LiFePo4 Power -- Controlling Alternator and Inverter/Charger

[CatNewBee]

Originally Posted by Pete7

Interesting YT video and on topic for this forum. Nice to see real life testing and demonstrating problems rather than just internet hearsay.

Pete

Nice, but somewhat unscientific....

I would expect when comparing different alternators and brands to each other to use the exact same preconditions.

It makes no sense to run the one at 3000rpm and 1500 rpm and compare it to the other 2200rpm and 3600rpm, especially if you want to show that one is better than the other.

Also I would compare apples to apples, means both alternators should be the same amp rating at same rpm.

Modern alternators with internal regulators do have heat protection. Even car alternators nowadays have a lot of electronics inside, especially for fuel efficiency, that regulate the current down on high rpm to reduce the load on the engine while accelerating by CAN bus integration etc. It is a whole plethora of things that may get wrong.

In modern boats there is a FET distribution device with feedback to the alt, that decouples the start battery from the bord battery too and may limit the current also. I do not agree with the statement that FLA batteries limit the current in general, it depends on the capacity connected, the alternator tested was not dessigned to charge a big battery bank either, but to keep the engine running, feed some low power systems and recharge a small start battery. It was not fit for purpose. A 300Ah LFP can be charged with 1C and quick charged with 3C, a boat usually has more than one of those, I am not aware of 12V alternators with this high ratings.

A current limiter is also not the solution if the findings are right. Assume the first alternator tested, 3000rpm 78.9A and 1500rpm 65A. A current limiting device is not sensitive to the temp, assume you choose one for 60A deeming right for a 80A alternator (75% of the rating) , it would limit your charging at safe high revs and will not save your alternator at 1500 or lower revs. So this recommendation is only valid if you drastically limit the current lets say below 50% of the rating, sacrificing a lot of charging capability.

Even in cars today the electric system draws a lot of amps, constantly rechargestart/stop batteries, strong headlights, heating systems for the rear window, mirrors, power seats, fans running for heating and cooling, hifi components and electronic toys. Alternators and batteries must cope with the much higher energy demands.

Best solution is to chose an alternator for low rpm with internal protection from overheating.

[CatNewBee]

... BTW they test the alternator at alternator rpm of 1500. Usually there is a transmission of a big pulley at the engine and a small at the alternator multiplying the rpm on the cranc shaft.

Many alternators work efficiently between 6000 and 12000rpm given a ratio of 2:1 for gas engines or 3:1 for diesel (3000..6000 rpm vs 2000...4000rpm at the cranc shaft)

[john61ct]

Originally Posted by CatNewBee

Modern alternators with internal regulators do have heat protection
…
Best solution is to chose an alternator for low rpm with internal protection from overheating.

The usual overtemp protection is to lower the voltage, in effect render charging ineffective.

The only way afaik to continue to maintain setpoint voltage during overtemp condition,

IOW using **reduced current** loading to keep temperature reasonable, is with an external VR designed to do so, or with a DC-DC charger.

And maybe boats currently sold as new are better, but that is irrelevant to 99.99% of members here.

As are car charging systems, often the ECU acts as the VR, making B2B DCDC chargers **much more** essential, not less.


> I do not agree with the statement that FLA batteries limit the current in general, it depends on the capacity connected

I've seen very brief acceptance at 0.18C, but usually 0.12 - .15C, and that only when depleted, 40-60% SoC.

And of course by definition, C-rate is proportional to capacity.

This "natural current limiting" is why traditional alternator/VR design has not included this functionality, now essential for newer banks pulling huge ampacities.


> the alternator tested was not dessigned to charge a big battery bank either, but to keep the engine running, feed some low power systems and recharge a small start battery. It was not fit for purpose.

Bingo! yes exactly, that is why retrofitting / upgrading is required!

> A current limiting device is not sensitive to the temp, assume you choose one for 60A deeming right for a 80A alternator (75% of the rating) , it would limit your charging at safe high revs and will not save your alternator at 1500 or lower revs. So this recommendation is only valid if you drastically limit the current lets say below 50% of the rating, sacrificing a lot of charging capability.

Current limiting is usually required, not optional. A DCDC charger has its "crude" limited amps choices.

An MC-614 can be configured with the "cruder" lower resolution "small engine" mode, or the slightly more elegant "belt manager".

Apparently the Wakespeed WS500/AP500 VRs have a lot more flexibility.

But **of course** however you do it,charging speed is reduced, that is the goal, the too-high amps demanded by LFP **is the problem being solved** here, limiting the current is not "optional", the only question is **how** to do so.


> A 300Ah LFP can be charged with 1C and quick charged with 3C

Can be, does not mean should be, doing so regularly will greatly reduce longevity. Not just the charge source, but also the LFP bank requires the protection from its own too-high current acceptance.

0.5C rate IMO should be the fastest charge rate, 0.3C is healthier.

> boat usually has more than one of those, I am not aware of 12V alternators with this high ratings.

So go to 24V or higher if you need over say 3kW, or multiple alts.

The difficulty is installation space on smaller boats.

[evm1024]

Originally Posted by john61ct

The usual overtemp protection is to lower the voltage, in effect render charging ineffective.

The only way afaik to continue to maintain setpoint voltage during overtemp condition,

SNIP

Perhaps you would like to think about this statement and revise it....

Specifically a voltage regular senses the voltage at some nodal point (the battery terminal in a typical external voltage sense case) and outputs a field current in proportion to the difference between the voltage set point (the desired voltage a the nodal point) and the actual sensed voltage.

The statement I quoted appears to be nonsense.

Perhaps it is my lack of understanding of what you meant to say or your lack of understanding of how voltage regulators work or perhaps just a plain misstatement.

Simply scaling the error voltage factor based on temp is quite easy to implement but hard to get optimal. And of course with microprocessor controlled voltage regulators we can get all kinds of control loops.

[john61ct]

Sorry, do not see any conflict there. I am not referencing the internal workings at all, just dealing with the resulting output, as if from a black box.

Obviously the "striving for" setpoint during CC/Bulk stage is different from the holding CV max during Absorb stage, if that's what you're referring to.

My point is the extreme dropping of voltage stock VRs use to deal with overtemp conditions, in effect halts effective charging for long periods of time.

While lowering the current output while keeping voltage at the desired setpoint, as accomplished by the devices under discussion here, results in the battery actually reaching the 100% Full point both **more quickly and** with greater precision, before dropping voltage to Float.

Most stock alt setups don't even know from Float at all. . .

[evm1024]

Exactly how do you maintain a setpoint voltage in a less than 100% SOC battery?

You do not. Clearly a misstatement or misunderstanding.

As for effective charging - It the battery is accepting a charge then the charging is effective. Of course you did not mean it that way. You implied that the reduced charging current due to heating would reduce the charging current to an unacceptable level and therefore be ineffective. Very subjective. And not true.

The interaction between the battery voltage (at the V sense nodal point), The error voltage between the setpoint and battery voltage, the field current, and the temp of the alternator and the alternator output is really quite complex.

Simple regulators just reduce the field current in a effort to keep the alternator from death. Reduced field current results in lower output current which causes a reduction heat generation in the alternator. The reduced current from the alternator causes the sensed voltage to be reduced which increases the error voltage which causes voltage regulator to increase the field current. In an ideal world these all come into balance and your battery gets charged without burning up the alternator. The time constant of the heat mass of the alternator tends to dampen out what would otherwise be a very unstable control loop.

Enter the microprocessor controlled voltage regulators and we can start doing some PID control loops on both the voltage regulation and also on the temp correction voltage.

Toss in Al's VSR with current sensing and we are no longer limited to the thermal mass of the alternator for temp corrections. This is much more like adding a rudder transducer to an AP that you might suspect.

Current sensing ALSO allows the introduction of some very advanced control. For example this current sensing allows for true constant current regulation within the limits of the ability of the alternator to supply current without overheating and the batteries acceptance rate.

As I said, this is really quite complex and when we are talking at a technical level such as this I personally think it pays dividends to be precise in our choice of words.

[CatNewBee]

Should not be that complicated. Just use a thermistor in series with the field current wire attached to the housing. Increased temp leads to increased resistance that leads to decreased field current, what weakens the magnetic field and results in lower current output. It can be that simple.

The voltage regulator in series increases the resistance too if the voltage on the sense wire goes up a certain point, usually set to 14.2V for FLA start batteries. The effect is the same, weaker field current, weaker magnetic field, less output current, less energy transferred, less heat.

Both regulators are independend of each other, while the voltage regulator passes current trough (minimal resistance) the thermistor reduces it, so voltage cannot be increased further the current is smaller, no matter what the difference is on the sense wire, and vica versa. When the alternator is cold and the termistor is at his low resistance, but the voltage is too high, the voltage regulator limits the current.

[evm1024]

Agreed that often "in practice" temp compensation is implemented simply. How well that performs has many factors to consider.

My main point was that the statements:

The usual overtemp protection is to lower the voltage, in effect render charging ineffective.

The only way afaik to continue to maintain setpoint voltage during overtemp condition,

are in error. They appear to be the result of a profound misunderstanding of how alternators and regulators work and interact. I did attempt to correct that.

[tanglewood]

Originally Posted by evm1024

Agreed that often "in practice" temp compensation is implemented simply. How well that performs has many factors to consider.



My main point was that the statements:







are in error. They appear to be the result of a profound misunderstanding of how alternators and regulators work and interact. I did attempt to correct that.

Agreed. A profound lack of understanding of electrical behavior. Reducing field reduces output current. The resulting change in voltage depends on the load which in this case is the battery. Under pretty much all charge scenarios, current will reduce and voltage will stay the same.

Everyone (mostly) knows Ohm’s formulas. But using them requires understanding which are the dependent and independent variables in any given circuit, and it varies widely.

It can be really tricky to sort out what is constant (independent), and what follows (dependent) in those formulas.

[Dockhead]

Originally Posted by tanglewood

. . . Reducing field reduces output current. The resulting change in voltage depends on the load which in this case is the battery. Under pretty much all charge scenarios, current will reduce and voltage will stay the same.

Everyone (mostly) knows Ohm’s formulas. But using them requires understanding which are the dependent and independent variables in any given circuit, and it varies widely.

It can be really tricky to sort out what is constant (independent), and what follows (dependent) in those formulas.

Thanks for this remarkably lucid explanation, from me, an amateur who has long wondered how all this works. Really useful light bulb moment for me

[CatNewBee]

It's not so tricky, it is always the battery that dictates the voltage measured, the open circuit voltage of the charger is only relevant to limit from overcharging at the end of absorption and to limit the charge current to not fry the charger and the cabling. You need a voltage difference to make a current flow across a connection. So if battery and charger are at the same voltage, current will be zero.

[Dockhead]

Originally Posted by CatNewBee

It's not so tricky, it is always the battery that dictates the voltage measured, the open circuit voltage of the charger is only relevant to limit from overcharging at the end of absorption and to limit the charge current to not fry the charger and the cabling. You need a voltage difference to make a current flow across a connection. So if battery and charger are at the same voltage, current will be zero.

"Not so tricky" for folks like you with an engineering background.


Requires effort (and curiosity) for the rest of us to figure out.