someone please enlighten me.

[joseph nicholas]

I've heard this about disconecting the battery in both cars and boats, but it has no merit, of course you don't want to short these two connectionsl

[MK14]

https://en.wikipedia.org/wiki/Load_dump

http://automotive.electronicspecifier.com/design/having-trouble-with-your-car-s-load-dump

[BradC]

There seems to be a myth amongst boaters that disconnecting the output of an alternator while it is producing charge will "blow" the rectification diodes.

I can see no reason for this to happen.

Could someone please enlighten me.

It's not common, but it's worse than the load dump referenced elsewhere as when you disconnect the alternator output there is *nowhere* for the power stored in the coils of the alternator to go, so like an ignition coil, the voltage builds until it finds somewhere, usually an insulation breakdown in the winding or rectifier. At least with a load dump there is a stack of electronics *outside* the alternator to fry which ultimately keeps the peak voltages below breakdown levels.

[james_s]

I've had an alternator fail when it was accidentally run without a load. I don't know how common it is but it can happen. Without a load the voltage can go very high even if the field is not driven hard. It's possible to exceed the maximum reverse voltage of the diodes and they fail.

[rstofer]

Back 50+ years ago, we had the boat alternator diode problem several times.  The diodes just couldn't hold up to the voltage created when there was no battery in the circuit.

But that was a long time ago!  I don't know if it is still a problem and have no intention of checking it on my truck!

[james_s]

I suspect it's much less likely since external electromechanical regulators went away, those were pretty much on or off, and without a battery the voltage would go WAY up.

You can actually run 120V universal motors as used in power tools from an externally regulated alternator in an emergency. Connect the field directly to 12V so it runs balls out and give it a bit of RPM and it can do over 100V. I've heard of more than one person doing this.

[Brumby]

There seems to be a myth amongst boaters that disconnecting the output of an alternator while it is producing charge will "blow" the rectification diodes.

I can see no reason for this to happen.

This is a trap for young players - and such traps can be found in many places.

The key is to understand that components of an interconnected system do not always behave in the way we might expect when they are operating in a different configuration to those with which we are familiar.  The automotive environment is a classic - specifically when the configuration difference is whether the battery is in circuit or not.

The automotive battery is a very low impedance component - which means it takes a LOT of energy to make big voltage changes across it quickly.  Since sources of electrical energy within the automotive environment are not particularly huge, the result is that changes in voltage across the battery are comparatively slow.

This is the scenario we are very familiar with.

However, when you remove the battery from the circuit, the circuit impedance increases dramatically, which means any energy sources or drains will have a much greater influence on circuit voltages.

Now you might ask ... "Well, the alternator has a regulator - so shouldn't that stop excessive voltages from reaching the rest of the vehicle's circuitry?"  The answer to this is both "Yes" and "No".

In the case where you have an engine running at a given speed with an alternator connected to circuitry without a battery, then it is possible that this can run and keep within tolerable limits.  But this is a "steady state" scenario - and pays no heed as to how this state was achieved.  This, again, is a familiar line of thinking which makes it seem strange that there would be any significant risk.

What is not included in this thinking is what happens at the moment of a configuration change.  It's when things change that we have to pay extra attention.  We encounter this in all areas of electrical engineering - from things like switch bounce to back EMF.

Such is the case with the alternator and the battery.  The alternator's pre-change state was one where it had significant magnetic field energy as part of a current generating system - where that current was being absorbed by the battery.  When the battery is removed, that current generating system has a key parameter change - and it will respond according to the laws of physics, as people have described above.


Always remember to not only consider state 'A' and state 'B' of a system - but also the transition between the two.

[johansen]

its worth doing some experiments on. but i don't think you're going to blow the diodes but rather blow up the regulator.

the inductance of the exterior circuit manifests as voltage across the arc when you break the circuit, but it is the inductance of the alternator that can cause the 600 volts or so needed to appear across the diode block. however, modern alternators usually have a capacitor across the output and i would expect it to be large enough to absorb the energy.

it is not the energy stored in the field coil that causes the alternator to over volt the diodes, because the alternator can't generate the 400 vac needed to blow the diodes, rather it is the current flowing in the stator coils (which have a leakage inductance that is not coupled to the rotor) that causes the voltage at the diode block to rise to nearly infinity when the circuit is broken. but again, most alternators have a capacitor on their output and this should not be a problem.

you can get an alternator to produce 400 volts or so if you spin it at about 40,000 rpm at full field excition. that would correspond to 40 volts at 4000 rpm, which would mean you could recharge the battery at about 1000 rpm. i think this is pretty typical but modern cars run the alternator at about twice or three times the engine rpm in order to get enough current out of the alternator at idle.

[ProBang2]

The correct answers are already here:

The posting of "MK14" and additional the posting of "Brumby".   

"its worth doing some experiments on. but i don't think you're going to blow the diodes but rather blow up the regulator."
Nice theory. In reality you blow up at least one of this nine diodes:



The best way for sure diagnose of this failure is an oscilloscope
as it can be seen on this web-page: Alternator (error analysis)

Sure: Manufacturers have tried to make their alternators less sensitive against "running without battery connected/ with no load".  But no one can make a prophecy about the behavior of every alternator in any car.
The only possible recommendation is: Don´t let your alternator (engine) run without a connected battery. It is that simple. Better safe than sorry...

[Zero999]

The correct answers are already here:

The posting of "MK14" and additional the posting of "Brumby".   

"its worth doing some experiments on. but i don't think you're going to blow the diodes but rather blow up the regulator."
Nice theory. In reality you blow up at least one of this nine diodes:



The best way for sure diagnose of this failure is an oscilloscope
as it can be seen on this web-page: Alternator (error analysis)

Sure: Manufacturers have tried to make their alternators less sensitive against "running without battery connected/ with no load".  But no one can make a prophecy about the behavior of every alternator in any car.
The only possible recommendation is: Don´t let your alternator (engine) run without a connected battery. It is that simple. Better safe than sorry...

That looks like quite an old schematic. Nowadays alternator regulators use PWM to switch the field, rather than linear control. Here are some datasheets for ICs commonly used for this:

http://www.st.com/content/ccc/resource/technical/document/datasheet/group3/df/67/29/ef/26/7e/4f/02/DM00320826/files/DM00320826.pdf/jcr:content/translations/en.DM00320826.pdf
https://www.onsemi.com/pub/Collateral/CS3361-D.PDF
http://www.nxp.com/assets/documents/data/en/data-sheets/TC80310.pdf

[SeanB]

A lot od alternator diodes are deliberately made with a low avalanche voltage, around 40-60V, so as to limit the open circuit voltage it can produce if it has an open circuit. This often though results in one or more going short circuit from the power they suddenly have to dissipate, and this then means the output of the alternator will be short circuited, and then you will be grateful that the manufacturers put in a fusible link in the power lead. Thus a blown fusible lead means it is likely the diode block is shorted, and you are noty going to be sure which failed first.

Easy to replace them, provided you have a press to get the old ones out, and a method to impact the locked screws holding the parts together, and you can solder or weld in the new pack.