Inverter/Charger in conjunction with MPPT?

[TheArgobus]

What I have right now: 24V LiFePo4 Battery Bank (8S4P) at 2850 watts. The anode is connected to the Victron Energy Battery Shunt BMS-700, then goes to the negative bus bar at the electrical panel. The cathode is connected to a 100 amp inline fuse, which then goes to the battery switch at the panel. This bank is charged via shore power through the AIMS Power 2kW inverter/charger, which is connected to the bank through the circuit side of the shunt and fuse.



Where do I attach the positive and negative lines coming from the MPPT controller? Conceptually, it seems that I could just connect them to the negative bus bar at the panel, and the battery side of the battery switch. That would be convenient because the MPPT is next to the electric control panel. But given that this is electricity, and I can be quite a dummy, I'm not sure if having solar juice coming down at the same time as being connected to shore power would work well.


If I was plugged in, wouldn't the MPPT see the battery voltage as "28" and say "okay cool no need to charge these batteries" or am I misunderstanding a core concept here?

[john61ct]

Originally Posted by TheArgobus

What I have right now: 24V LiFePo4 Battery Bank (8S4P) at 2850 watts

A bank cannot "have" watts. Maybe you mean Wh (watt hours, Ah * volts)?

A pack can be "rated" for a wattage output for a certain amount of time, but you likely don't mean that and that is a very nebulous subjective attribute anyway.

OK, nit pick yes but units are important.


> Where do I attach the positive and negative lines coming from the MPPT controller?

If there is no dedicated voltage sensor wire, then directly to the positive post / bolt on the bank.

Negative usually to a "ground" bussbar, must be "outside" the BM's shunt along with all loads and sources

> If I was plugged in, wouldn't the MPPT see the battery voltage as "28" and say "okay cool no need to charge these batteries"

Yes and perfectly appropriate.

Multiple disparate charge sources can operate concurrently, no problem IRL.

[bus-bro]

Might want to have your controller's output switched and fused (or use a circuit breaker) and attach electrically as close to the batteries as possible. And switch and fuse the solar controller's input too. Do what the controller's installation docs says.

[TheArgobus]

Awesome, thanks. I'll pick up two more inline fuses for pre and post MPPT and I'll hook it up back at the battery bank. Whenever I get around to doing that, I'll post an update.

[kazetsukai]

I have Tesla packs, in series (as of yesterday) for 48V (12S). I connect my packs to the bus bars via an inline DC breaker on positive, through an anderson/winch connector.




The MPPT connects to the negative bus bar via a 100A shunt, and to the positive bus bar via a 100A DC breaker. My DC fuse panel connects to the positive bus bar via a 40A inline DC breaker and connects to the negative bus bar via another 100A shunt.

The inverter connects directly to the bus bars with four 4/0 cables. Basically, everything meets at the bus bar and it has been this way for over a year without issues (at 24V until yesterday). The batteries and MPPT can be disconnected from the positive bus bar by tripping the breaker, the inverter cannot (I don't know of anything large enough to facilitate the 4/0 cables).


I think the Victron shunt can measure bidirectionally, but in your configuration it may only measure current to/from the batteries, so it isn't going to measure general DC load if you connect the MPPT to the bus bars. This may be desirable, or undesirable to you. I know in my case I have two shunts each with its own meter because I wanted to measure DC input and output, not battery charge/discharge.

[TheArgobus]

Thanks for the input, it's reassuring. My shunt measures only what's going in or out of the batteries, which is fine for now. It sounds like what you have is what I'm envisioning for my own system, although I am envious of all that space available for your electrics.



Also where are people getting Tesla packs? Are they buying from totaled Teslas or what?

[kazetsukai]

Originally Posted by TheArgobus

Also where are people getting Tesla packs? Are they buying from totaled Teslas or what?

Yeah, these come from totaled vehicles. Personally I love mine, but they have their challenges. These are pretty awesome batteries, because of heavy internal parallelization (6S74P) and built in water cooling they can handle extremely high charge and discharge current no problem.


The biggest drawbacks are:

• They are each 6S instead of 7S or 8S so most inverters can only tap into about 70% of their total capacity. To me, this is more a feature than a bug, I'd rather the outlets (AC) go out and my lights (DC) stay on.
• Some people claim they are not a safe chemistry. They will not tolerate overcharging or discharging is what it comes down to, when managed within specifications they are perfectly safe.
• Proper BMS solutions for them are plentiful or scarce depending on your point of view. Most folks get by with RC balancers (like I have until now). Now that I'm charging and discharging a >1kW I'm switching to a proper BMS that can disable the inverter before they over discharge and stop the MPPT before any cell overcharges.

[john61ct]

I would very highly recommend only using LFP chemistry out of the many dozens of other LI variants used for propulsion in passenger vehicles

for use cases where the pack is anywhere near even large permanent human habitation, much less small mobile living spaces.

The risk of thermal runaway (fire bad, boom bad) is just astronomically higher if any of your infrastructure fails in the wrong way without robust failsafes taking all possible failure modes into account.

I am certain there will be commercially sold professionally engineered solutions coming out one day in the decades to come, but for noobs & amateurs like me DIYing, no way.

And I highly doubt insurance would pay out on a DIY setup either, if a large liability claim were made.

[john61ct]

And those cheap looking circuit breakers make me very nervous.

My reco is fuses, Eaton, Cooper / Bussmann, Blue Sea etc ignition protected types like Class T, MRBF, and ANL.

To get as secure in a breaker costs hundreds each.

Stuff advertised to automotive / stereo hobbyists is crap.

[kazetsukai]

Originally Posted by john61ct

The risk of thermal runaway (fire bad, boom bad) is just astronomically higher if any of your infrastructure fails in the wrong way without robust failsafes taking all possible failure modes into account.

I keep seeing this claimed, but I don't see this actually happening. I don't care what chemistry you use, if you're on lithium and you overcharge (any cell) you are risking a fire. Lead acid carries the same risks when overcharging as well- pop, zap, flames. Personally I think this is FUD.

Do not overcharge your batteries, do not overcharge any cell. Do not overdischarge, and do not charge when near or below freezing. Simple rules to follow. Every individual cylinder in the pack is fused, these are built for a high standard of safety. Not to say there haven't been improvements in safety.

Originally Posted by john61ct

I am certain there will be commercially sold professionally engineered solutions coming out one day in the decades to come, but for noobs & amateurs like me DIYing, no way.

And I get it. I also don't recommend these batteries to others- you really have to know what you're doing. Not saying I'm any more or less of an expert, but I do know their operating specifications and how to keep them happy.

Originally Posted by john61ct

And I highly doubt insurance would pay out on a DIY setup either, if a large liability claim were made.

Ah, but lets be real- even with any other battery vendor if you DIY your electrical do you honestly think insurance will pay out on a claim? Probably not if it comes to light that the cause came back to _any_ component in your system. I think you either start with a licensed electrician or you are at risk here.

Feedback on the "cheap" breakers taken, I'll just have to cycle these out over time. They're properly sized and installed securely. I'll agree on wanting high quality here, I'll disagree in that they're probably better than nothing. Will they work when it counts? Who knows regardless of the brand, but you're probably right that Blue Sea and friends are on the more likely side of that equation.

[john61ct]

Originally Posted by kazetsukai

I keep seeing this claimed, but I don't see this actually happening. I don't care what chemistry you use, if you're on lithium and you overcharge (any cell) you are risking a fire. Lead acid carries the same risks when overcharging as well- pop, zap, flames. Personally I think this is FUD.

I very strongly stand by what I said.

Of course any chemistry has risks. Lead is more dangerous than LFP IMO.

LMO is also very safe.

But the LI chemistries used by passenger EV in the west are 50-100 times more likely to go into thermal runaway than any of the above.

LiPo that much again more so than LI.

Fundamental characteristics are very very different.

Go hang out in the RC hobby, DIY ebike and EV forums, many members have burnt down their homes.

It is very easy to **say** do not overcharge. But **every** such forum says never take your eyes off your LI cells, do **not** automate charging, do not leave them alone for a second.


> And I get it. I also don't recommend these batteries to others

It would be more responsible to state that whenever discussing your setup then.

[kazetsukai]

Originally Posted by john61ct

But the LI chemistries used by passenger EV in the west are 50-100 times more likely to go into thermal runaway than any of the above.

_Maybe_ when overcharged, but not when operating within spec. And overcharging is a recipe for fire in any chemistry, yet for some reason this critique only comes up in the context of EV batteries. What about purely homemade 18650 banks or custom built LFP banks from China? Is it more the chemistry or the fact that there isn't a paired BMS solution?

Originally Posted by john61ct

Go hang out in the RC hobby, DIY ebike and EV forums, many members have burnt down their homes.

I've had some of my own lithium RC batteries catch fire- but then again these are not RC batteries. I'm not saying it can't happen, I just don't see this happening to the degree warranting the FUD. You're not claiming homes haven't burned to the ground with misused LFP banks, you know they have.

Originally Posted by john61ct

It is very easy to **say** do not overcharge. But **every** such forum says never take your eyes off your LI cells, do **not** automate charging, do not leave them alone for a second.

And again, I don't care what chemistry you use, there is no system where this rule does not apply. If you are living in a confined space with a battery bank you better have good monitoring and use it!

Originally Posted by john61ct

> And I get it. I also don't recommend these batteries to others

It would be more responsible to state that whenever discussing your setup then.

This is over-the-top ridiculous. I'm not going to pick a technology to power my rig only to disavow it at every opportunity. It works exceptionally well and has for a long time with minimal oversight.

[john61ct]

Of course they only go thermal when something fails.

But stuff fails all. the. time.

I understand we have different profiles for risk aversion, so of course do as you like, but please don't get (too) annoyed if I make this point once in each thread when it arises, noobs are too easily led astray by bad examples.

[kazetsukai]

Originally Posted by john61ct

I understand we have different profiles for risk aversion, so of course do as you like, but please don't get (too) annoyed if I make this point once in each thread when it arises, noobs are too easily led astray by bad examples.

And you're 100% correct... "Tesla" is not synonymous with "good" here. Had I known what I was getting into, I may have made different choices.

[john61ct]

All depends on context.

If only all the million-dollar professionally engineered redundant protective and temp-management systems the packs are usually embedded within, came along with them when they get removed.

But they don't, and replicating them is beyond a DIYer's skillset.

As I said, the day will come that trustworthy professionally packaged systems using those chemistries are sold for use in a mobile House bank context.

There may be some now we just hven't come across yet.

But for me I'd wait a few years, collect trusted user reports a while first before jumping in.

Even if it carried the Tesla marque