Adding 3rd battery, LiFePO4, to AGM dual battery setup

[Xterabl]

See existing solar/dual-batt/electrical rough layout in picture, with proposed addition in yellow shaded region (LiFePO4 system).
Having MPPT in parallel with (1) dual-AGM (separated by diode) system and (2) DC-DC charger makes me wonder about conflicts.
Will MPPT (being set to AGM profile) "look at" the AGM battery(ies) to decide charge profile, or will the addition of DC-DC charger "confuse" the MPPT and output will be unpredictable?
Expert opinions appreciated.
(Background: vehicle is 2000 Xterra that I've recently re-dedicated myself to...new MT transmission, ARB air locker installed...currently rolling at 250k miles and hoping to get 400k or who knows???)

 

[Dave in AZ]

That all looks ok to me, particularly if its all working and just the yellow dcdc lifepo4 is new. Limit the dcdc to around 30A max, such that full car power (engine, fans, air-conditioning, wipers, hibeams, turn signal, radio) plus the dcdc totals to 70A or so. Want to keep alternator demand to 60% of max or so to keep its temps down and not burn out.

My Tacoma 130A alternator with all that listed stuff pulls 45A or so. Here is a thread on that test you may like:

Info only: 2021 Tacoma amp draw and alternator loads tested

This thread is just for reference for those looking to use excess alternator capacity to charge a battery. Other cars should have similar loads to...

www.tacomaworld.com

 

[DaveInDenver]

There's a question with the diode drop to the starting battery. What are you thinking for an ideal diode? One of those N-channel FET controllers? The question I would have is current capability vs cost.

As you know, any drop may cause problems with the regulator in the alternator, which will compensate for it if the voltage sense is from that battery but potentially overcharge the dual use in your diagram.

Alternatively when using the solar charger you may undercharge the starting battery for similar reasons, the dual use being the one it sees will determine what it does as it steps through it's constant current to constant voltage to float stages.

A good MOSFET isolator would work, though for someone thinking about it the trade-off is it may be preferred (or just cheaper) to use a mechanical isolator such as an ML-ACR or similar relay or contactor. The advantage to an ideal diode is no moving parts.

Another option may be to isolate one of the AGM with it's own DC-DC charger instead of trying to parallel the AGM batteries. The starting/factory one being probably easier. Switch the alternator voltage sense to the dual use and let the DC-DC handle topping up the starter, which shouldn't be seriously stressed most of the time. This would have the benefit that the two AGMs could be completely different types and ages, too.

As far as the DC-DC, it probably won't care as long as the upstream chargers stay within it's parameters. You may run out of current to supply them all. The alternator will probably work fine but as you mention the Morningstar may not operate reliably.

It's hard to predict that without testing them together. Sometimes two switching supplies will work fine together, sometimes one or both can get pushed into an odd state.

It's possible the battery/batteries serve to keep the solar control loop happy insofar as it runs the AGM profile right.

What you have to watch is that it settles to and stays in float with sufficient current available when it senses a load. If it doesn't it may want to drop back to bulk->absorption in a never ending cycle, which would be hard on the AGM. It would basically think the added load is being asked of the AGM it's charging, when it would not in fact be putting any load on the AGM at all.

It's also possible the DC-DC consumes most of the current during bulk such that the AGM(s) never reach absorption voltage. In effect if the apparent input impedance of the DC-DC is low enough the AGM will only trickle charge.

Until you run it you may not be able to say for sure.

I'd plan for the eventuality that the DC-DC may need to be isolated from the solar at least with a relay or switch ahead of it. At least have that option the first few times using it so that you can put things in a state to keep from being unable to start the engine or something. You could in a pinch run the fridge by bringing a long 12V extension cord and use the dual use AGM to get home if the lithium path turns out to go weird.

 

[Xterabl]

Thanks to the two Daves for the comments/feedback. AZ Dave that data is interesting, I've always kind of wondered. Especially, on a recent remote outing I fried my alternator and "big" ("house") AGM battery and had to limp home 6 hours with just the 35AH AGM and an alternator that wouldn't charge, above some small loading. Basically, when driving, my voltage monitor would show ~13.5V; but if I turned on my headlights or air conditioner it would drop to 12.5V (draining battery) after 10-20 seconds. It made me ponder what were the actual battery drains (bare minimum) for daytime driving...I came up: with fuel pump, spark, brake lights, and dash instruments. Anyways, I made it home safely--even enjoyed a route over the Sierra-Nevada mountains that I had never been on before--but was definitely sweating bullets for the first hour or two.
Dave Colorado: The "ideal diode" I use for the two AGM batteries is Perfect Switch PowerGate single rectifier, found here: https://perfectswitch.com/isolators/single-rectifier-isolator/
I've been using this product for 14 years and am very happy with it as well as the customer support. I went 10+ years without having to talk to them, and when after the above-described adventure (fried big battery/alternator) I had to place a call to get a replacement unit the same founder/owner/operator was there for me, spending time on the phone and getting me back up and running as quickly as possible. He is also an expedition enthusiast which makes everything pleasant even when plunking down a decent chunk of change. My experience is: money well spent.
Also, something relevant I forgot to mention: I only use the solar when parked, not driving (it's a 2-panel custom clam shell folding desing), so there will be no alternator in-play whe using solar.
With this new LiFePO4 addition to the setup, I was hoping to minimizing additional switches but I'm tending to agree with you that that's probably the best thing to do. I ordered some heavy duty 3-way toggle switch that I'll strategically place...not only for solar routing but also for my "LiFePO4-only" loads so I can switch them to use the AGM house battery only.
Incidentally, I've been looking for a LiFePO4 battery monitor (current shunt-based) to track charging/discharging (coulomb-meter) and couldn't find anything with (1) simple/compact wired visual display unit and (2) bluetooth functionality that fit within any kind of reasonable budget. I like to buy American, but I just couldn't find something satisfactory that fit this simple criteria for < $200, which I thought was kind of absurd. So, I've ordered and received a Chinesium product that fits the bill and was only $50. I'm currently testing it with my LiFePO4 battery (on a bench in the garage, loading with a couple 12V headlight lamps and re-charging with AC-DC charger), and so far, so good. If anyone is interested to try this product, it's a pretty small investment. It comes in 50A, 100A, and 400A versions and i went with the 100A (the shunt becomes quite large with the 400A). Search terms in amazon are " "groupchn" battery coulomb counter". Picture of the one I got shown below:

 

[dstefan]

Incidentally, I've been looking for a LiFePO4 battery monitor (current shunt-based) to track charging/discharging (coulomb-meter) and couldn't find anything with (1) simple/compact wired visual display unit and (2) bluetooth functionality that fit within any kind of reasonable budget. I like to buy American, but I just couldn't find something satisfactory that fit this simple criteria for < $200, which I thought was kind of absurd. So, I've ordered and received a Chinesium product that fits the bill and was only $50. I'm currently testing it with my LiFePO4 battery (on a bench in the garage, loading with a couple 12V headlight lamps and re-charging with AC-DC charger), and so far, so good. If anyone is interested to try this product, it's a pretty small investment. It comes in 50A, 100A, and 400A versions and i went with the 100A (the shunt becomes quite large with the 400A). Search terms in amazon are " "groupchn" battery coulomb counter". Picture of the one I got shown below:

FWIW if your shunt doesnt work out, or for others looking Victron’s 500 amp Smart Shunt is a bit over $100 on Amazon or about $160 with the wired monitor. Seems to fit all your criteria and is a quality piece of equipment. Mine’s been great for 3 years so far.

 

[Xterabl]

FWIW if your shunt doesnt work out, or for others looking Victron’s 500 amp Smart Shunt is a bit over $100 on Amazon or about $160 with the wired monitor. Seems to fit all your criteria and is a quality piece of equipment. Mine’s beeing been great for 3 years so far.

Thank you for this heads-up. I did see the smart shunt but somehow never noticed a pairing with a wired monitor. Probably I will learn a painful lesson with this budget monitor and get something like Victron down the road. Having said that, bluetooth is such a cheap, commoditized technology it's somewhat appalling the kind of price premium that quality, well-known brands put on it. Anyways, we shall see, thanks again!