Myth of alternator damage cause by BMS disconnect?

[team karst]

Originally Posted by goboatingnow

10000uF would be far too small values in excess of 1F would be needed , batteries are equivalent to 100F

The VR can do nothing to regulate the load dump response. The output spike is not a function of the field current at that point and is not influenced by it.

Yes, there are two transient energy sources. Field and stator. one would need to integrate I*V over time for the waveform you speak of. I suspect it won't take anywhere near a Farad to absorb it within a reasonable voltage. And, most all systems will have other loads besides the battery that are absorbing energy. Someone needs to do the math.

[LeighC]

"Lets go back to a generator and a black box regulator with a voltage and current coil as I recall. No pun intended. Most won't have the slightest" : quote from above.
I have plenty of S/H 20 amp Lucas ones of these old 3 core regs lying around, same as the VW's, could I try one of these to drive my rotor circuit of about 8 amps, I have 2 exposed wires on the positive brush & also 2 the same on the earth brush on my car type Bosch alternator, I want to control belt slip & cooling.
Cheers - Leigh

[Wotname]

Originally Posted by Cadence

Lets go back to a generator and a black box regulator with a voltage and current coil as I recall. No pun intended. Most won't have the slightest. :-)

However IIRC, the 'current coil' was really a 'reverse current coil'. Its purpose was to prevent current flowing from the battery and through the commutator/rotor windings when the generator was stationary. Of course alternators do not have this problem as the stator diode pack prevents 'reverse current flow' when the alternator is stationary.

Today one could use a diode for this function (if using a generator) but when generators were common, diodes weren't.

[team karst]

Originally Posted by Wotname

However IIRC, the 'current coil' was really a 'reverse current coil'. Its purpose was to prevent current flowing from the battery and through the commutator/rotor windings when the generator was stationary. Of course alternators do not have this problem as the stator diode pack prevents 'reverse current flow' when the alternator is stationary.

Today one could use a diode for this function (if using a generator) but when generators were common, diodes weren't.

yeah, and if that failed shorted, the generator operated as a motor. Usually, an unwelcome change. Not quite enuf to propel my VW tho.

[goboatingnow]

Originally Posted by team karst

Yes, there are two transient energy sources. Field and stator. one would need to integrate I*V over time for the waveform you speak of. I suspect it won't take anywhere near a Farad to absorb it within a reasonable voltage. And, most all systems will have other loads besides the battery that are absorbing energy. Someone needs to do the math.

The issue of load dump protection is actually very complex. Firstly you have to exactly what you are protecting. What is the Peak spike you will accept and for how long. On my situations in the past especially in emergency vehicles etc a big high power 2000uf was used but this was backstopped with many avalanche diodes distributed around the wiring loom.

Even then , these systems clamped at around 45V , trying to completely remove the spike is actually difficult which is why using a battery with its effectively very very large capacitor, is used as the primary means of controlling load dump spikes. ( because these occur anytime an alternator is exposed to any rapid current change )

So a capacitor is nothing like a comprehensive solution and only works with making the consumers ( and the alternator electronics ) rigged enough to cope with the attenuated spike.

[Wotname]

Originally Posted by team karst

yeah, and if that failed shorted, the generator operated as a motor. Usually, an unwelcome change. Not quite enuf to propel my VW tho.

Thread Drift Alert

Was that a 6V or 12V VW?

And presumably +ve ground!

[team karst]

Well past page 2. All entertainment at this stage.

12V There were POS ground VW’s?

[Wotname]

Originally Posted by team karst

Well past page 2. All entertainment at this stage.

12V There were POS ground VW’s?

I dunno, that is why I was asking!

There were lots of generator equipped cars that were +ve ground but there also plenty that were -ve ground. My 60 year old English tractor +ve ground and generator equipped.

[team karst]

Originally Posted by goboatingnow

...

Even then , these systems clamped at around 45V , trying to completely remove the spike is actually difficult which is why using a battery with its effectively very very large capacitor, is used as the primary means of controlling load dump spikes. ( because these occur anytime an alternator is exposed to any rapid current change )
....

good info on the field situation. I thought "load dump" is the scenario that resulted from the quick removal of THE battery. So, in your mitigation scenario, one of the proposed solutions is a SECOND battery, that somehow won't be quickly removed??

[goboatingnow]

Originally Posted by team karst

good info on the field situation. I thought "load dump" is the scenario that resulted from the quick removal of THE battery. So, in your mitigation scenario, one of the proposed solutions is a SECOND battery, that somehow won't be quickly removed??

Yes it’s one way to mitigate the situation , however a fully charged relatively small battery may not fully damp a worse case main load dump scenario. It may be that TVS diodes make need to be fitted as well

My view is that lithium disconnects should remove the alternator from the circuit and not leave any electronics stranded on the alternator side. Hence only the alternator then needs protecting

Of course field current disconnect ( or control via the regulator ) is best.

Really in a high cost lithium upgrade , modifying the alternator to support an external regulator ( and even better external Diodes ) makes sense rather then leaving the stock alternator. Extracting the field supply is not difficult.

[team karst]

or, make sure the alternator is charging more than the house bank. I've reported this earlier, but each of my two alternators are charging 3 independent banks. AND, the house bank has multiple LiFeP batteries, therefore multiple, independent BMS.
Then, the most likely fault scenario, and one that has already happened, is the Alt regulator looses its mind, and full on charging to high voltage is occurring. One by one, the BMS shut down its battery, and its only the FLA, in my case, keeping the charge voltage from going truly skyhigh. By now, the nut holding the wheel has heard the rather dim Victron HV alarm, and takes action...

[Dieseldude]

Originally Posted by goboatingnow

There’s already a “ micro processor “ inside your boats alternator.

The LinBus based regulators are a single dedicatedintegrated IC. Not any different to the current non linbus alternators.

Electronics are far more reliable then mechanical things.

Your “ external “ system of course is also a microprocessor.

Loads and loads of microprocessors on boats these days. Mine has 28. !!

You can’t protect alternators with breakers by the way or fuses.

Not sure what equipment you are including in your 28 count. If including such accessories as nav and coms gear, of course, computerization is needed to make this stuff tick. To be prepared for failures, you need to carry 28 spare microprocessors (assuming that they are all different, and that they are in critical equipment). But most chips today are surface mount, so that is a whole other repair issue. Perhaps electronics are relatively reliable compared to mechanical devices, but they can still fail.

The old school guys like to keep things simple and not have electronic overload.



Why can't an alternator be protected from overload by a breaker or fuse? Agreed that they cannot be protected by these means from overheating. A simple system with thermal sensor inside or on the housing coupled to a relay in the exciter coil should be a simple preventative for overheating. When the sensor trips, it opens the relay contacts which would switch off current to the exciter. When the machine cools down the circuit resets. If you want to prevent loading an alternator at low RPM, a simple speed sensing circuit connected to the exciter relay can handle that. Indication that the alternator is not providing charge can be included. Old systems had a charge/discharge meter. Such an arrangement would eliminate complex circuitry. Spares could be reduced to a relay and a few sensors. Anyone with basic electrical knowledge could repair any failure.

[PaulCrawhorn]

It is usually silly to have more than one general House bank, but there are specialised exceptions.

This bears repeating, with all these complex hypothetical speculations, the simple old-school well-proven solution:

Nothing wrong with a kill switch for other reasons, but in this context there is NO Need to shut down the alternator!

Just take the LFP House bank offline from the charge bus - which is likely to also have solar and/or input, shore power charger, maybe a generator, all these also need to be prevented from hurting your expensive bank.

To protect the alt from the resulting load dump - not at all mythical - simply leave the dedicated Starter batt on the engine/alternator side, or if you do not use one, any cheap lead batt will do. Small in capacity, fully charged or not, internal or external VR, matters not a whit, the lead batt WILL effectively buffer the load dump and protect the alt diodes engine circuitry etc.

And the DC loads, including inverters, can continue being fed by the alternator, not energy stored in House, so long as the motor is running. Once the engine is shut down, reconnect House to the load buss; this can be done via VSR, or manually if you prefer.

Remember, we are not just talking about taking House off-line when it gets to fully charged. The whole rationale here is that LFP sitting at high SoC reduces longevity, so unless there WILL be loads ready to start pulling down House SoC, within a few hours anyway, you should be isolating House from the charge bus at around 3.2Vpc (12.8V), not allowing the LFP to get up to anywhere near full.

[goboatingnow]

Why can't an alternator be protected from overload by a breaker or fuse? Agreed that they cannot be protected by these means from overheating. A simple system with thermal sensor inside or on the housing coupled to a relay in the exciter coil should be a simple preventative for overheating. When the sensor trips, it opens the relay contacts which would switch off current to the exciter. When the machine cools down the circuit resets. If you want to prevent loading an alternator at low RPM, a simple speed sensing circuit connected to the exciter relay can handle that. Indication that the alternator is not providing charge can be included. Old systems had a charge/discharge meter. Such an arrangement would eliminate complex circuitry. Spares could be reduced to a relay and a few sensors. Anyone with basic electrical knowledge could repair any failure

The issue is not about protecting the alternator from temperature . The issue is what to do if a battery protection trips and the alternator is disconnected from its load

You cant fix that ( once the disconnect has occurred ) by controlling the field current.

we all agree the best way for a BMS to shut the alternator is either control the regulator, ie via a signal to the regulator , or directly via a field current control system that overrides the regulator.

by the way theres nothing particularly complex in a LinBus controlled regulator, its just a form of serial port . All alternators these days have microelectronics inside them anyway

[goboatingnow]

Just take the LFP House bank offline from the charge bus - which is likely to also have solar and/or input, shore power charger, maybe a generator, all these also need to be prevented from hurting your expensive bank.

To protect the alt from the resulting load dump - not at all mythical - simply leave the dedicated Starter batt on the engine/alternator side, or if you do not use one, any cheap lead batt will do. Small in capacity, fully charged matters not a whit, will effectively buffer the load dump and protect the alt diodes engine circuitry etc

And the DC loads including inverters continue on fed by the alternator not House long as the motor is running. Once the engine is shut down, reconnect House to the load buss, can be via VSR or manually if you prefer.

Its really not quite as simple as that , a small ( say starter battery ) which is fully charged will not fully soak up the inductive spike due to the rapidly changing current . say where the alternator goes from fully loaded to virtually unloaded . Your electronics attached to the alternator needs to be quite robust to handle the resulting spike. Rapidly changing alternator currents is not something you normally see on a boat ( ie within 400 mS)

If you leave electronics still connected to the alternator as that happens at the very least these electronics must be capable of withstanding 35 volts for 400mS as per ISO 7637-2 Test pulse 5B, If the alternator doesn't have suppression , ie avalanche diodes that the electronics have to withstand voltages to Test 5A , closer to 100V !!

The later IEC/ISO spec ISO 16750-2 , is even worse because the protection must work for a least 10 consecutive overloads , Thats a lot of heat in those TVS didoes

Now everything in your car is supposed to meet these standards , but on a boat we have all sorts of stuff connected to the 12V system and its actually a big task to adequately build electronics such that about 160 Joules of energy have to be handled as over voltage and current . I have just finished a LINbus current sensor and the TVS diode are huge things from ST micro and that's just to 35V ( ie protected alternators. My valeo alternator has avalanche diodes .

I fully agree that HVC events in a Lithium system should never actually occur, certainly its not the way to stop charging , but its not a trivial design exercise to make the system robust under worse case disconnects , ie a fully loaded alternator suddenly disconnected.

Again you have to specify for worst case , ie the disconnect occurs when the alternator is working hard and the remaining loads after disconnect are light

remember its a simple formula , V=L(di/dt), L is about 350uH for a typical alternator .

relay contacts for example open initially in about 5uS,

plug that into the formula at a 70A disconnect !!!