Parallel lead acid and lithium circa 2018

[markopolo]

Booster - I was looking at this site: https://www.solacity.com/how-to-keep...tteries-happy/ - for the discharge curve and SOC chart. That info is approx 2/3 down the page.

[booster]

Thanks Marko, just what I was looking for.


Discharge curve: Seems to show that the AGM will not start to discharge until the lithium is at about 15% (12.7v) so that would be their priority discharging of the lithium







Charge curve: The charge curve seems to show that the lithium will not charge much past 10% until the AGM is nearing the bulk to absorption transition point. The voltage of AGM charging seems to go from 13.6 to 14.3v pretty quickly, so they probably would be approaching 70% full. This would pretty much fullfill their priority charging for the AGM, I think.








If all this is right, you would not need any fancy controls to do what the literature claims for the system.


This is very similar to what we had guessed in the initial and later discussions on a hybrid system that culuminated with Harry building his. Our biggest issues were the limiting of the charge and discharge amps from the lithium, which in the system likely done with the BMS, and the fact that while the lithium will finish off the the AGMs, sort of, the charge voltage is going to be too low to do a good job except at the very beginning. After that it is basically charging at float voltage, which won't take care of sulfate fixing and maintain capacity. Better than leaving it at 70%, but not as good as a full charge would be.

[peteco]

Originally Posted by booster

Charge curve: The charge curve seems to show that the lithium will not charge much past 10% until the AGM is nearing the bulk to absorption transition point. The voltage of AGM charging seems to go from 13.6 to 14.3v pretty quickly, so they probably would be approaching 70% full. This would pretty much fullfill their priority charging for the AGM, I think.

Why isn't the priority to charge the lithium first, as that gives you access to the most power? Then charge your reserve AGM or LA batteries.

[booster]

Originally Posted by peteco

Why isn't the priority to charge the lithium first, as that gives you access to the most power? Then charge your reserve AGM or LA batteries.

Yes, if you are going to charge just one first like they indicated they do, and called it an advantage to charge the one first. As Marko and I both indicated earlier, it would make more sense, as you say, to charge the lithium first if you had the choice.


Since then, and looking at the charge and discharge curves it looks like they probably don't do anything special at all and charge both at the same time. Physics takes care of all their "priority charging and discharging" which makes the AGM charge mostly first and the lithium to discharge mostly first. What they did, most likely, is just make that fact appear to be intentional and an advantage by claiming it was something revolutionary.

[markopolo]

Booster's observation looks to be spot on.

My understanding:

When charging, you bring the system voltage up by supplying current at a voltage that overcomes the internal resistance of the batteries. When charging the combo AGM/LFP bank from a deeply discharged state like 12.0V OCV the AGM has to accept the charge current to increase in voltage and therefor also cause the system voltage rise to where the LFP battery begins to charge. The competition for charge current between the two probably begins at around 13.6V.

I'm not sure exactly how the current will be shared from there. The LFP would want to rapidly charge but it will probably be restrained because the system voltage also has to rise and the rate of that rise is tempered by the AGM acceptance rate. When the AGM begins to taper the LFP will happily and rapidly absorb the surplus current.

[markopolo]

Now I'm thinking an A, B, Both, Make before break, switch should be in the system. The main reason would be to disconnect the LFP battery to prevent below freezing charging. Heating pads are always an option of course but they could fail without notice.

[hbn7hj]

As soon as I get my system back I'll run some tests. Up to now I haven't paused at the both setting but switched through it. Pretty easy to let it sit there, run the sat TV and see what happens. I'll do the same thing while charging.

I expect a switchable system with separate chargers to be better but we will see how the other one works with nothing but battery physics controlling things.
Harry

[booster]

Originally Posted by hbn7hj

As soon as I get my system back I'll run some tests. Up to now I haven't paused at the both setting but switched through it. Pretty easy to let it sit there, run the sat TV and see what happens. I'll do the same thing while charging.

I expect a switchable system with separate chargers to be better but we will see how the other one works with nothing but battery physics controlling things.
Harry

The one unknown that might bite you is that a big part of the reason to have the B to B charger is to limit current from the lithiums to a low AGM bank. You might wind up with way high current and blow fuses or melt something. The Hymer system seems to indicate they limit current out of the lithium battery to prevent that issue, but we don't know that for sure at this point.

[Luv2Go]

The system in our Roadtrek has the normal 2 - 6V flooded golf cart batteries in series (not sure what capacity, maybe 220A/H) and 320A/H of LFP. At first I had them separated with separate chargers but the lead acid charger failed (and the replacement failed, both because they overheated in the compartment RT puts them in in my year 210V) so I've had them connected in parallel for several months now. They are charged from the alternator or the inverter/charger (we have 400W of portable solar too, but the portable panels have not seen much use and I don't really want to put panels on the roof of this coach).

They operate pretty much as you all have posted, I've verified that the LFP discharges first, but I haven't discharged the LFP to the point where the lead acid has to discharge substantially. I have Thornwave battery monitors on both batteries feeding a Rpi with Domoticz on it to collect the data, I'll post the results when I do discharge them, maybe sometime this winter.

[kite_rider]

Originally Posted by booster

The one unknown that might bite you is that a big part of the reason to have the B to B charger is to limit current from the lithiums to a low AGM bank. You might wind up with way high current and blow fuses or melt something. The Hymer system seems to indicate they limit current out of the lithium battery to prevent that issue, but we don't know that for sure at this point.

This has been a great thread and I'm now seriously considering adding a LFP battery to my existing AGM/solar/alternator setup. On the point mentioned above; if the hybrid battery is directly connected, I don't see how you would get a very high current from the lithiums to the low AGM bank. They would both go low together, right?

What situation would allow the AGM to get low while the lithium remains full of energy? Connecting them together let's the lithiums do most of the cycling by draining them first).

I'm thinking I might have an isolation cutout relay on the lithium battery for the case where the battery temp goes to 0C (or lower), but otherwise it would just be connected to the existing system in parallel. A basic BMS on the lithium battery would protect it from overcharge and overdischarge. Any major 'gotcha's' with this proposed setup?

[booster]

Originally Posted by kite_rider

This has been a great thread and I'm now seriously considering adding a LFP battery to my existing AGM/solar/alternator setup. On the point mentioned above; if the hybrid battery is directly connected, I don't see how you would get a very high current from the lithiums to the low AGM bank. They would both go low together, right?

What situation would allow the AGM to get low while the lithium remains full of energy? Connecting them together let's the lithiums do most of the cycling by draining them first).

I'm thinking I might have an isolation cutout relay on the lithium battery for the case where the battery temp goes to 0C (or lower), but otherwise it would just be connected to the existing system in parallel. A basic BMS on the lithium battery would protect it from overcharge and overdischarge. Any major 'gotcha's' with this proposed setup?

I think that the largest chance that you could get a high current out of th lithium would be if both batteries were very low when the charge started. It is likely that the low internal resistance lithium would take most of the the charge for at least a while, depending on charger size, I think. Even if both batteries got to 50% full and the charger was shut off, you might get the high current situation from the lithium to the AGM. At 50% the AGM will still be able to take it's full acceptance of 1C or more, if the lithium would supply it, which is probably possible .


This certainly is a what if at this point, but I think a real possibility. It would likely be very dependent on both banks sizes and the actual max and safe currents that each can handle. Charger size would definitely be a factor, I think, as a big charger could fill hold voltage higher and put faster charge into the lowest internal resistance, which would be the lithium.


The fact that the Hymer system has pretty low current limits in and out of the lithium battery could be an indication that it is necessary, maybe for this reason.


The charge curves listed earlier probably won't tell you much about it all as the curves look to have been run with a current limited charger at .5C. We would need to see information that showed acceptance for both battery banks at the various voltages they would see to know how the SOC curves would look like during charges of different depths.

[hbn7hj]

Originally Posted by kite_rider

I'm thinking I might have an isolation cutout relay on the lithium battery for the case where the battery temp goes to 0C (or lower), but otherwise it would just be connected to the existing system in parallel. A basic BMS on the lithium battery would protect it from overcharge and overdischarge. Any major 'gotcha's' with this proposed setup?

Probably better to put in an A, B, Both marine high current battery switch to retain configuration flexibility along with two battery monitors to see what is happening. You need to be able to handle a failure in either battery.

[kite_rider]

Good suggestion Harry on the A/B/Both switch; agreed that it would make human management much easier. However, I'd like to design this so it could be in the 'both' position all the time without much worry...

Booster - what products would you use to limit the in/out current from the LFP? I've been looking around and found the Victron 12/1200 as a possibility. It seems perfect, except for the part where it tells you to isolate the ground on the LB side. I think I would need a system where the loads are all connected to the chassis ground for the hybrid system to work. So I'm sort of stumped on this...

[booster]

Originally Posted by kite_rider

Good suggestion Harry on the A/B/Both switch; agreed that it would make human management much easier. However, I'd like to design this so it could be in the 'both' position all the time without much worry...

Booster - what products would you use to limit the in/out current from the LFP? I've been looking around and found the Victron 12/1200 as a possibility. It seems perfect, except for the part where it tells you to isolate the ground on the LB side. I think I would need a system where the loads are all connected to the chassis ground for the hybrid system to work. So I'm sort of stumped on this...

The current limiting was what, at first, was our hangup in trying to figure out how to put together a system when Harry was looking at it. We couldn't find anything that would limit current except the DC-DC charger like Harry used. At that time, we were looking at a permanently connected setup exclusively, and where looking a step further than the Hymer system in that we wanted to completely finish the charging of the AGM battery from the lithium. At 13.6v out of the lithium, for most of it's discharge, that is not enough to fill and AGM, which needs closer to 14.4v to top off, and lots of time at that voltage.


The original way I looked at the system would be to have the batteries parallel when charging, with the lithium shutting off when full. The AGM would continue to charge as long as there was power source to do it. Once the power source was gone, the lithium would top off the AGM through the DC-DC charger. If wanted, the lithium could then be quickly filled with the charge source and all would be full, or the van could run on the AGM, lithium, or both with a 3 way switch.



The main reason to do the system at the time was for the generator recharge crowd, be it Onan or alternator, to limit generator run time which would be many hours if you try to completely fill and AGM (8+hours). By doing short generator runs to quickly fill the AGM, lots of generator time would be saved as the lithium would take care of the longer time, lower output, charging needed to complete the charge in the AGM.


The parts int he Victron layout don't have part numbers on them, and I couldn't find them on their site by appearance, so I can't really tell what does what. The charger looks like their normal charger, there is a BMS and what looks like an isolator that is more lithium specific with some kind of manual shutoff and lithium control connection. Hard to tell where any current limiting would be for me.

[hbn7hj]

I have yet to see over 40 amps going into the lead acid batteries for any length of time so it may be self limiting. We will test it for you soon. My lead acid batteries are 7 years old so new ones may allow higher current.

You will need to be able to turn it off so a switch between lead acid and lithium would be important.

My guess is the lead acid batteries will become superfluous as they won't provide power till the lithiums are nearly discharged.

[booster]

Are your lead acid wet cells or AGMs? I don't recall. If they are wet cells they will accept much lower amps than AGMs of the same size will, so more likely to be less max from the lithium. Of course the wet cell can't handle as much either so they still could be hurt.

[hbn7hj]

I use lead acid wet cell batteries.

[kite_rider]

Ah, I should describe the existing hardware, batteries, and proposed LFP.

I'll be doing most of my charging from the upgraded alternator that came with the Ram Promaster - it's the 220A model. My camping style usually has me driving somewhere each day and I'm fine with firing up the engine if I need an occasional power boost.

The shore power charger is the common Hymer installed Chinese made 2000W inverter / charger that ships in a lot of their products (Axion, Banff, Zion, Active, etc) as well as many Roadtreks I think... This has a programmable charging setting that has a pre-set AGM/Lithium mode. As it is, I rarely connect to shore power and expect to do this even less with the additional LPF battery.

There is also a factory installed MPPT solar charge controller and 200W of installed solar on the roof.

The house battery is a 105Ah AGM and the starting battery is also AGM and I think it's about 100Ah as well. The two are currently isolated with whatever isolator is pre-installed on the Axion. It's pretty beefy and there are large cables pre-installed; probably designed for the E Trek lithium battery option. The LFP that will be based on 60Ah cells in either a 2P or 3P and 4S configuration - still trying to decide between 120 and 180Ah of additional lithium for the system. The BMS will be determined based on the final design - but looking for simplicity!

No generator in the picture.

[kite_rider]

Originally Posted by booster

The parts int he Victron layout don't have part numbers on them, and I couldn't find them on their site by appearance, so I can't really tell what does what. The charger looks like their normal charger, there is a BMS and what looks like an isolator that is more lithium specific with some kind of manual shutoff and lithium control connection. Hard to tell where any current limiting would be for me.

Sorry about that. I was only interested in the "Victron BMS 12/200" since it has a current limiting function (based on a fuse) and can also be used to switch off the lithium batteries when they are full. I attached a mini-datasheet for this part with that has some specs.

The diagram also says that you need to use a "Victron isolated Orion DC-DC converter" for DC loads that are connected the vehicle chassis. This is where the system falls apart since all of the loads I want to run will be grounded to the chassis..

[booster]

Originally Posted by kite_rider

Sorry about that. I was only interested in the "Victron BMS 12/200" since it has a current limiting function (based on a fuse) and can also be used to switch off the lithium batteries when they are full. I attached a mini-datasheet for this part with that has some specs.

The diagram also says that you need to use a "Victron isolated Orion DC-DC converter" for DC loads that are connected the vehicle chassis. This is where the system falls apart since all of the loads I want to run will be grounded to the chassis..

The Victron descriptions are a little bit confusing to me, but nothing new about that . Victron tends toward very, very, integrated systems consisting of their own (lots of times private label parts) products, so using single items out of their product line can sometimes get to be tough, both from the reliance it may have on other components and also in trying to duplicate any control or data sharing that it normally might have in a system.


The text description mentions lots of different "isolated" comments, including the one for the 12v to 12v charger that you mentioned. It also says as you say it is needed for the stuff that is currently connected to the chassis.


I think the "phraseology" may be getting in the way of the just taking the wiring diagram at face value.


* Everything in the diagram, including the alternator, a connection on the isolator, the chassis grounded components are marked as "chassis ground"


* The only thing that is "reverse isolated", for a made up term, would be the batteries, the charger, and the LB terminal on the Isolator which also connects to the INPUT of the 12v to 12v charger.


All this is doing is taking the batteries and main charger off of the chassis it appeas, so not a hard rewiring, I think to do that. The starter battery even stays on the chassis side.


* The positive stays the same as normal, except that the chassis based loads connect to 12v to 12v charger output.


That just leaves the positive and negative connections that are marked as isolated for the from the van loads and grounds related to the chassis, marked as "other DC loads"


* I would assume that would be for any big power users in the system like a large inverter. All that would mean is that the input to the inverter would connect to the batteries directly, as would the negative, and not be connected to the van loads that are using the chassis as ground.


If I am interpreting things correctly, I don't think you would be looking at any amount of huge wiring changes, depending on where the current components are located, of course.