DC DC Charger for alternator and lithium charging

[WaldPinkler]

There was another thread on this forum to ask about a 60A DC DC charger from alternator. Unfortunately no real answer.

The main issue is the price. Like $500 for a 20A charger or $700 for a 40A. It seems excessive. Is there really something special about a boost converter? I mean the lithium charge cycle isn't rocket science.

I believe the basic idea is the alternator charges the lead acid bank, which when running, switches on the DC DC charger to the lithium bank. Many alternators may cook themselves going straight to lithium due to their low resistance and alternator outputting at maximum. Also is the issue of the lithium BMS switching off when full so there's nowhere for alternator power to go.

My alternator puts out 80A so much would be wasted unless multiple chargers are used. It's a dumb alternator but says it has temperature compensation however that works? But how to let it output to lithium bank and have the starter charge also?

The question is, what is the high price for? Are there cheaper alternatives for DC DC charging alternator to 12V (14.6V max) lithium? It seems to be a such a simple operation.

[richard.cs]

The question is, what is the high price for? Are there cheaper alternatives for DC DC charging alternator to 12V (14.6V max) lithium? It seems to be a such a simple operation.

If the alternator is doing nothing else but charge the lithium battery then I have no idea why you would use a DC-DC for this or have a lead-acid stage. Instead rip out the lead-acid-optimised alternator regulator and replace it with one that produces the output voltage and current limit you want, with whatever temperature compensation you like.

I would start by deciding on the lithium charging algorithm, that then gives you voltages, currents, temperature compensation, time profiles, etc. That can be implemented in a small micro as the top-level behaviour, which provides inputs to two control loops (probably also in the micro, though these loops could also be implemented externally with opamps), one voltage and one current. The hardware has to provide suitably scaled voltage and current measurements, and the control loop adjusts the alternator field current servo the output to whatever it should be according to the charging profile.

One of the advantages of pulling the control loops into the micro is you can better optimise the transition behaviour between CC and CV modes, but it also just means fewer parts and easier to tweak.

[WaldPinkler]

There's a lithium house bank and a lead acid engine starter. Makes sense to keep it this way.

Programming a micro with the lithium charge cycle is no problem. I have voltage and current sense devices already that can be attached to battery and alternator. Some temperature monitors on the way actually.

Unfortunately alternator regular is internal. Not sure why kind of rabbit hole it is to replace but I think it's doable. A nice project and possibly not that complicated.

[richard.cs]

There's a lithium house bank and a lead acid engine starter. Makes sense to keep it this way.

I agree, I would keep a lead acid for the starter, but if all your load is on the lithium it makes sense to optimise the alternator to lithium charging and charge the starter battery either with a small DC-DC from there or with a separate small alternator. Unless you're starting and stopping many times an hour there's not a lot of average load on the starter battery.

Unfortunately alternator regular is internal. Not sure why kind of rabbit hole it is to replace but I think it's doable. A nice project and possibly not that complicated.

Looks like you have a combined regulator and brush assembly, that's pretty common, a little harder to mod than when they're separate but not too big a deal. I have modded one like that before, I don't remember the specifics but I think I pretty much just pinged it open, snipped the regulator wires and soldered my own on and clipped/glued it back together. If you trash it they're cheap enough to buy and try again. The only annoyance is that you have to do it again if/when you change the brushes.

[WaldPinkler]

I thought a big advantage of alternator to lead acid was to avoid a load dump where the lithium BMS disconnects when full then burns out the diodes.

If a modified regulator for lithium is added then it's not really a point. And you're right, 10A buck boost is pretty cheap to go DC-DC for the starter bank. Unfortunately I think sail winch, anchor which and thruster are all on the engine circuit. Nothing is simple.

[richard.cs]

Hmmm, with those loads on the starting battery, I would probably go to a dual alternator setup unless you really can't manage it mechanically.

I guess the point I am really making is that adding in power electronics is a relatively expensive and complex way to do something that could be achieved by modulating a field winding, an alternator is after all a kind of magnetic amplifier.

Could you perhaps share a little more about your application? Clearly it's a sailing boat, but is this a propulsion engine where charging is secondary (charging usually solar, etc.), or is the engine routinely used for charging only? The latter would mean that it spends long-hours lightly loaded with probably quite poor efficiency and then there would be benefit in upping the charging power as much as battery will allow. The efficiency of conventional car-type alternators is also pretty dire, especially at high speeds where the iron losses become dominant.

[David Hess]

Unfortunately alternator regular is internal. Not sure why kind of rabbit hole it is to replace but I think it's doable. A nice project and possibly not that complicated.

Regulator design for an alternator is fairly simple; I designed and built one from scratch once.  The alternator acts as a current amplifier so that the current into the field terminal, which appears something like a 4 ohm resistance to ground, is multiplied by a factor of like 10 times to produce the output current.

I thought a big advantage of alternator to lead acid was to avoid a load dump where the lithium BMS disconnects when full then burns out the diodes.

Some type of power shunt protection would be a good idea to limit the output voltage when battery charging stops if the battery is not directly connected to the alternator.

[WaldPinkler]

The details are that I just bought the boat. There are 2 lead acid batteries in the starter bank. The house lead acid bank was dead so I replaced with 270Ah lithium but that leads to learning a bunch about that. It's not a "drop-in" replacement as marketing suggests. There is a VSR (voltage sensitive relay) to switch on house charging when starter reaches a high enough voltage.

Those high power winches, etc are on the engine battery side of the circuit to be used when alternator is running. I'm sure the new lithium can handle them though.

The alternator is secondary charging. Yes, solar primarily. A 2nd alternator is a bit of a pain requiring space, mounts, pulleys.

Normally problems with lithium and alternators are the low resistance causing the alternator regulator overheat and the BMS disconnecting suddenly when full. The other is what happens when the lithium is full but the alternator is running? This could be happen in the unfortunate case of needing to motor a whole day. It's just floating the charge voltage at whatever level that can't be adjusted.

The regulator has a thermistor but is that just lowering the voltage as temperature increases? Defeats any precise charging.

[NiHaoMike]

In the long term, perhaps look into replacing the starter battery with some supercaps? That should solve the problem with providing a low impedance path for transients if the BMS disconnects the pack.

[richard.cs]

Normally problems with lithium and alternators are the low resistance causing the alternator regulator overheat and the BMS disconnecting suddenly when full. The other is what happens when the lithium is full but the alternator is running?
...
The regulator has a thermistor but is that just lowering the voltage as temperature increases? Defeats any precise charging.

Essentially this is just a symptom of the alternator being optimised for lead acid charging. It's not got current limiting (beyond its inherent impedance), it's got the wrong temperature coefficient, etc.

Given you can't simultaneously optimise the alternator for both I guess you have a few options:

• Standard alternator regulator, power electronics to charge the lithium, as per your original suggestion.
• Custom, lithium-optimised alternator regulator, power electronics to charge the starting battery, probably move the heavy loads to the lithium so you only need say 10A of charging.
• Two alternators, giving full electrical flexibility at the expense of mechanical complexity. It's a shame no-one makes a device with two independent alternators in one package, it'd be good for split charge applications generally (hmmm, not too bad a machining project to Siamese two alternators).
• Some sort of compromise regulator which gives output characteristics suitable for both lithium and lead, with a split charge relay. Kind of what you have now but add in some current limiting and tweak the temperature coefficients, etc. so it's better for lithium and just accept the suboptimal lead-acid charging that results.

If you have the alternator connected to the lithium battery then something to aid load dump survival if the BMS decides to disconnect (it shouldn't if you get the regulator right, but still might) is sensible. Shunting it to the starting battery might be an option.

[David Hess]

Normally problems with lithium and alternators are the low resistance causing the alternator regulator overheat and the BMS disconnecting suddenly when full.

Overheating of the alternator happens with lead acid batteries also if the lead acid battery is large enough or dead enough.  More than once I have diagnosed a bad alternator after someone jumped their car from a dead battery because the alternator burned out recharging the dead battery.  In principle the alternator should protect itself but but car stuff is cheap.

[f4eru]

Many alternators may cook themselves going straight to lithium due to their low resistance and alternator outputting at maximum

Very very simple solution :
replace or modify the regulation circuit of the alternator in order to achieve CC-CV regulation, and take input from the BMS (original is only CV)

No need for Pb batteries at all. Keep it simple and straightforward.

[richard.cs]

Yes, standard alternators don't regulate current at all, they don't measure it. All that saves them from cooking most of the time is that the output impedance is high, the windings have a lot of leakage inductance and at high RPM they act as a pretty good current source. With constant high loads they can still die, but usually the battery gets full before the alternator catches fire.  Old cars with DC generators on the other hand do have regulators that measure and control for current because Zout is much lower and they're more like voltage sources. Generally more efficient too, but with the downside of lower speed range and lesser cooling.

[f4eru]

A big beefy specialized TVS will protect against load dump.

[WaldPinkler]

Great advice here. There's lot of confusing information out there regarding lithium change overs. Mostly from companies selling over priced DC-DC chargers. It really makes more sense with a CC-CV regulator to the lithium and a smaller DC-DC charger for the lead acid.

[BrokenYugo]

Depending on the alternator already in there you may be able to get an old style external regulator alternator in the same case. I've done that swap the other way many times.