Myth of alternator damage cause by BMS disconnect?

[s/v Jedi]

Originally Posted by rgleason

Frans Veldman replies to "tytower" who is spreading more misinformation here and he finally locks the thread after a rude response.


I had linked to littlefuse some time ago when I wanted to understand this better and because it looked like it could be an interesting backup. I don't recall who referenced it earlier, but one of the knowledgeable regulars here.

Do I understand correctly that littlefuse is a good solution for automobiles, but not our use case, in a marine environment, using larger alternators to charge large batteries?

Should we even bother to consider one as backup?

I wonder how the Balmar Alternator Protection Device works and what it's design approach is?

In this PDF

I believe the APM is intended to be used as a backup only. But sailorboy perhaps could use this, and try a full dump to see how it works out.

It will have a bunch of TVS’s as well as monitoring circuitry to see if a protection event lead to self sacrifice of the parts or not. There are other similar devices. I don’t need them because I use DC-DC converters, but may put one on the central busbars to help in case of near lightning strikes etc.

[wsmurdoch]

Originally Posted by s/v Jedi

I ... may put one on the central busbars to help in case of near lightning strikes etc.

For lightning protections, does the TVS need to be a bi-directional device?

[s/v Jedi]

Originally Posted by wsmurdoch

For lightning protections, does the TVS need to be a bi-directional device?

TVS diodes exist in bidirectional polarities but you can also connect them anti-parallel when using unidirectional types. I have no idea what is actually inside these products like from Balmar but would hope they list that and explain use cases incl. for main busbars. I studied a different brand but forgot which one for now…

[team karst]

Yeah, lightning surges include both polarities.

[wsmurdoch]

Originally Posted by s/v Jedi

I have no idea what is actually inside these products like from Balmar but would hope they list that and explain use cases incl. for main busbars.

Yes, Chris (witzgall) in post #111 (page 8) said, "The [Balmar] APM could be used to protect electronics installed at the breaker panel, but we are not making this claim as we have done no testing for this, and do not plan to do this testing at this time." He did not say whether the electronics protection offered by their APM would be from an alternator surge where the red wires would be positive with respect to the black or from a lightning strike where the red wires could be either polarity.

[Jes]

Originally Posted by goboatingnow

People need time understand there’s about 160 Joules of energy in the alternator spike. Mosfet crowbars typically have VDs of 40-60 V. Way too low , keeping the mosfet within its Safe Operating Area is a huge challenge and rapid Di/Dt changes cause unintentional mosfet operation

Secondly using TVS diode even big ones is a challenge to keep them within the absolute maximums. The problem being that each excursion beyond those limits weakens the TVS structure.

The other thing is TVS limits will be well above the operational voltage so electronics remaining in the circuit still needs to handle over voltage spikes.

It’s actually a very challenging design situation , I’ve a current auto qualified board on my pcb system and it’s a big challenge to make it repeatedly survive these incidents.

The best way is to avoid the

Do you have references of measurements or calculations for the amount of energy in an alternator spike to be 160 Joules?


In an earlier response, the "typical" inductance of an alternator was estimated to be 300 .. 400 micro Henry.
Energy contents of an inductor is calculated by:
Energy = 0.5 * L * I^2
For an alternator producing 100A of current at the moment of battery disconnect that results in a calculated energy contents of 20J.
That's in a range which can be filtered and absorbed by Metal Oxide Varistors.

[team karst]

I come up with 2 Joules using 400 uH. A quick searched showed a 12v 100 A delco has 390uH windings.
So thats just the energy stored in the winding. But, the alternator is still spinning, with Field current, still making perhaps far more energy than what is stored in the windings.
So, now its V*I for some amount of time, controlled by the regulator trying to shut down.
Half a second of 100A @ 12V is now 600 J.
This would be the sudden loss of battery scenario.
OTOH, I would propose that the loss of the alternator connection would be a high voltage issue, taking out alternator semis.

[Jes]

Thanks for correcting my calculation: the coil energy is only 2 Joule.
And you are right that additional potential energy is available in the alternator as long as the field current remains.

I expect that the induction energy from the coil is the most important reason for the high peak voltage.
I do not expect that the energy caused by the continuing field agitation has an inductive behaviour: I do expect an open terminal voltage of the alternator as a result of remaining generator field (during 200 .. 400 msec as determined by the regulator).
Unfortunately there is almost no information available on the internet available how exactly the load dump pulse energy is build up.

The investigations of several research institutes have resulted in international standards (ISO 16750, ISO 7637), where for test purposes the load dump pulse is defined allowing laboratory test generators to be developped.
These test generators can be used to test load dump resistance of alternators and connected consumers.
Test results seems to be satisfactory: probably the dump pulse definition for generators is at the "safe side".

There are some articles to calculate the energy contents of the (ISO 16750, ISO 7637) load dump pulse
Based on the ISO 16750-2 dump load pulse definition, I made next calculation to estimate amount of energy in the dump load pulse.
Since 16750-2 defines two situations (maximum induction voltage combined with minimum alternator impedance and minimum induction voltage combined with maximum alternator impedance) my estimation is:

Dump Load pulse for 12V alternator (ISO 16750-2):
=================================================
V alternator 14 V 14 V
Ri (0,5 .. 4) 4 Ohm 0,5 Ohm
Vs pulse (79 .. 1010) 101 V 79 V
T pulse (40 ..400) 400 msec 400 msec
V pulse start stop 22,7 V 20,5 V

V clamp (SLD15-018) 20 V 20 V
I Peak 20,25 Amp 118 Amp
P Peak 405 Watt 2360 Watt
T Conduct 400 msec 400 msec
P Avg 202,5 Watt 1180 Watt
E adsorb 81 Joule 472 Joule

(please correct me when this calculation set-up is wrong).

The website of Littelfuse provide information for a full protection of load dump pulse by using TVS diodes (SLD and SDLS series).
Littelfuse also provide an application note to explain the degree of protection:
https://www.littelfuse.com/media?res...plication-note
Wonder how these diodes are able to maintain their junction temperature under 175 Celcius.
But the application note is very interesting, and I wonder who has field experiences with these kind of TVS diodes.

[team karst]

Originally Posted by Jes

...

I expect that the induction energy from the coil is the most important reason for the high peak voltage.
I do not expect that the energy caused by the continuing field agitation has an inductive behaviour:...

I tend to agree. Load dump is a transient condition of generated power with no place to go. It causes distress unless countermeasures are in place. Either absorptive devices, or possibly much faster acting regulators eliminating field current.

[s/v Jedi]

Originally Posted by team karst

I tend to agree. Load dump is a transient condition of generated power with no place to go. It causes distress unless countermeasures are in place. Either absorptive devices, or possibly much faster acting regulators eliminating field current.

Unless the BMS can give a warning signal before actually disconnecting the battery and the regulator being triggered by that warning signal toshut down field current, it won’t be fast enough.

People must get sick of this but I’ll repeat the optimal solution once more: there are two scenarios; one is for people who depend on their alternator for charging the house battery and the other for those who have a different primary charging method like solar panels or a genset:

scenario 1: primary solar and/or genset for battery charging: scale down the plans for a super high output alternator with Rolls Royce class regulators because this is just a secondary charging source. I used to have three (!!) engine mounted alternators totaling 530A charge current at 12V. After the change to solar power, I removed all three and put one nice Balmar 170A alternator with serpentine belt upgrade back with their smart regulator with temperature sensors on both the alternator and the start batteries it charges. Also, I installed two dc-dc converters that can take up to 60A ofthe alternator output and charge the lithium house bank. Nothing can go wrong by design, works like a charm.

scenario 2: alternator is the primary charging source for the house bank because there is no room or budget for solar. This can he done in two good, fail proof ways. If the house bank is 12V you can set any smart regulator for lithium profile but connect its output to an AGM start battery. Now add an automatic charging relay that parallels the house bank when alternator output is detected. The AGM start battery will have a reduced lifespan because it isn’t correctly charged but when you start with a good battery like Odyssey, Lifeline etc. it will still last for 5 years or more. It keeps the alternator completely safe.
the other fail proof way requires much more budget. You connect the alternator output to the house battery, then add a dc-dc converter to charge the start battery from the house battery when a charge source is detected (Victron Smart Orion). You can configure that converter for the correct charge profile of the start battery for optimal lifespan. However, this requires that you have a BMS that is able to send a warning signal to the regulator about an imminent disconnect and a regulator that stops charging (kills field current) when that signal is received. For the Balmar regulator I have, you use that signal to disconnect the brown ignition wire from the regulator using a relay. You don’t need a special regulator but do need that special BMS. The popular drop-in batteries can’t do this; youneed expensive systems from companies like Victron or you have to build it yourself.

That’s it. Forget special protection devices, those are nice to add to the solutions described above for backup in case something doesn’t work as designed, not for primary use depending on it because there is no guarantee it’ll work.

[Stu Jackson]

Originally Posted by s/v Jedi

Unless the BMS can give a warning signal before actually disconnecting the battery and the regulator being triggered by that warning signal toshut down field current, it won’t be fast enough.

People must get sick of this but I’ll repeat the optimal solution once more: there are two scenarios; one is for people who depend on their alternator for charging the house battery and the other for those who have a different primary charging method like solar panels or a genset:

scenario 1: ...........

scenario 2: .....................
That’s it. Forget special protection devices, those are nice to add to the solutions described above for backup in case something doesn’t work as designed, not for primary use depending on it because there is no guarantee it’ll work.

Thanks so much for a dose of reality. It bears repeating: no one can build something that doesn't exist, so you use what is available that supports the mission. That and KISS without breaking anything, knowing well ahead of time what the pratfalls are. Thx again.

[RaymondR]

Nah,

I have a 5 hp Honda belt driving two alternators as a battery charger.

Both alternators have internal regulators which are mounted with two screws and which connect to the alternator through a couple of spring devices when contact when you insert the regulator into the alternator body.

I have previously soldered wires to these terminals on other alternators to allow boosting the alternator current and voltage output (Generally by using the wiring to connect to a DC motor speed controller in parallel with the regulator)

There is absolutely nothing to stop me modifying the regulator wiring so that the BMS relay disconnects the alternator's excitation current with a few dollars worth of wire and a half hour with a soldering iron.

[LeighC]

Hi RaymondR , could you share with us the DC motor speed controller you use & a quick diagram please, I have the brush wires extended so I can remotely switch the rotor (field)
current off or reduce it.
Would love to hear your experience so I can pre control the alternator output so the alternator is not damaged in any way before it does a BMS disconnect if it ever does.
If you think this going off topic could you PM me please.
Thanks - Leigh

[RaymondR]

Originally Posted by LeighC

Hi RaymondR , could you share with us the DC motor speed controller you use & a quick diagram please, I have the brush wires extended so I can remotely switch the rotor (field)
current off or reduce it.
Would love to hear your experience so I can pre control the alternator output so the alternator is not damaged in any way before it does a BMS disconnect if it ever does.
If you think this going off topic could you PM me please.
Thanks - Leigh

The ones I generally used were built from a kit marketed by a corporation which no longer exists but they were designed for 12V and 10 amp output so any motor speed controller with these specs should do the job. They were wired on parallel with the internal regulator so that when you wound the control back the internal regulator took over.

Not a problem with off topic and I am a repeat offender re thread drifting and have a fairly thick skin anyway.

There are two types of excitation regulation depending on where in the circuit the regulator is placed. Also there are dozens of wiring arrangements of wiring in alternators so it's difficult to generalize a circuit but what you want to do is place a relay in the regulator circuit to break the circuit when the BMS wants to remove the charging current to the battery.

If you are concerned about a voltage spike damaging electronics put a diode in the circuit to short out the rotor coil similar to what we do with electromechanical relays.

[LeighC]

Thanks Raymond, I have bought a 10 amp motor speed controller (as you suggest) that I wanted to put in series with my negative field coil brush before the reg but it has a common + (pos), (eg controls the - neg) I cant see how it will let the original reg. work as a back up. My alternator is 12 volt 80 amp driving from one 10mm gates V belt so I want to back it off a bit.
Cheers - Leigh