Mixing lithium & AGM batteries

Hi all. Well I went and did it. I jumped into upgrading my 10-year-old house AGM bank to lithiums. But I didn’t quite realize what I was getting into...typical.

I like the advantages of lithium. Because they can be deeply discharged, a much smaller nominal bank can provide the same amount of usable energy as a much bigger lead acid bank. They can also absorb charge faster and more efficiently, so I don’t have to run as long to recharge my bank. They’re much lighter which has performance benefits for my catamaran, and makes them a lot easier to muscle into place and secure. They also will last far longer, probably longer than I will. They’re certainly more expensive, but prices have come down, so considering the advantages above the costs are comparable.

So I’m installing a lithium house bank but retaining the AGM start battery for now. The house bank is charged by the alternators and the start bank is charged through a Blue Sea combiner. The house bank lithiums are 300Ah.

I have two concerns. First is that because lithiums can charge at a much higher rate, they work the alternator much harder. I’m thinking that with only 300Ah and two fairly high output alternators this shouldn’t be a major deal.

The second concern is that the lithiums have an internal management system that controls charging. Once the battery is full it shuts off. If that happens while the alternators are running, the sudden loss of load will trash the alternator electronics. There are sophisticated battery/charger systems from the likes of Victron and Balmar that overcome this, but I don’t have those expensive systems (yet). So my plan is to keep one of my AGM house batteries in place, mixing it with the lithiums. When the lithiums shut off, the AGM will still be there to absorb some load. The lithiums fill up much faster than the AGM so I’m assuming that the AGM will have lots of capacity remaining to be charged. I know the cardinal rule against mixing battery types, but I think the battery that will suffer is the AGM, and it’s near the end of its life, so I consider it disposable.

Any battery gurus out there who have lithium experience and can advise? Thanks for reading this long-winded post.

 

I don't know what kind of management the batteries you bought have. However, the cheapest solution would be to have a small light or other small load so the alternator would never be without a load. I suppose that having the instrument lights on all the time would work.

 

First at all, a) is your AGM parallel to your Lithium batteries? b) is it 12.8-13.2 Volt? c) are they LiFeP04 batteries?
I have for about 8 years my 40 Lithium batteries parallel and in serial, giving me between 4 x 10 cells parallel i.e. 4 x 3.2 Volt a good service for all those years. I charge my lithium battery with a normal 14 Volt charger, but just keep an eye on each 10 x parallel that they stay within the 14.4 dived by 4 i.e. None of the group of 10 is allowed to go over the 3.65 Volt. So far it has never happened. Yes other battery chemistry may need balancing due to the fact that they discharge the battery with very high currents, like hobby airplanes which I have avoided. I learned that trick 8 years ago from an Englishman. I have no idea what batteries you have and what the internal connections are, thus you may be able to do the same or maybe not. I have time and charge the batteries at 1C , while you may not have the time to do that. Charging the batteries with very high currents, promote easier out of balance of your batteries and then you need a battery management system to do the correction. . Bert

 

I do not think your alternator works harder. If they are maximum 50 Ampere, then it depends on the battery voltage whether you get 30 Ampere charged or that you go over the maximum 50 Ampere. That can only happen when your Lithium battery is gone down so low in Voltage, that the difference in voltage will course the alternator to be overloaded. What kind of regulator do you have on the alternator? Bert

 

Hi Deering, Sorry, I overlooked the real problem and just gave some nonsense, my apology for that . I use 2 Schottky diodes, one to the lithium battery and 1 to the Deep charging AGM. They are only 70 Ampere/30 Volt each. This in your case, it could solve the problem, if you take a diode with higher current. It prevent the AGM to discharge into the Lithium batteries and the Lithium battery feeding the AGM. Thus in your case if the internal management shuts the Li batteries down, the load of the AGM will still prevent your concern most likely. But it totally depends on the voltage regulator on the alternator. In the olden types, in microseconds it shut the the output down, if all batteries are fully charged. Bert

 

Thanks Bertku. I think I follow you. If the AGM is ‘full’ will it reject any current from the alternator in the event that the lithiums shut down? Or will it always provide some place for the current to go in the short term (milliseconds I assume)?

 

I’m guessing that by the time that the lithiums are close to full the amperage feeding them would be small, so if the lithiums shut down suddenly it wouldn’t be that large of a drop relative to the other loads. My engines are fed from the alternators and they always require some power for the fuel injection system and control unit. I almost always have my nav instruments running, and they take about 100 watts.

I admit my ignorance with alternators. Maybe I really don’t have a problem?

 

A good voltage regulator always shuts down the main source of the alternator and therefore I dot think you should be worried. That is the principle of the voltage regulator, The only problem you must check, if the AGM is 14.40 it shuts down (szghould be, but check) , but if the AGM is 14 volt, it will still charge and 14.4 divided by 4 = 3.6 and it very close to the maximum voltage a LiFeP04 is allowed during charging. i.e 3.65 volt. Any imbalance could make one set exceeding the limit for charging a LiFeP04. Let me think about it in how to solve that problem. Also could you check what the current is at 14.4 Volt when charging the AGM, if only a few mili Ampere it answer your question. even the LiFeP04 can handle that. If more please let me know. Bert

 

These are Renogy batteries. Their literature says that their max voltage is 14.4 and their management system shuts them down at 14.6. With an AGM in parallel I haven’t seen the voltage from my Volvo internally regulated alternators exceed 13.5. If the AGM wasn’t present would the alternator voltage go higher?

Will the lithiums shut down when they’re full even if the voltage never exceeds 14V?

In looking at the Victron battery systems, they have external control systems that can communicate with external alternator regulators, such as the newer Balmar systems, to warn the Balmars in advance that they’re on the verge of shutting down. My batteries and alternators have nothing so sophisticated.

 

Let us both design a voltage regulator, don't laugh, it is a good way to understand what a voltage regulator will do. But let me first answer your question. I don't know your type of battery with the internal management system. A good system charges high currents, compare various voltage from sections and then balance them out buy connecting too high voltage with those of too low voltage and the last up to 1 hour it is voltage regulating at very low currents like 100 to 200 mA maximum. i.e. it keeps the voltage constant at 13.6 or 14 Volt.
But we are going together design a voltage regulator like we did 50 years ago with relays. Thereafter we design one with software system, just not real, but let us come to an agreement what it should do.
1) we take a car.
2) we switch the ignition on, the voltage regulator should check the battery voltage. agreed?
3) then we start the car and the battery collapses to a very low voltage and high currents. agreed so far?
and the engine starts.
4) at this moment the voltage regulator should kick in, by closing the relay whereby the alternator is connected to the battery. Agreed? or do you think the relay should come in immediately when the ignition switches on whereby the alternator is helping the battery. I doubt it, but yes maybe.
5) We measure with a thermistor the temperature and that we use to regulate the output voltage for 14.3, 4 or 14,5 Volt. happy?
5) At 13.8 volt a lead acid battery, any type AGM, sealed , liquid battery will start gassing by high currents and a little bit by low currents
6) we should scale down the current from the alternator by pulling a relay in which places a resistor in series with the battery to limit the current. Remember we are 50 years younger. Happy with that?
7) Now the voltage regulator which was giving 14.,4 or 14.5 Volt depending on the temperature at high currents, drops down to 13.6 Volt.
8) until the driver put his lights on at full blast, the battery voltage drops and the relay pulls in to give 14.4 Volt again and high currents to help the battery to maintain more than 12,5 Volt battery voltage.

Deering do you think the automotive industry will buy from us? Tomorrow we do a modern micro processor one.
Bert

 

Ok, I’m game.

4. It makes no sense to have the alternator start producing current until the engine is started. Otherwise you’re using the battery to power the alternator...to power the battery. Perpetual motion machines only works in politics and religion.

5. By thermistor you mean the temperature of the alternator windings?

6. I assume that the selection of the resistor is based on the type of battery. So a different resistor would be chosen for a lithium to achieve ~14V?

I’m not clear about battery resistance. Does it rise as the battery gets closer to being ‘full’? How is full defined anyway, or is that an inappropriate term in this case?

I am certain that the industry would buy from us. I’m eagerly awaiting the direct deposit to my bank account.

 

I agree with you. on point 4, because the revolution during starting of the engine is so low, that the engine and therefore the starting motor need more energy to turn the crankshaft.
5) The thermistor must sense the temperature of the battery. At -10 degrees Celsius, the battery has a lower output voltage than at ambient temperature of 25 degrees Celsius, thus one must compensate for that condition.
6) No, in principle the Lithium battery has also similar cutoff voltages then the Lead Acid battery. You can charge it at a constant voltage of 3.65 Volt per cell i.e 14.4 Volt and preferable constant same current, not being tapered. But that does not mean that I found great problems with charging my LiFeP04 with a normal charger over those 9 years I have them on my boat. But I do the cut off at 13.6 Volt and then only have a small constant current of 30 mA for some time, which has varied from 1/2 hr to 5 hours, until pulling the mains plug. I do not care whether my Li batteries are charged to 97% or 100-%. Remember 13.6 Volt means 3.4 Volt per cell. That is good enough for me. I am in a realistic user environment and not in a laboratory were 1% means the world. I prefer also to have my 9 year old set of AGM batteries to be charged till 13.6 Volt and not higher. Bert

 

Hi Deering, I have my doubt that the automotive industry will buy our, first at all they only select something new for their next 7 year model run. The entry into that industry is a very long one and a very costly one. Thus we just try by designing our own device what should be the best for your situation, i..e. having your Lead acid battery parallel receiving a charge with the Lithium batteries. So let us design now one with a microprocessor.

1) You switch your diesel or gas(petrol) on and the generator starts feeding batteries.

2) The difference between charging the lead acid battery and Li batteries is, that the lead acid battery has a gassing point (13.8 Volt at 5 degree Celsius) and the Li not.

3) Both batteries can be charged with a 14.4 Volt charge

4) Personally I don't like to charge a near full battery AGM battery with 14.4 Volt, but stop at 13,6 to 13.8 Volt gassing point charging, that maybe the reason also that all my AGM's are lasting all for 9 years already. I could not give a hood that it is charged to 85% instead of 100 % the same for the lithium batteries.

the problem is you have a 300 Ah lithium battery bank and only maybe a 100Ah AGM

5) Thus we design now an alternator voltage regulator with a microprocessor. 40 Mhz Microchip PIC18F2431 with a 40 Mhz crystal. although 4 Mhz is good enough

6) we start with specifying the settings. I want bore you with that.

7) We first have a routine which senses the temperature of the battery and check a table which has some Lead Acid battery voltage reference at those temperatures. (refer the graph further on. )

8) We now start the generator

9) we measure and senses the Voltage of the batteries, Lithium separate from the AGM.

10) both lower than 12,0 Volt? start charging at 14,4 Volt both batteries via the 2 Shottky diodes of 70 Ampere/30 Volt

11) One high 13 Volt and the other lower than 12 Volt? . So what 14.4 Volt minus the 0.5 Volt over the Shottky diode means in anyway only charging to 13.9 Volt which is 0.1 Volt higher than the gassing voltage at 5 degrees Celsius and if the temperature of the Shottky diodes goes higher it compensate for the lower gassing voltage at a higher temperature. refer chart here under.

12)Thus I stop designing a new Voltage regulator as you now have understood why I have those 2 Shottky diodes between the batteries and are not worried if the batteries are nearly full and not 100% but maybe 96 or 97% so it lengthen the lifetime of all batteries.

Deering do you mind if I stop now, as you probably have understood, that you now can sleep without worries, but please buy 2 Shottky diodes. Bear in mind that if a battery is full, it will not absorb more current, it stops and trickle charge, provided the voltages of both are the same, battery and charging. I still have to finish a design with the PIC18F2431 for somebody. Bert

 

Deering I forgot to mention, if two voltages are the same, there is no current flow between them. Thus if our battery is fully charged and the generator is still going and also has the same voltage as your battery, nothing will happen. But it is nice to understand a voltage regulator by: go into your car, switch the ignition on, but do not start the car. Now your switch your lights on at full blast. Now you start your engine and suddenly your lights are brighter. You know now that your regulator and your alternator is working fine. Something like that you could also put over your batteries and temporarily switch and then start the generator, you will see how well the batteries are being charged. If both batteries are full and you put those bright heavy lights over the batteries for a short time, just to see whether your regulator is working. The difference in light will tell you whether the batteries are still being charged or no longer.

 

Hi Deering, I see that you are living in Alaska. If you do have a problem in getting 2 Schottky diodes of 70 - 80 Ampere, please give me a private message with a poste restante address (to keep you address confidential) in Alaska and I will see whether I can post you 2 used 40CPQ100 which each, as I have understood 2 diodes of each 40 Ampere (parallel thus 80 Ampere) and max 2750 Ampere for 5uS Seconds. The only problem is that the RSA postal agency is having wages negotiations and sometimes we have a delay in the post, please bear with me in that case.
You must see it my way, If you buy a new car and the speedometer tells you that you can drive the car to 200 miles per hour, how long does it take before the car is giving trouble, now you only drive around the 100 miles per hour and you have very little problems. I see it, the same with charging lithium batteries, by charging and discharging maximum allowable currents, the cells go easier out of balance and you need a good balancing system in place, while when you use 1/2 of the maximum currents, you have very little imbalance. Bert