Mixing lithium with AGM's - bashing around some ideas.

[booster]

Originally Posted by hbn7hj

You don't allow for the lithium battery to charge the SLA. For those of us that have a system like this that is a key feature.

I don't have such a system but was in on all work that hb7hj did on his, and the whole premise of having the combo setup was to be able to charge the AGM from the lithium because of the long charge times to totally fill the AGM to get max capacity and prevent walk down failure.


Basically, charge the lithium quickly off the genny or engine, shut off the charging source and finish the charging of the AGM off the lithium, then another quick run of the genny to refill the lithium.



You basically are able to cut you genny time to probable a quarter of the time or less.

[@Michael]

Thanks for the feedback. Much appreciated.

Originally Posted by hbn7hj

Don't forget a circuit breaker coming out of the lithium. If the low voltage cutout doesn't work as expected you can remove it.

I'll have a 60A fuse on the + of the lithium battery and another at the point where the SLA's tie into the rest of the circuit. The chassis batteries already have a 40A breaker in line.

The low voltage cutout is already in place as a part of the Coachmen factory design. It's also used as the coach battery disconnect via a push button on a user-facing control panel.

Originally Posted by hbn7hj

You don't allow for the lithium battery to charge the SLA. For those of us that have a system like this that is a key feature.

Hmmm....

I didn't forsee a situation where that would be useful.

Originally Posted by booster

Basically, charge the lithium quickly off the genny or engine, shut off the charging source and finish the charging of the AGM off the lithium, then another quick run of the genny to refill the lithium.

You basically are able to cut you genny time to probable a quarter of the time or less.

Now I do....

The A/B is really an A/B/Both, which (I think) would charge the SLA's from the higher voltage lithium, but at an unregulated rate. Otherwise I assume that I could have a cheap 10 or 15A B2B switched in line to accomplish that.

I assume that if set to 'both', any charge sources would be applied to both batteries at a relative current based on internal resistance or <??>.

If I use the 'both' position to tie the batteries together, I'll have to shut the input side of the Redarc B2B off. AFAIK, B2B's don't like having input and output jumpered.

I don't know at what relative rate the Redarc and the SLA's will draw current when any of the alternator, solar, or converter are charging. The Redarc defaults to a 13.2v threshold on input, so it should draw current any time any of the charge sources is charging. I'm not sure how much the SLA's will charge while the Redarc is drawing current, or which will charge first/faster.

Maybe have a fie extinguisher handy when I light this up.

[booster]

If and when you do this system, it will be very interesting to see how the Redarc does it's claimed "proven 100% full battery" charging from their algorithm. We have seen similar claims from other algorithm based chargers and they were not as accurate in charging as that. If they have managed to do a better job with it, that would be good for a lot of users because there currently are not any off the shelf products to control alternator charging.

[hbn7hj]

Don't know why but my attempt to charge both batteries from the alternator with the switch in the both position didn't work. The lithium didn't charge. Just a test and I don't use it that way but I expected it to work.

[@Michael]

Originally Posted by booster

If and when you do this system, it will be very interesting to see how the Redarc does it's claimed "proven 100% full battery" charging from their algorithm. We have seen similar claims from other algorithm based chargers and they were not as accurate in charging as that. If they have managed to do a better job with it, that would be good for a lot of users because there currently are not any off the shelf products to control alternator charging.

booster - The Battleborn shows up early next week. Watch the horizon in the direction of the SE corner of St Paul. If you see a mushroom cloud, you know I got it built.

If you are referring to managing a 'smart' alternator, AFAIK my Transit alternator isn't 'smart' like my Silverado was, where the alternator voltage drops way down while the truck is running and the start battery is charged. From what I see, the alternator voltage seems to be high any time the engine is running.

[booster]

Originally Posted by @Michael

booster - The Battleborn shows up early next week. Watch the horizon in the direction of the SE corner of St Paul. If you see a mushroom cloud, you know I got it built.

If you are referring to managing a 'smart' alternator, AFAIK my Transit alternator isn't 'smart' like my Silverado was, where the alternator voltage drops way down while the truck is running and the start battery is charged. From what I see, the alternator voltage seems to be high any time the engine is running.

We will be watching for the cloud


Not really as much toward the "smart" or dumb alternators, just in general with all charging off the engine, including standalone second alternators. Even if you add a B to B charger, or have a remote smart regulator, you are dependent on the algorithm in that control, and all I have seen and used do not do a very good job of charging. The "test" I use is to check the charging amps at the transition point, like a monitor would do to indicate a full charge based on amps at absorption voltage.

Long drives on systems without anything in between the alternator and batteries can overcharge them for long periods of time, and the smart regulators seem to tend toward undercharging.


On ours, with remote regulator but parallel dual alternators, the regulator did very poorly for charging curve but it had an input so I can lock it in absorption from and dash switch. I also put an ammeter on the dash to watch battery charging amps. I just lock it in absorption until the amps are correct for full, or just until the amps are low enough for the solar to finish off as it has shunt controlled charging and charges right on to full and not over.

[peteco]

Sorry if I missed this in all the discussion, but how does a lithium battery (~13.4v) with a battery-to-battery charger put out enough voltage (14.6+) to charge an AGM or FLA battery? Does the B2B boost the voltage somehow?

[booster]

Originally Posted by peteco

Sorry if I missed this in all the discussion, but how does a lithium battery (~13.4v) with a battery-to-battery charger put out enough voltage (14.6+) to charge an AGM or FLA battery? Does the B2B boost the voltage somehow?

Yes, and that is why it has to be there-good question. It will also limit current if needed so the lithium can't overcurrent the AGM or itself.

[@Michael]

Originally Posted by booster

We will be watching for the cloud

I installed the Battleborn/Redarc combination and lit it up. Nothing smoked, nothing popped, and no mushroom clouds.

Superficial testing indicates that it's working as expected, so far.

Here's the final wiring diagram - hot side:



The new stuff is in the lower left, all installed under the passenger side jump seat in place of what was a small storage cubbyhole next to the wheel well.

It's right above a Truma duct, so it'll stay warm in winter.

Photos of the install:







I need to make the wiring a bit neater and figure out how to fasten the battery better, and then hide the battery with a dark panel of some sort.

Thanks for the feedback hbn7hj, booster, marcopolo, kite_rider and others.

--Mike

[@Michael]

Originally Posted by booster

If and when you do this system, it will be very interesting to see how the Redarc does it's claimed "proven 100% full battery" charging from their algorithm.

Recent Changes:

I replaced the OEM flooded lead acid coach batteries with AGM's for various reasons.

I installed Battle Born's official and expensive battery heater.

I added a Thornwave solid state relay in line with the on/off switch on the source side of the Redarc B2B. The Thornwave battery monitor can now control the Redarc input, allowing me to programmatically disable/enable the lithium charger based on temperature, voltage, current or time.

I swapped out the Xantrax battery monitor that I had on the AGM's for another Thornwave. The way the coach is grounded, there is no place where you can insert a shut on the ground side that captures all current. Unlike the Xantrax, Thornwave's are happy to operate on the +12v side. So I now have a shunt on the +12v side of the AGM's and hopefully can get a good read on their state of charge.

Six month update:

I now have six months, fifty nights, and 9000 miles of experience with this AGM/Lithium parallel setup, and for the most part it works. The Redarc charges the lithium at a rate of 40A any time that it sees source voltage greater than its 13.2v threshold, and stops charging when either the lithium battery is full or the source voltage drops below its threshold - but with an important caveat described below.

As far as I can tell, the Redarc's lithium profile fully charges the lithium battery.

The Ford Transit heavy duty alternator easily supplies enough current to charge both house and lithium batteries. I'm quite sure that this particular alternator can safely supply much more current than what I'm allowing.

Issues

When the Redarc charges, it charges at 40A for a minimum cycle time of at least a minute regardless of source voltage. I.E. even though in theory it is supposed to cut off at 12.7v, in reality it keeps charging for some period of time after the source voltage drops below 12.7v. If I'm plugged in or driving, this is not a problem, as the alternator and converter can keep the system above 12.7.

However, when parked and on solar, the solar brings the source voltage up as it is charging the AGM batteries, and when the voltage is high enough (13.2v) the Redarc starts charging at 40A for whatever its minimum cycle time is. That has the effect of temporarily drawing the AGM battery bank down to about 11.9 volts, after which the Redarc stops charging and the solar attempts to recover the AGM's.

The net is that the AGM's get discharged at 40A every time the solar gets them partially recharged, after which the solar recharges them. The charge/discharge cycle occurs many times during the day. I'm not sure of the effect of this short duration charge/discharge cycle on the AGM's.

I'm experimenting with voltage thresholds on the SSR so that I can effectively favor either the AGM's or lithium batteries when charging from any of the charge sources. My assumption is that if I set a higher threshold than the 13.2v that the Redarc uses (I.E. a 14.x volt threshold) I can effectively force the solar to favor the AGM's over the lithiums. Likewise if I set a low voltage threshold with a short timer I can cut the Redarc off soon enough that it doesn't draw down the AGM's as far.

Battery Heater
The Battle Born battery heater works - sort of. As best as I can tell it cycles on at somewhere around 26F, but because it is only about a 15w heater, it has trouble keeping the battery warm in Minnesota's -24F overnight temps. To keep me out of trouble I ran the furnace overnight a few times last winter. I set up the newly installed SSR to disconnect the Redarc charger at 34F, and the Battle Born has built-in low temp disconnect, so in theory I don't have an issue until the battery gets below -4F.

It's too warm out now to experiment, so I'm leaving it as is until next fall.

[markopolo]

Not sure if this applies to your Redarc - can you use the Standard trigger instead of the Low Voltage trigger?

standard vs low.JPG


Edit: Or toggle between them?

[@Michael]

Originally Posted by markopolo

Not sure if this applies to your Redarc - can you use the Standard trigger instead of the Low Voltage trigger?

Attachment 8984

Edit: Or toggle between them?

That applies, and I have it set to the standard 12.7v threshold, but the footnote indicates a 100 second poll rate.

*1 Tested every 100 seconds

I think what's happening is that the AGM's drop below 12.7 fairly quickly, but the Redarc doesn't poll the input voltage until up to 100 seconds later. During that time, the AGM's are already down to +/-12.0 volts.

Edit: What I could do is to tie the ignition signal (Blue) of the Redarc to a 12v source and run it in low voltage mode, then use the SSR that I have inline ahead of the Redarc to manage the charging. Or tie the Redarc low voltage signal to the SSR and toggle between low voltage and standard mode based on source voltage.

[booster]

I haven't been watching all the threads and posts very closely, but it seems that battery and/or charger activation seems to be an issue when no 12v source visible to the charging equipment.



With a B to B charger from the alternator, it will normally lock out that source if no downstream battery online for reference, same with the shore charger and maybe solar controller, which really should never be unreferenced if in the sun.


Has anyone tried to just use a large relay around the B to B controlled by the starting battery circuit and an manual switch? That way you could get it to activate the coach stuff, even with the coach battery off referenced off the engine side, and then turn it off. As mentioned, not following closely so may be way off on this, but it may allow going without an AGM to activate things.

[markopolo]

Boosters idea of going around the Redarc is a possible solution here.

I think Michael only needs the Redarc to limit current when the engine is running. It might be useful on shore power as well.

Is it needed when off-engine and off-grid?

[@Michael]

Originally Posted by markopolo

Boosters idea of going around the Redarc is a possible solution here.

I think Michael only needs the Redarc to limit current when the engine is running. It might be useful on shore power as well.

Correct. Current limiting, and to make sure the alternator doesn't overcharge the lithium or run at too high of voltage.

The alternator has at least 120A of excess capacity at idle, and if I had heavier gauge wire from the chassis batteries back to the coach, I could charge at that rate. But that would be 1.2C.

Originally Posted by markopolo

Is it needed when off-engine and off-grid

No. My solar is only 200W & a PWM controller, so about 11A max. No current limiting needed. All I need is a compatible charge profile.

You guys have me thinking.....

One idea would be to simply connect the solar to the common side of the A/B switch ahead of the low voltage disconnect. Then the solar would charge whichever battery was connected to the load via the A/B. Or else have another A/B that switches the solar to whichever battery I want to charge.

This depends on the AGM profile on the solar being close enough to what Battle Born can tolerate so it can charge either. It's a GoPower PWM-30-SQ, with an AGM profile of 14.4v absorption/13.6V float.

[booster]

As I alluded to earlier, solar controller should always have a battery visible to their battery connection, unless there is a fool proof way to assure that the panels are never connected when there is no battery on the controller. The spikes from the uncontrolled panels will be in the 22v range for most panels which can do some damage. IMO, it probably should be wired to starting battery if the main batteries can disconnect themselves, or the panels will need to be relayed on and off appropriately. The benefit of going to the starting side of the B to B would be that the solar and starting battery could also be use to activate the lithium bank by jumping the B to B, as you mentioned, without the engine running.



IIRC correctly another member got some information from Battleborn that said the could accept double the current they list as max. I don't recall who it was though or even positive it wasn't another brand, but depending on the number of batteries 120 amps is not all that much current for even a couple of AGM batteries for a short duration. A single #2 would handle it, or even a single number four for a shorter time. My guess would be that you would only have to jump the B to B until the batteries came back online, which short be a very short time, as once connected they should still have plenty of charge to hold the charger on. I think you could also probably rig the jump around switch with another switch to turn down the field current to the alternator to lower the current output.

[hbn7hj]

What kind of money are we looking at for this system?

[booster]

The idea of somehow bypassing the B to B or probably any other charger to activate lithium batteries or charging sources that are on dead batteries comes up occasionally but I don't think it ever has been done or tried much.

Bouncing around the concerns of @Michael of wire size in particular would be nice to be able to address to make it simpler with less rewiring or over complication.

Since the B to B is there to limit current to coach anyway, an attempt to use it to limit the current during the reactivation may be possible.

The goal would be to get the B to B activated without connecting it directly to the starting battery parallel to the lithium bank in @Michaels case.

The B to B needs to see the battery voltage for sure, but I am not certain if it also checks that it is really a connection to a battery that would actually use some power and not just a voltage reference, so you may not be able to, or want to, just give it a voltage reference to starting battery that is not capable of current use.

I think that this product from Victron

https://www.amazon.com/Victron-Energ...078RD87KC?th=1

may be the answer to it all, if it will activate the B to B charger well. It is designed for activating alternators, so not certain it would work on the B to B, but it very likely it would. It would depend on how much current it would be using during activation maybe, as you supply a voltage to the B to B output, but may not accept current back out of the B to B if it checks for acceptance. I think it will because the battery is connected to the starting battery (for output only) when reactivating if wired the way I think would work.

The way I see it would be wired:

Output from B to B to the Victron input

Lithium pack to one output of the Victron

Starting battery to the other output of the Victron through a 40 amp switch (for a 40 amp B to B bigger or smaller for other sizes)

Wire from "run" circuit of engine to "energize" on the Victron with a blocking diode to prevent backfeed.

Small momentary switch to the "energize" contact on the Victron from the starting battery.

I think that is all that it would take. It would be used as follows.

* If the lithium pack is in low charge shutdown, you would turn on 40 amp switch to connect the output of the Victron to the starting battery.

* Start the engine

Power should go to the activate on the Victron which will put power from the starting battery to output of the B to B so it activates.

The B to B outputs will share the power while not being connected to each other as it is an isolator (FET style so not voltage drop). If the lithium takes some time to come on line, no big deal as the starting battery is there.

Once the lithium comes back on line just shut off the 40 amp switch so all the B to B output goes to the now charging lithium pack.

The momentary pushbutton would only be there for if you wanted to activate the B to B to get the lithium online without the engine running, so could be eliminated if you never would do that. That might be good for activating a shore shore charger assuming that the input of the B to B has a constant connection to the starting battery.

Probably about $125 in parts, so not bad if it works out well.

Opinions, ideas, holes in theory all welcome, as this is not something I think has been done that I know of.

[@Michael]

Originally Posted by hbn7hj

What kind of money are we looking at for this system?

Roughly:

• Battle Born 100AH: $1000
• Battle Born battery heater: $220
• Redarc 40A B2B = $420
• Thornwave Battery Monitors: $120 ea.
• Thornwave SSR: $40
• Switches, wires, shunts, connectors, gizmos, do-dads and beer: $200

So probably just over $2k.

There are B2B's that are pretty close in functionality at half the price of the Redarc, and much cheaper ways of heating a battery.

The solar, converter, battery isolation relay and low voltage disconnect came with the coach.

I replaced the 2 @100AH OEM flooded coach batteries with 2 @105AH AGM ($500) which I'd have done anyway.

[@Michael]

Thanks for the thoughts. Much appreciated.

RE: The Victron FET:

If I ever need to charge the lithium from the alternator at a rate greater than 40A and don't want to put another ($420) Redarc in parallel, A FET based isolator might be interesting. I'd have the input side attached directly to the chassis AGM staring batteries, an output directly on the lithium battery, and an on/off signal with some kind of voltage sensitive relay, so when the alternator raises the voltage above 'n' the FET is activated. I'd leave the alternator wiring alone, as it's likely using some kind of intelligent PWM signaling that's not well understood.

The isolation relay in between the chassis and coach batteries is a Precision Circuits BIM-160, which seems to be intelligent enough that I see no reason to mess with it.

The low voltage disconnect protects the coach batteries, so I have no plans for recovering for a situation where either or both are dead.

Likewise, I'm not going to worry about having a completely dead chassis battery. If that happens, and if my coach batteries are also dead and my generator doesn't work, I'll call Coachnet.

Here's current state:



It works, with the caveat that when the solar sees sun and the lithium is not full, I have the 40A on/off cycling every few minutes. If no sun, or if the lithium is full, or if I trigger the SSR, there is no cycling. If I don't care about the cycling, the current design works.

Simply moving the solar to the load side would mean that the solar would charge whichever battery bank is selected via the A/B, and the load would be from the same bank. If I select the AGM bank, the Solar will charge the house AGM's directly, the chassis AGM's via the isolation relay, and the lithium via the Redarc. Load will be drawn from the AGM. It'll be essentially as it is today. OTOH If I select the lithium bank, neither AGM bank will get charged, lithium will be charged via solar, and load will be drawn from lithium/solar.



This would be an easy 5 min change - the solar output is already one post and 3" away from a where it would need to be.

Some flexibility would be gained by adding an A/B to separate solar charge from load.



If I think that the solar controller must see a battery at all times I'd make sure the A/B's are never set to 'off'. I'm not sure that is does - the docs are ambiguous and I've disconnected it many times without incident. I also am careful not to use 'Both' for more than the second it takes to go from 'A' to 'B', especially if either bank is significantly discharged. I only have 60A fuses and 6AWG wire.

If I want to shut off the load, I already have the low voltage disconnect that has a remote shutoff, and as placed would not disconnect the solar from the batteries.

As far as the cycling, I'm only concerned if the cycling shorten the life of the AGM batteries, or if the inefficient-ness of charge/discharge on lead acid is high enough that I'm wasting solar.