Best mppt charge controller for LiFePO4?

[fafrd]

Victron 250/100.

Nice! So 6000W powering 100A @ 48V = 5440W (10.3% overpanel). That’s a sweet set-up.

$600 came from $736 being 123% the price of the panels. I didn't consider additional controllers.

Something to consider is paralleling more panels on the same controller. A 2S4P array would only suffer in the individual string being shaded even on the same controller.

My shading situation is peculiar and greatly compounded by my choice of half-cut panels.

The rules change with half-cut panels because they can put out 50% Imp @ 100% Vmp and a Series string has no way to improve upon that through bypass diodes so the entire string runs @ 50%.

I have a specific set of 4 consecutive panels which will be shaded only on their lower-halves for a good hour late in the morning and the only way I can capture that energy is to either string them in parallel or within their own individual string (either 4S1P or 2S2P).

And I’ve got another 3-4 panels that will be shaded vertically and I shade 1/3rd at a time, so a 3S or 4S string would be best for them.

And the final two panels are a mess and will suffer from half-shade, full-shade, and partial shade at various points throughout the morning, so just connecting them in parallel will probably work best (with half-cut panels).

I’m not sure optimizing output by planning for the best string configuration on a sun-array basis is worth the trouble, but since we’re talking about trying to save 50% to 67% of the 40% of full daily production that would otherwise be lost with a full-series or full-parallel string, my gut tells me it’s an effort worth making.

Keeping your array to a controller capable of 150V would help with the budget. 200 and 250V units carry a premium. I believe a 48V system coupled with paralleling within the array can accomplish the same results as multiple SCC.

Yes, if I want the option to go to a 24V battery, I need SCCs Abe to handle at least 120V or 150V. That’s my plan.

But I may spring a bit more for one 170V+ MPPT if I decide I want to capture some of the energy on those 3 vertically-shaded panels as the shadow from the tree slowly shifts off of them...

While only half is installed (3S3P), my array will be 3S6P. I could shade 1 entire string, and the other 5 would be unaffected on the single controller. It would be no different if I had 6X 3S arrays on 6X SCC, 3X 3S2P arrays on 3X SCC, etc. I have also oriented them horizontally such that only one "string" within the panel will be impacted rather than all strings within the panel.

I’ve been considering all of those same issues as I struggle to settle on the best way to configure my 10 panels. It would be rather straightforward if they were 72-cell panels like the ones in my AC-coupled array, but half-cut panels dramatically changed the picture.

Each half-cut panel is essentially 2 half-panels connected in parallel, and this means a half-cut panel can put out Imp/2 @ Vmp (which a standard panel can never do).

If your panels are also half-cut, I’m interested to understand more about how you’ve configured them to deal with shade..,

 

[fafrd]

I had a 20A EPEVER. External Temp Sensor. USB cable adapter. A bit difficult to get it sync with laptop. Also dropped Custom for default setting.
I upgraded my camper to a 27' travel trailer. 100w suitcase style panel to two 250W used panels.
I also upgraded to Victron for the blue tooth networking with the Smart Shunt (with voltage sense) and Smart Phone display.

This is my biggest concern with the Epevers (of any of these 2nd-tier budget MPPTs, frankly).

So your Epever would occasionally revert to default setting and abandon your custom (‘user’) settings? Was their any specific triggering event (battery discharging below low-voltage cutoff, etc...?).

And in comparison, your Victron always wakes up and does what you programmed it to do (after a similar triggering event)?

 

[fafrd]

I am tossing this out in a general response.
Many people SCREW UP their charging profile simply because they do NOT understand the gritty details. This often leads to overcharging cells and having BMS's cutoff and worse. We ALL know that LFP only requires CC which get's it to 85-90% and CV to top the battery pack to 100%. 100% IS NOT 3.65V per cell ! Anything above 3.55 is a WASTE 3.500-3.650 only really represents <5% and the cells will ALWAYS settle within 1 hour of charging to 3.65 to 3.55 or a tad lower.

Chargers / SCC's use BULK (CC) which pushes as much current as available to the packs.
FLOAT is CV for varying amperage as demanded by either the battery packs OR inverter draw. An SCC will provide a Fixed Voltage and floating amps as far as the incoming solar power can provide.

IE: My system ATM is generating 7.5A just to service the demand of the Inverter and the battery bank is getting 0.4A because it is at 99% - 27.9V - 3.487V per cell AVG. If I turn on my coffee maker which pulls 51A, the SCC will try to provide as much as it can to the Inverter and draw the rest off the batteries for any shortfall. If my battery pack drops to 27.4 or lower, that initiates a RE-Bulk Mode which the pushes teh Amps & Volts to bring the battery bank back up before it flips into Float Mode.

FLOAT allows for the cells to Top Off and Balance because it is a Trickle Charge related to capacity demand. As balancing is happening some cells will drop while others go up and often that creates a bit of differential so the Pack will take an Amp or Two for a few minutes or more to top the pack before it drops into Storage Mode (not available on all BMS'). It is NOT overcharging or pushing the cells too high, it is capped by the voltage setting for it.

Stacking SCC's works fine and you are correct, the settings MUST be the same or unpleasantries happen. Smart Linking in Parallel adds more management capability usually with one interface as opposed to two or more. Also pending on SCC, there may be Ground Fault Protections and more functions and features which only ONE SCC needs to address. Having two trying to manage Ground Faults can cause havoc, a friend spent 2 weeks chasing that wild monkey, finally ended up listening to what he was told, at his wife's insistence.

Sigineer, YiYen are two Huge OEM's who produce equipment for VARS (Value Added Resellers) such as AIMS and many others. Their equipment is generally Tier-3 & Tier-2 level gear, they don't do "Value Tier-4" stuff. Yiyen will not sell direct but Sigineer will if you do it right. Sigineer now also have full on EV Charging "Kits" and manufactured linkable smart battery packs as well which link to their higher end gear. We do have users here using Both Brands and assorted models of their products. I started out with a Yiyen APC-3024 and the thing is a TANK and only cost me about $550 USD back in 2015. Since been replaced by my Samlex EVO which is a world of difference. NOTE, Yiyen/Sigineer average between 82-90% efficient and they have a higher Idle Draw as well. There are tradeoffs to consider.

Hope it helps, Good Luck

I think I just found YiYen’s MPPT offerings: https://www.yiyen.com/product/mppt-ii-solar-charge-discharge-controller/

Unfortunately, they seem to be limited to Lead Acid batteries...

 

[100 Proof]

I've got a Morningstar TriStar MPPT that is highly programmable. It accepts up to 150V. The 60 Amp version can be had for just under $600. Morningstar also has awesome customer service!

TriStar MPPT - Morningstar Corporation

TriStar MPPT, Morningstar Corporation, Solar, Charge Controller, Maximum Power Point Tracking

www.morningstarcorp.com

100 Proof

I was getting set to jump on an Epever MPPT charge controller but between reading some of the struggles members have had getting their Epever MPPTs to not overcharge their LiFePO4 batteries and this warning I found in this Epever document: https://www.epsolarpv.com/upload/file/1812/EPEVER Tracer AN_Main_Pres.pdf

“High current series doesn’t have self-activation function for lithium-ion battery, thus it cannot be used with lithium-ion battery.”

I’m having second thoughts about Epever and broadening my search for the easiest/best MPPT charge controllers for LiFePO4 batteries.

Specifically, what I am looking for:

8S/24V
60A (or possibly 50A)
100V minimum (but ideally 120V or 170V)

As easy/failsafe to use when charging LiFePO4 as my 10A AC charger (specifically not having to worry about either overcharging or wasting solar power when the battery is discharged enough to take charge).

 

[Pappion]

So your Epever would occasionally revert to default setting and abandon your custom (‘user’) settings? Was their any specific triggering event (battery discharging below low-voltage cutoff, etc...?).

I think it was dormant for too long. I don't travel with it connected. Don't bother with it while on RV park AC.

 

[fafrd]

I think it was dormant for too long. I don't travel with it connected. Don't bother with it while on RV park AC.

Thanks. So after disconnecting for a long time period, it did not wake up in the same state it was in when you disconnected it.

Did that happen more than once?

I’m considering Epever and this issue may be my single greatest concern

 

[fafrd]

I've got a Morningstar TriStar MPPT that is highly programmable. It accepts up to 150V. The 60 Amp version can be had for just under $600. Morningstar also has awesome customer service!

TriStar MPPT - Morningstar Corporation

TriStar MPPT, Morningstar Corporation, Solar, Charge Controller, Maximum Power Point Tracking

www.morningstarcorp.com

100 Proof

Not a brand I’m familiar with, so thanks for the recommendation.

Unfortunately, 150V will limit me to 2S.

As I mull-over my options to charge both 24V and possibly 48V batteries, I’m thinking 3 160V+ 60A MPPTs may be worth a modest premium.

Charging 24V, I can go 2S2P + 1S3P + 3S1P (the best to deal with my shade pattern).

If I ever switch to 48V, I can switch to either 2x2S2P + 2S1P or 2S2P + 2x3S1P.

So yeah, 150V is the minimum so so can go 2S (panel Voc_max is 52V) and 170V while 200V would provide me the freedom of going 3S on a string or two.

 

[Pappion]

Thanks. So after disconnecting for a long time period, it did not wake up in the same state it was in when you disconnected it.

Did that happen more than once?

I’m considering Epever and this issue may be my single greatest concern

Lost custom settings every year, maybe twice a year.

 

[fafrd]

Lost custom settings every year, maybe twice a year.

Well that’s concerning. But if you always remember to check / refresh settings after a long period of disconnect, was the issue manageable?

Did you ever have an episode of it losing custom settings without being disconnected for some period of time?

 

[Pappion]

Well that’s concerning. But if you always remember to check / refresh settings after a long period of disconnect, was the issue manageable?

Did you ever have an episode of it losing custom settings without being disconnected for some period of time?

I don't keep a log book. It held for 4 weeks continuous use. Pain in the but to program, had to disconnect comm cable and reconnect.
I'm going LiFePO4, pricey, so I would rather not damage them. At the time Victron 100/50 was little more $ than EPEVER 60A with MT50.

 

[Pappion]

Multiple Victron Smart SCC can form a network. One will dominate, no telling which one, so set them the same.
The Smart Shunt with Voltage Sense optimizes network charging. Many people use 2 or more SCC.
So you can group the panels by shading position, or have some on top and some on the ground.
Victron Smart 150/35 (150v, 35A) is $323 at Amazon. Smart has Bluetooth, Blue Solar does not.

They also have a wired network.
Xscapers boondocking stage trailer

 

[HRTKD]

Multiple Victron Smart SCC can form a network. One will dominate, no telling which one, so set them the same.
The Smart Shunt with Voltage Sense optimizes network charging. Many people use 2 or more SCC.
So you can group the panels by shading position, or have some on top and some on the ground.
Victron Smart 150/35 (150v, 35A) is $323 at Amazon. Smart has Bluetooth, Blue Solar does not.

They also have a wired network.
Xscapers boondocking stage trailer

I have an account over on IRV2, but only 7 posts. I don't get over there much. The threads relevant for my trailer are dead, dead, dead.

 

[Pappion]

With Victron your also paying for the connectivity and the software.
I can call up 30 days of performance logs and plot trends on my Smart Phone.

 

[fafrd]

I am a user of Victron products in all my builds and have close ties to the Manufacturer. There is nothing better. Only MFG with a 5 year warranty, integration with their other products, smart devices and an impressive product line. But Victron, unfortunately does carry a premium cost UP FRONT.. Something to be said about not buying it twice. There are places to find Un-heard-of deals on Victron from top sources. Not Amazon or Ebay.

The Victron products look beautiful, but outside my budget.

The SmartSolar 250V / 60A would allow me to feed a 4S string of shaded panels to optimize power out as the shade shifts away (1/3rd of a panel at a time).

But I can’t find it for under $640.

So I’m interested in your top sources for unheard-of-deals...

 

[snoobler]

I did a little research and don't see how your split cell panels present any more concern about shading than standard panels. In fact, it appears that they are more shading tolerant. I think your pursuit of high PV input voltage units is counterproductive to your shading concerns and a 2SXP array will be likely be more productive than a 3 or 4S array.

My panels are oriented horizontally to ensure only the lower 3rd of the panel is subjected to limited shading to perform as shown in this video:

Split cell will likely have different results.

 

[fafrd]

I did a little research and don't see how your split cell panels present any more concern about shading than standard panels. In fact, it appears that they are more shading tolerant. I think your pursuit of high PV input voltage units is counterproductive to your shading concerns and a 2SXP array will be likely be more productive than a 3 or 4S array.

My panels are oriented horizontally to ensure only the lower 3rd of the panel is subjected to limited shading to perform as shown in this video:

Split cell will likely have different results.

Easiest to think of my 10-panel array as 3 subarrays:

-subarray #1 is 4 panels that start out shaded and the shadow shifts down until they are only shaded at the bottom halves and the top halves are clear. Half-cut panels will work well for this sub-array and I don’t need high PV input voltage (2S / 110V will suffice). A 2S2P string feeding a 150V 60A MPPT is my plan for this subarray.

-subarray #2 is 4 panels that start out shaded and the shade shifts across them vertically (so similar to the video you linked to). The half-cut panels aren’t buying anything in that subarray and a 4S string would maximize the power out (unblocking one bypass diode at a time, in sequence).

-and subarray #3 is two remaining panels in between these other 2 subarray that are more of a mess (sometimes half-shaded horizontally, sometimes 1/3 or 2/3 shaded vertically). For those I think the best I can do is connect 2P to a 30A MPPT if charging @ 24V or connect 2S to a 30A MPPT if charging @ 48V.

So perfection would be:

Subarray #1 2S2P 150V 60A
Subarray #2 1S4P 250V 60A
Subarray #3 2P1S or 1P2S 150V 30A

The alternative I’m considering is to take the first ‘clean’ panel from subarray #2 and add it to the 3rd subarray (at least if charging @ 24V):

Subarray #1 2S2P 150V 60A
Subarray #2 3S1P 200V 50A
Subarray #3 1S3P 100V 50A

But getting a second 2S2P subarray out of this is a challenge - the problem is that the half-cut panels only deliver half-current when half-shaded (so putting those 2 central panels in any 2S2P chokes off a lot of power).

But I’m not in a rush and still evaluating alternatives...

 

[snoobler]

I would question if a single 2S5P array on a 150V controller would have any inherent disadvantage compared to 3 sub arrays on 3 controllers, but your situation sounds like a mess, and I don't pretend to understand it based on the description.

Are you sure solar is even a viable option for you?

 

[Pappion]

Just watched a Will Prowse video comparing solar controllers. On my tablet, so no link. Anyway he said multiple controllers will not fight with each other.
Caveats:
1. Some will not go to bulk charge if battery is sufficiently charged when the startup. like the SCC that is late to get the sun.
2. Most SCC have a dumb absorption time of 2-3 hours at max volts. SCC late to get the sun may extend this time and overcharge.

Anyway the absorption time for LiFePO4 should be short, per Battle Born setup instructions.

IMHO: For Lead Acid the set absorption time is dumb. The time should end early when the current drops. It should lengthen for low charge state. Victron Smart SCC Lead Acid setting has an Adaptive Absorption Time option and a Tail current exit option. Tail current accuracy is better with a Smart Shunt in the network. Multiple Smart SCC work together. They will take far better care of your Lead Acid than EP-Whatever. I like Will and think he should do a deeper dive on Victron Smart SCC.

 

[fafrd]

I would question if a single 2S5P array on a 150V controller would have any inherent disadvantage compared to 3 sub arrays on 3 controllers, but your situation sounds like a mess, and I don't pretend to understand it based on the description.

Are you sure solar is even a viable option for you?

In reverse order:

Yes, I’m sure solar is a viable option for me. If I went with a Microinverter-based array, the shading would cost me ~15% of the potential output with no shading.

If I use a bone-headed string configuration that only captures the energy from panels when they are not shaded at all, I’ll lose 40% of that potential output with no shade issues.

The reason a 2S5P string will basically deliver that full 40% shading loss is because all 5 of those 2S strings will always need to be at the same voltage and that means they will always be at 2xVmp (because their will always be at least 1 2S substring which is unshaded) and all panels with any shade at all will choke-off not only themselves, but also their series string panel (unless they are only half-shaded in which case they will only half-choke their series string partner because of the fact that I’m using half-cut panels).

Now what is interesting is that after writing that description of my 3 subarrays, I had a new idea I had not thought of before. By composing two 2S2P arrays using a 2S string from what I called Subarray #1 and a 2S string from what I called Subarray #2, I found a way to form two 2S2P strings which will deliver close to optimal shading performance.

So thanks for the discussion (even if the details of my mess escaped you) - who knows whether I would have settled on this idea all on my own?

So I’m now pretty settled on 3 MPPT charge controllers, two for 2S2P and 1 for either 1S2P @ 24V or 2S1P @ 48V.

This means I can get by with 150V MPPTs and am thinking even though the 3rd only needs 30A of output, I may just get 3 identical 150V 60A MPPTs for uniformity and redundancy (and because once you pay up for 150V, the premium for 60A over 30A is modest).

Where an optimal 10-MPPT Microinverter-based solution might limit my losses from shading to ~15%, I’m estimating this new 3-MPPT configuration will only lose ~0.36% to 1.1% beyond that (so pretty darned good).

To come back to the earlier point about the drawback of parallel series strings when dealing with shading, separate MPPTs mean 1 2S2P string can harvest from one or two panels in parallel at a string voltage of ~1xVmp_panel (bypassing the other 2 or 3 panels with exact string voltage of Vmp_panel - 3xVbypass_diode), while the second 2S2P string can harvest from 2 or 4 panels at a string voltage = 2xVmp_panel.

This near-optimum performance is only going to be when charging a 24V battery (since output of 1 panel in a 2S series string will be sufficient to charge). If using this same configuration to charge a 48V battery, I’ll lose a lot more power, probably closer to 40%, because I’ll only get output at high enough voltage to charge @ 48V when both panels in a series string are not blocked by shade (at least 4 out of 6 substrings within those 2 series panels, to be precise).

This is one of the factors leading me to try my system at 24V first - I’ll get close to 25% more power charging at that voltage...

 

[fafrd]

Just watched a Will Prowse video comparing solar controllers. On my tablet, so no link. Anyway he said multiple controllers will not fight with each other.

Yeah, I’ve watched that video several times. It’s one of the main reasons I’m considering Epever...

Caveats:
1. Some will not go to bulk charge if battery is sufficiently charged when the startup. like the SCs that is late to get the sun.

Is that because the late-waking-up MPPTs see the voltage of the already-wakened MPPTs already actively charging the battery rather than voltage of the battery itself?

I need to understand more about that since I will have at least one MPPT getting significant shade until later than the others. Perhaps this is the value of any kind of communication between the various MPPTs - no confusion about best battery charging mode... Thanks for raising this issue, I’ll look more deeply into it.

2. Most SCC have a dumb absorption time of 2-3 hours at max volts. SCC late to get the sun may extend this time and overcharge.

Anyway the absorption time for LiFePO4 should be short, per Battle Born setup instructions.

Yeah, my understanding is that it’s vital to be able to disable Equalize function for LiFePO4 and Absorb should either be skipped as well or short. Another thing I’ll put on my list of requirements - thanks.

IMHO: For Lead Acid the set absorption time is dumb. The time should end early when the current drops. It should lengthen for low charge state. Victron Smart SCC Lead Acid setting has an Adaptive Absorbtion Time option and a Tail current exit option. Tail current accuracy is better with a Smart Shunt in the network. Multiple Smart SCC work together. They will take far better care of your Lead Acid than EP-Whatever. I like Will and think he should do a deeper dive on Victron Smart SCC.

One of the few things I (happily) don’t need to worry about is charging LA batteries...