Musings Regarding External Voltage Regulation

[Will Gilmore]

Hope this was a somewhat helpful topic on alternator regulation.

Thank you so much for this post, Maine Sail.
I am very interested in this subject and there is a lot of superb information here.
My long-term goals for installing electric power is to run a multi-phase brushless outboard with solar and wind generators for the battery bank. This helps tremendously.
- Will (Dragonfly)

 

[dLj]

If you are getting 15-20 years then something is working in your favor, but it's not likely to be EQ. If I did not know better I would guess you have GEL batteries or industrial batteries like a Rolls or similar. By industry standards a battery is considered dead when it can no longer deliver 80% of its original 20 hour Ah capacity rating. I routinely see this occur in less than 2 years with the typical automotive case size batteries typically used on a Hunter 30.. Even just calendar life kills most basic flooded batteries before 15-20 years and that is without any cycling at all.

So to clarify, I have only ever used flooded lead acid batteries. However, reading all that you have posted on what to do or not do, seems I'm being really quite a "good" battery operator. I never let my batteries go below 75% charge before recharging if I can avoid it. If I do, then I immediately re-charge to 100% and then perform an equalization (my flavor) on the battery. To date, I've not been without shore power for more than about a month, and more frequently only a week at a time. Work so gets in the way of the things I like to do....

Lifeline AGM batteries can be equalized but they are currently the only AGM's that can be.. For FLA batteries the biggest reasons we need EQing are due to improper charging by so called "smart chargers", most of which are are not smart at all & PSOC (partial state of charge cycling) use.

In a perfect world deep cycle flooded batteries should be charged at 14.7V to 15.3V yet very few battery charger makers are anywhere close to that with their rather stupid "dip switch" set points.

Very interesting!

I don't use commercial battery chargers, I only use laboratory power supplies. I've always used 14.8V as my set point for charging FLA's. I have no real experience with commercial battery chargers, and that is one of the motivating reasons for me getting into this discussion. I really need to understand the fundamentals of battery charging beyond my simple working knowledge that is what I've used to date. And I REALLY thank you for taking such an enormous amount of time to answer all the questions that are thrown at you.

It is my desire (time will tell if it is realized) to retire onto a sailboat and sail for some undetermined amount of time, but at this point I'm thinking until my bones no longer want/can do it. Once I am at the point of living on a boat, I then need to have fully developed a system of battery maintenance that does not include having access to shore power as I currently have.

A true EQ is not really done as the charger makers want us to believe, but is the only practical way of doing it in the field.. The way a manufacturer would do it, and some call it a reforming charge, is to charge from a very low SOC such as 0% SOC / 10.5V and use a very low charge current such as .005C to .01C (0.5A to 1A for a 100Ah battery) and to let the voltage wind up where it will. With healthy batteries it can go over 17V towards the end of charge but requires in excess of 100 hours to do so...

First you say the above...

EQ's in the field should ultimately not begin until the bank is already at 100% SOC with your existing charger. If you do that even a 1.3A power supply should be able to push individual batteries over 15.5V. EQ should always be temp compensated and done at manufacturers voltage guidance.

then you say this above. Can you please explain this a bit. It seems to be somewhat contradictory?

The effects of sulfation actually happens far faster than folks assume it does. In a 30 cycle PSOC (partial state of charge cycling) test done for Practical Sailor (May 2015 & August 2015) one popular AGM battery lost 1% in capacity for each PSOC cycle or 30% of it's Ah capacity lost due to sulfation in just 30 cycles..

Oh boy, is this the darned truth! Sulfation happens very fast, especially the lower you go in state of discharge! I have direct hands on knowledge of just how right MS is on this one!

So, you have published in other posts you've made what the end of charge is when charging to achieve 100% SOC (can't recall right now, will have to go look for it again).

When do you know you've reached the end of an equalization cycle?

Another question - I had been told (no idea from whom - not the electrical engineer from Hoover) that when a FLA had been discharged to below 50% charge, that you should use low current charging (what we used to call trickle charging) until it reaches at least 50% charged, then you can hit it with lots of current to bring up to full charge. Is that correct? And if not, what would be correct?

Thank you very much for your time and answers. I'm going to have more questions, I'm sure...

dj

 

[Tribologist]

Thanks you for one of the most comprehensive writeups I ever seen on this subject.! Such a great reading. I been messing around with the charging system on my little 3 cyl Isuzu 3CB1Diesel that is in my Albin 25 that came with a Denso 40A alternator. I replaced it wit a Delco 12si but quickly learned that the Super Dumb regulator drops the voltage as the engine bay heats up. I was able to snatch a Balmar ARS-5 on fleabay for under $50 with shipping so I will most likely convert to that. One thing I been seriously thinking of is to connect a standard Delco regulator but put it remote from the alternator on a heatsink near the batteries. For wiring, I'm thinking it could be as simple as putting a insulated washer under the Diode Trio input and a ring terminal on top of it to run a wire to the external regulator, same for the field conneection. If the regulator dont see any voltage, Im thinking it should stay passive. I could off course make a dummy regulator out of phenolic or something too.

 

[dLj]

With the current state of Lithium Ion battery technology, lead acid on boats is pretty much a historical anecdote.. Just an opinion. I'm now running LiFePo house batteries and am extremely happy!

dj

 

[Tribologist]

With the current state of Lithium Ion battery technology, lead acid on boats is pretty much a historical anecdote.. Just an opinion. I'm now running LiFePo house batteries and am extremely happy!

dj

That is true. I will go that route when my current wallyworld. 3x100 Ah call it quit. Currently i have a m-ACR, a victron 25 A AC-DC charger and a 30 A Victron DC-DC charger. At that point i probably keep a FLA for starting and for my bowthruster and let the alternator feed that and then use the DC-DC to manage the lipos but so far so good. I have 7 years on the wallyworld house batteries.

 

[dLj]

That is true. I will go that route when my current wallyworld. 3x100 Ah call it quit. Currently i have a m-ACR, a victron 25 A AC-DC charger and a 30 A Victron DC-DC charger. At that point i probably keep a FLA for starting and for my bowthruster and let the alternator feed that and then use the DC-DC to manage the lipos but so far so good. I have 7 years on the wallyworld house batteries.

I see folks doing this fairly frequently but it's really penny wise and pound foolish.

I have an AGM start battery and two AGM batteries in the bow running my bow thruster and windlass. All charging systems run into my LiFePo house bank and then I run DC to DC charging systems off my house bank to charge the lead acid chemistry. This allows full advantage of both systems - I can charge my LiFePo batteries at the rate they can absorb - and in turn they will charge the lead acid batteries to 100% charge, what those batteries need, over whatever time frame may be needed.

Compare to charging through your lead acid battery - you are throttled back so you can't charge at a rate above what the lead acid permits. You probably never fully charge the lead acid batteries as that takes hours and you'll stop after getting to maybe around 90% maybe a bit more. But you'll never fully charge them back up. So here you are back to lead acid land where you can't take advantage of the benefits of the rapid charging that minimizes costs and time to charge. You still aren't getting your lead acid batteries fully charged so you are still killing your lead acid batteries ... Makes no sense to me...

dj

 

[Tribologist]

I could do it that way too if i simply upgrade the alternator regulator. That would allow for faster charging of the house bank. The 30A victron dc-dc would be happy charging the start battery and the thruster battery. But for now my main issue is to get enough juice from the alternator. Im getting ARS-5 on thursday to play with.

 

[Davidasailor26]

Compare to charging through your lead acid battery - you are throttled back so you can't charge at a rate above what the lead acid permits. You probably never fully charge the lead acid batteries as that takes hours and you'll stop after getting to maybe around 90% maybe a bit more. But you'll never fully charge them back up. So here you are back to lead acid land where you can't take advantage of the benefits of the rapid charging that minimizes costs and time to charge. You still aren't getting your lead acid batteries fully charged so you are still killing your lead acid batteries ... Makes no sense to me...

Your system is better, but I don’t think the indirect charging approach needs to be as bad as you describe it. The lead acid banks for a starter or a thruster never get deeply discharged, so they should recharge quite quickly even if the alternator’s voltage sags a little from the DC-DC load. The slower lithium charging is the typical drawback of this setup - the lithium charging rate will be limited to the size of the DC-DC, and it’s true those are generally smaller than a good high output alternator. Still, for someone that doesn’t want to replace the alternator and is just looking for more capacity regardless of charge speed, it might be a worthwhile compromise.

 

[Tribologist]

Your system is better, but I don’t think the indirect charging approach needs to be as bad as you describe it. The lead acid banks for a starter or a thruster never get deeply discharged, so they should recharge quite quickly even if the alternator’s voltage sags a little from the DC-DC load. The slower lithium charging is the typical drawback of this setup - the lithium charging rate will be limited to the size of the DC-DC, and it’s true those are generally smaller than a good high output alternator. Still, for someone that doesn’t want to replace the alternator and is just looking for more capacity regardless of charge speed, it might be a worthwhile compromise.

Cost vise its better to upgrade the regulator. I paid $100 for a used 30A victron, a 50A unit is $350 but a Li capable Balmar regulator is $289 at Defender. Someone posted that the temp sensors are cheap LM230 10mV/K diode sensors so one can get full charge and thermal management for less than the DC-DC charger . I will keep playing with what i have and when the FLA’s give in I will take a second look.

 

[dLj]

Your system is better, but I don’t think the indirect charging approach needs to be as bad as you describe it. The lead acid banks for a starter or a thruster never get deeply discharged, so they should recharge quite quickly even if the alternator’s voltage sags a little from the DC-DC load. The slower lithium charging is the typical drawback of this setup - the lithium charging rate will be limited to the size of the DC-DC, and it’s true those are generally smaller than a good high output alternator. Still, for someone that doesn’t want to replace the alternator and is just looking for more capacity regardless of charge speed, it might be a worthwhile compromise.

I've never seen state of charge curves for the lead acid battery in the system where you charge your lithium batteries through the lead acid battery so I don't have first hand knowledge. But knowing how the two battery systems charge and the time it takes to fully charge the lead acid, I think I'm pretty accurate.

Your lithium batteries will take as much current as you can throw at them, your lead acid battery is probably going to charge up to somewhere probably around 90%. Then you are going to have to hold the entire system charging until your lithium system is fully charged and the BMS shuts down acceptance to force the lead acid to continue to charge to 100%. I seriously doubt anyone in practice does that. So you are back to never fully charging your lead acid batteries and you cannot take advantage of the lithium ability to charge quickly because it's throttled back by the lead acid.

If you have actual charge profiles of a lead acid battery used this way, I'd love to see them. if I'm wrong, I'm happy to admit it.

dj

 

[Davidasailor26]

Your lithium batteries will take as much current as you can throw at them, your lead acid battery is probably going to charge up to somewhere probably around 90%. Then you are going to have to hold the entire system charging until your lithium system is fully charged and the BMS shuts down acceptance to force the lead acid to continue to charge to 100%. I seriously doubt anyone in practice does that. So you are back to never fully charging your lead acid batteries and you cannot take advantage of the lithium ability to charge quickly because it's throttled back by the lead acid.

That's not really how a DC-DC would charge the lithium though. The DC-DC input would be in parallel with the lead acid, able to get current straight off the alternator. It's not really charging "through" the lead acid, just next to it. Assuming the alternator is an internally regulated "dumb" type it will self-regulate to keep the voltage in the low 14's, or as close to that as its temperature allows, so all the led acid banks will know is that they're connected to a 14+ volt source and able to charge off it until full. If the DC-DC pulls too much off the alternator, or the alternator heats up too much, its output will drop into the mid 13's, but that's still not that different from how those banks would've charged years ago with a larger lead acid house bank.

 

[dLj]

That's not really how a DC-DC would charge the lithium though. The DC-DC input would be in parallel with the lead acid, able to get current straight off the alternator. It's not really charging "through" the lead acid, just next to it. Assuming the alternator is an internally regulated "dumb" type it will self-regulate to keep the voltage in the low 14's, or as close to that as its temperature allows, so all the led acid banks will know is that they're connected to a 14+ volt source and able to charge off it until full. If the DC-DC pulls too much off the alternator, or the alternator heats up too much, its output will drop into the mid 13's, but that's still not that different from how those banks would've charged years ago with a larger lead acid house bank.

This is the point I don't see happening - "so all the led acid banks will know is that they're connected to a 14+ volt source and able to charge off it until full". Once a lead acid battery reaches about 90%, it really slows down in accepting charge. In order to get that last 10% charge - it takes a really long time. Your lithium bank will be fully charged and the BMS will have shut down not allowing more charge coming in. That in itself is not something I would want happening with my LiFePo batteries every time I'm trying to get my lead acid batteries fully charged.

Here's a simple quote "Charging the last 10% of a lead-acid battery takes a disproportionately long time, often several hours or more, because the battery's charge acceptance slows significantly as it approaches full capacity." Of course the time depends upon battery size, but the battery itself will not allow fast charging through this region. For any battery we would be using, it would be hours. At anything less that 100% full charge, the battery will be sulfating. This is one of the main reasons that sailboat lead acid batteries die an early death. If you read the IEEE standards on equalizing lead acid batteries, you will find that the expected lifespan is stated as 12 to 15 years. But sailboat owners seem happy when they get say 8 years from their lead acid batteries. Often they get less.

If you are charging your system using your engine, or a generator, you are using diesel fuel to produce electricity. The cost of charging is quite high in terms of $/KW. It's also quite high in terms of the wear and tear on the unit you are using to charge those batteries. That's the main reason why I say it's penny wise and pound foolish.

Now look at the system where you charge the lithium bank. In my case, I have a 400 amp hours house bank. I run twin alternators and charge at roughly 200 amps. So if my house bank is fully depleted, it takes me 2 hours to fully charge it back up. But I don't even have to worry about that. I can run my engine an hour or hour and a half and have plenty of power in my house bank. That is notably less time to charge any lead acid battery I might be using. Now since lithium batteries don't need to be kept fully charged, in fact they like being charged up to about 90%, I then use those batteries to charge the lead acid batteries in my system. It doesn't matter at all how long that takes. That last 10% is coming out of my house batteries - slowly. My lead acid batteries are fully charged after each and every time I use them. Most of the time I don't even have to worry about what state of charge my house bank is at, they have plenty of power to recharge my lead acid batteries. It doesn't matter what state of charge my LiFePo batteries are in as long as I have enough to run the systems I need running.

I fully expect to get 12 to 15 years from my lead acid batteries running in this fashion. That in itself is much more economical than charging through, or in parallel whatever it may be, my lead acid batteries in the system you are "defending". Then add to that, I rarely have to run my engine to do this as all my charge sources, solar, wind, whatever, go straight into my house batteries. Most of the time I don't need to charge using my engine. So way less engine wear and tear, way less fuel consumed - even if I have no other charge sources but my alternators.

Like I said, show me the charge curve for the lead acid batteries that are set up that way. I'm sure there is not likely a single system that fully charges that lead acid battery. Doing it that way, you are shortening the life of the lead acid battery and you are spending way too much money doing an inadequate job of charging.

dj

 

[Tribologist]

This is the point I don't see happening - "so all the led acid banks will know is t

dj

Thanks! Good food for thought! Very much appreciated. My system is kind of a breadboard. I can certainly experiment with that approach with my current battery bank. My today project is looking at external regulation of the Delco 12si. What I like to do is to is to take the internal regulator and put it where the batteries are so the the temp compensation act on the battery temperature instead. I'm also going to play with the ADR-5 regulator to see how that works.

Ulf

 

[Davidasailor26]

This is the point I don't see happening - "so all the led acid banks will know is that they're connected to a 14+ volt source and able to charge off it until full". Once a lead acid battery reaches about 90%, it really slows down in accepting charge. In order to get that last 10% charge - it takes a really long time. Your lithium bank will be fully charged and the BMS will have shut down not allowing more charge coming in. That in itself is not something I would want happening with my LiFePo batteries every time I'm trying to get my lead acid batteries fully charged.

Here's a simple quote "Charging the last 10% of a lead-acid battery takes a disproportionately long time, often several hours or more, because the battery's charge acceptance slows significantly as it approaches full capacity." Of course the time depends upon battery size, but the battery itself will not allow fast charging through this region. For any battery we would be using, it would be hours. At anything less that 100% full charge, the battery will be sulfating. This is one of the main reasons that sailboat lead acid batteries die an early death. If you read the IEEE standards on equalizing lead acid batteries, you will find that the expected lifespan is stated as 12 to 15 years. But sailboat owners seem happy when they get say 8 years from their lead acid batteries. Often they get less.

If you are charging your system using your engine, or a generator, you are using diesel fuel to produce electricity. The cost of charging is quite high in terms of $/KW. It's also quite high in terms of the wear and tear on the unit you are using to charge those batteries. That's the main reason why I say it's penny wise and pound foolish.

Now look at the system where you charge the lithium bank. In my case, I have a 400 amp hours house bank. I run twin alternators and charge at roughly 200 amps. So if my house bank is fully depleted, it takes me 2 hours to fully charge it back up. But I don't even have to worry about that. I can run my engine an hour or hour and a half and have plenty of power in my house bank. That is notably less time to charge any lead acid battery I might be using. Now since lithium batteries don't need to be kept fully charged, in fact they like being charged up to about 90%, I then use those batteries to charge the lead acid batteries in my system. It doesn't matter at all how long that takes. That last 10% is coming out of my house batteries - slowly. My lead acid batteries are fully charged after each and every time I use them. Most of the time I don't even have to worry about what state of charge my house bank is at, they have plenty of power to recharge my lead acid batteries. It doesn't matter what state of charge my LiFePo batteries are in as long as I have enough to run the systems I need running.

I fully expect to get 12 to 15 years from my lead acid batteries running in this fashion. That in itself is much more economical than charging through, or in parallel whatever it may be, my lead acid batteries in the system you are "defending". Then add to that, I rarely have to run my engine to do this as all my charge sources, solar, wind, whatever, go straight into my house batteries. Most of the time I don't need to charge using my engine. So way less engine wear and tear, way less fuel consumed - even if I have no other charge sources but my alternators.

Like I said, show me the charge curve for the lead acid batteries that are set up that way. I'm sure there is not likely a single system that fully charges that lead acid battery. Doing it that way, you are shortening the life of the lead acid battery and you are spending way too much money doing an inadequate job of charging.

dj

You make some good points about recharging depleted lead acid batteries, but one thing to remember is that the lead acid banks in this setup are never significantly discharged. They’re used for starting and a thruster only. Even though those loads are high they are very brief. 10 seconds of 300 amps should leave the bank over 99% charged. Recharging these banks is basically the same as recharging a car after starting it, and no one asserts a car needs to be run for hours to recoup each start. The rule I’ve heard for cars is that 15 minutes of driving is about break-even. That should be achievable for pretty much all boat engine use cases.

In order to get that last 10% charge - it takes a really long time. Your lithium bank will be fully charged and the BMS will have shut down not allowing more charge coming in. That in itself is not something I would want happening with my LiFePo batteries every time I'm trying to get my lead acid batteries fully charged.

Even if the engine runs for a long time, trying to charge the lead bank or just cruising, the BMS shouldn’t be shutting down the charging in a DC-DC setup. DC-DC chargers like the Orion and Orion XS have proper three stage charging profiles that can manage the lithium charging just like a conventional shore charger, smart alternator regulator, or MPPT.

I still think your setup makes more elegant use of the alternator, but I don’t think the DC-DC to Lithium arrangement is inherently bad for either bank, just less efficient.

 

[dLj]

You make some good points about recharging depleted lead acid batteries, but one thing to remember is that the lead acid banks in this setup are never significantly discharged. They’re used for starting and a thruster only. Even though those loads are high they are very brief. 10 seconds of 300 amps should leave the bank over 99% charged. Recharging these banks is basically the same as recharging a car after starting it, and no one asserts a car needs to be run for hours to recoup each start. The rule I’ve heard for cars is that 15 minutes of driving is about break-even. That should be achievable for pretty much all boat engine use cases.

Most cars are started very frequently, at least compared to boats. Boats can sit for weeks or even months between usage. Lead acid batteries self-discharge - here's a quick quote: "Lead acid batteries self-discharge at a rate of about 4% to 8% of their capacity per month under normal, moderate-temperature conditions, but this rate significantly increases with higher temperatures. For example, the self-discharge rate can double for every 15°F rise in temperature, while colder temperatures will slow the discharge rate considerably."

I don't think the comparison is valid between auto use and boat use. If in the heat of the summer, you don't use your boat for two weeks, your lead acid batteries could be at 8% discharge when you get to the boat. If you are gone a month - well below 10%.

Charging my lead acid off my lithium house bank keeps them at essentially 100% - 100% of the time....

Even if the engine runs for a long time, trying to charge the lead bank or just cruising, the BMS shouldn’t be shutting down the charging in a DC-DC setup. DC-DC chargers like the Orion and Orion XS have proper three stage charging profiles that can manage the lithium charging just like a conventional shore charger, smart alternator regulator, or MPPT.

I still think your setup makes more elegant use of the alternator, but I don’t think the DC-DC to Lithium arrangement is inherently bad for either bank, just less efficient.

Perhaps we'll just have to agree to disagree. I honestly can't see any benefit to the DC-DC to Lithium arrangement and feel it is inherently bad for the lead acid batteries used that way. Perhaps not worse than "normal" sailboat usage - but certainly worse than simply setting it up having your lead acid charged from you lithium bank...

dj

p.s. Here's a part I don't understand on your DC-DC to Lithium setup. How do you put in solar panels, for example? Are they also running through that same controller? What happens when you decide a couple years down the road to add in another solar panel? Or wind generator? Or hydro generator? How do you add that in? In the setup I'm describing, all charging sources go directly into my house bank. I can add in any number of additional sources and all I have to do is put the controller into the charging system of the house bank. Simple. Nothing more to do than just end up with more power available to charge my house bank...

 

[colemj]

What I like to do is to is to take the internal regulator and put it where the batteries are so the the temp compensation act on the battery temperature instead.

I'm not sure how you are going to do that. The voltage regulator is bolted to the brush holder, and they make electrical contact through being in physical contact with each other. The other end of the regulator is tied to the diode pack with a robust solid strap. I don't know how you are going to connect all of that with wiring, and bring that wiring outside the case.

If the alternator is folding back output that dramatically because of temperature, then you need to solve the temperature problem with airflow. Otherwise, you will just burn the windings or pop the diode pack.

Are you sure the output drops because of temperature? Have you ruled out wiring and connections? If the regulator thinks the battery is full because it's voltage sense sees a higher voltage than the batteries are actually at, it will pull back charging.

Pull back due to temperature usually doesn't cut the alternator output off like you describe. It also brings it back when cooled, which you aren't seeing. On the other hand, if the regulator thinks your batteries have reached a high voltage, it will cut back permanently.

Mark

 

[colemj]

Here's a part I don't understand on your DC-DC to Lithium setup. How do you put in solar panels, for example? Are they also running through that same controller? What happens when you decide a couple years down the road to add in another solar panel? Or wind generator? Or hydro generator?

Pretty much all cases of using a DC-DC from start to LFP are due to an unsuitable alternator, or not wanting to make changes to an alternator charging circuit. There is no need for other charging sources to go through this route, as they all have their own regulators/controllers and their original designs have no issues with charging LFP.

DC-DC start to LFP is just a kludge used when the alternator isn't the primary source of charging the LFP, but having some charging is a bonus, and there is no perceived value in spending more money to directly charge. In these cases, it is a reasonable choice.

But I agree it isn't optimum. For some reason, there is a good component of the LFP community who are adamant that the alternator should never be pointed directly at LFP (and they will go on and on about all the 0.000001% edge problem cases this solves), and they recommend using stacks of multiple DC-DC chargers in parallel to get their HO alternator output to the LFP. It's nuts, but it's their boat.

Mark