Looking for opinions

[VanKitten]

I have a MPPT controller....60amp.    700W solar panels up top...bright sunny days.
The controller was supposed to be for lithium, but clearly not.
I say that because in the set up, the selection on LI (lithium) freezes the remaining parameters to preset value.     BUt, oh boy...that is. Maxium voltage...16.0V, maxium voltage cutoff threshold.  16.0v.     Everyone agrees this will cause an explosion.

In reading through all the other parameters... (boost, float, etc) this is clearly really for lead acid.

So..here is where I am.   With consult (Electric Engineer) the parameters were all set thru "user defined".   Max. Voltage 14.6v. Threshold 14.6.  The other parameters were set to 13.8. With the user defined duration set to 5 seconds.    In other words to direct the mppt to end boost and float and return to max output.

So, this is what I am seeing...  the battery status cycles (about 5 seconds) between steady on...(all ok) and fast flash (low voltage).  Meanwhile, the system error light cycles the exact same...coming on when the battery indicator is flashing.  I watched the system while this cycles...the error light and battery flash is happening when voltage drops to 13.5.  -  13.3. There abouts.  And comes back to normal  with 14.2 - 14.6.   The voltage then continues to drop till it trips the system error again.....   However, the battery monitor says I have increased the batteries by 90 amps since yesterday.   I believe it.   Still at 13.1 volts..but that is normal for lithium.  

I am telling all of this so that you can understand the dilemma I am facing.

I believe this is only charging about half of the time...only when the system error light is not on.    

The seller knows absolutely nothing about this device.  THe manufacturer is not responding to queries. (Turns out it is in China)    Sellers tech staff are clueless (this is why I hired an electrical engineer...but, he is unable to get data sheets on this device)

So...sell says there is another MPPT that is for lithium from a company in Amsterdam.  
Offered to get it for me ....

If I drive away with the MPPT I have it amounts to acceptance of delivery.   But, no way to know for sure that the next one will be right.   (Remember   His staff cannot offer help).

I purchased a Morningstar previously...but cannot use it due to a serious design issue that would fail in my system. (In a different post for a couple weeks back). After som investigation..I found many mppt have this same flaw.

What do you think?   Roll the dice and wait for Amsterdam?  OR, Accept what I have and get out of here now?

Any opinions on what is really happening?   Think my interpretation is correct?

 

[John61CT]

Yes, every single lithium vendor and "lithium ready" piece of gear out there specs **way** too high voltage for LFP longevity.

But 16 is way out there, maybe for NMC?

You need to config custom setpoints IMO 13.8V and 12.1V, minimum Absorb hold time.

 

[John61CT]

Do not go with another no-name.

You don't need anything labeled for "lithium"

The Dutch vendor that is good is Victron.

 

[John61CT]

Their MPPT 75/15 is $100, handles 220-250w

Victron USA tech support:

Justin Larrabee

+1 207 3540493 · [email protected]

But I think just set your BMS' OVD to just cut off (or short out) the **inputs** to the controller would work with any vendor's.

Specs on your panels again?

VoC, Imp and nominal watts each, how many?

each

 

[jonyjoe303]

its strange how the mppt is cycling back and forth, it should be steady at one voltage and then slowly go up or if it hits float go down rapidly and stay there. Maybe the controller is designed to cut the voltage to check the resting voltage of the battery then go back up, my rc chargers do that when charging lithiums, but I don't understand why the error lights would come on.

What I'm thinking about the voltage being set at 16 volts on lithium, that might be intentional, to fast charge them. On my ecoworthy 20 amp mppt, I have to manually set the bulk voltage to 15.5 volts (as high as it will go) to fast charge my 110 ah lifepo4. I notice if set any lower than that it will go to float too soon, if set to 14.6 volts, it will charge very slowly (controller thinks battery is fully charge). I do have an overvoltage relay to disconnect the solar panel, if 14.6 volts is reached and the bms hasn't tripped, plus I always monitor the battery when fast charging. 

You could set to lithium setting and monitor how it behave, unplug it if the battery reaches 14.6 volts and it hasn't cutoff. There might be voltage drop in the system, so your lifepo4 probably won't get all 16 volts. It might be relying on the bms to cutoff when 14.6 volts is reached. My own experience (I have 2 x 110ah lifepo4) 14.6 volts won't fast charge a lifepo4, that would be more of a trickle charge. But check the actual voltage on the battery while charging, it will be different then what the controller is displaying. The voltage on the battery is the one that determines if the battery is getting close to full.

One thing I notice is your user define settings you set seem very low to me, if you have any voltage drop, the battery won't even be reaching 14.6 volts, the controller might read 14.6 volts and switch to float, but the battery might be at 14.2 volts, the float also might have nothing going into the battery. I rely on the float to top off my battery, float on mine puts out about 2 amps (at 13.8 volts I see the battery getting no charge and actually discharging). If there is no voltage drop you can keep them as they are, which is a safe modest charge. On my system I have a .5 volt voltage drop, I keep the bulk at 15.2 volts (unless I want to fast charge), and float at 14.4 volts. In 4 months this setting hasn't tripped the overvoltage relay. The battery is never fully charged but its at least about 95 percent.

 

[John61CT]

No, 13.8V is the ideal setting for LFP longevity.

It is important to avoid the shoulders on both ends of the voltage curve.

EV usage is much more intensive, they might only get 1000 cycles.

Being gentle with House usage patterns likely to get at least 3-4,000, fact is we don't know yet.

It is true you'll lose maybe 3-5% AH off setting 14.X as "full", but since quality cell vendors usually over-deliver on spec'd AH you should still get full rated capacity.

16V is just nuts, as I said designed for a different kind of lithium ion, likely NMC, 3.7Vpc rather than 3.2Vpc

 

[John61CT]

And no need to increase voltage to get fast charging, LFP has such high CAR a 100AH bank could be charged at 400A and completely refill at 13.8V in under 14 minutes.

Assuming your other infrastructure is robust enough to not get too hot, but for the LFP cells themselves no problem.

This is usually a moot issue, have you seen the pricing on 400+A charge sources? And most LFP banks are much larger than 100AH anyway.

 

[ZacLee]

Ideal max charging voltage for LiFePo4 cells is 3.65V, but this chemistry is tolerant of mild over voltage, to a max of 4.2V per cell. Under typical solar charging conditions, you can set set the max voltage to 16V (assuming a a 4 cell LiFePo4 battery and battery management device that prevents any cell from exceeding 4.2V) to get it closer to 100% charge without damaging the cells. Keep in mind that other lithium chemistries ARE NOT tolerant of even slight over voltage.

Some helpful info about LiFePo4 charging from powerstream (a battery vendor):

https://www.powerstream.com/LLLF.htm

 

[John61CT]

As I said, the vendors all state those (IMO too-high) charge voltages.

My theory is their business model suffers when people find out they need to get specialized replacements for their expensive existing infrastructure, not least charge sources.

More cynically, consider the effects of your product lasting "too long".

At that voltage you may well get 1500-2000 cycles, but wouldn't double or triple that be better? Especially it is only costing you a tiny percentage of your theoretical AH capacity?

 

[John61CT]

Some other relevant LFP notes, good further reading links below

Bare cells: ​Winston/Voltronix, CALB, GBS, A123 & Sinopoly

Systems: OceanPlanet (Lithionics, not in Oz, just FYI), Victron, MasterVolt, Redarc

EV usage is very different from much gentler House bank cycling. Most EV people talking "lithium-ion" mean other chemistries not as safe as LFP, much shorter lifetimes, and with completely different setpoints and behaviors.

My charge settings for LFP: 3.45Vpc which = 13.8V max for "12V".

Either stop when voltage is hit, or if you want another couple % SoC capacity, stop when tailing amps hits .02C, or 2A per 100AH.

If you can't "just stop" then Float at 13.2V, but may shorten life cycles.

Don't need to fill up all the way at all, as far as the cells are concerned, in fact it is bad for them to sit there more than a few minutes, so do so only if loads are present, ready to start discharging right away.

Store cool at 10-20% SoC, or higher if not getting topped up regularly (I would monthly), letting them go dead flat = permanent unrecoverable damage.

Same with charging in below freezing temps.

Persistent heat also drastically shortens life.

Charging at 3-4C or even higher is no problem, as long as you wiring is that robust, vendors spec lower out of legal caution.

Going above 14V won't add much AH capacity, but will shorten life cycles dramatically.

The point is to look at the SoC vs Voltage chart, and avoid the "shoulders" at both ends, stay in the smooth parts of the curve.

Following these tips, letting the BMS do active balancing is unnecessary and potentially harmful, just look for LVD / OVD and temp protection. Multiple layers if a very expensive bank, don't rely on any one device to work.

Check cell balance say monthly to start, then quarterly, finally every six months if that seems safe to you.

This thread is long but informative
http://www.cruisersforum.com/forums/showthread.php?t=65069 , give Maine Sail your close attention.

Also his summary notes here
http://www.pbase.com/mainecruising/lifepo4_on_boats

**Everything** at that site is worth reading, very valuable. He also has great articles in Practical Sailor.

 

[ZacLee]

As I said, the vendors all state those (IMO too-high) charge voltages.

My theory is their business model suffers when people find out they need to get specialized replacements for their expensive existing infrastructure, not least charge sources.

More cynically, consider the effects of your product lasting "too long".

At that voltage you may well get 1500-2000 cycles, but wouldn't double or triple that be better? Especially it is only costing you a tiny percentage of your theoretical AH capacity?

LiFePo4 is much more tolerant of excessive charging voltage than other lithium chemistries.  I'm not aware of any data that suggests cycle life or capacity will be reduced by slightly elevated charging voltage, especially in solar cyclic use when the charging voltage will not be sustained for longer than the solar is at peak production. 

Deeply discharging the cells substantially reduces cycle life and discharging the cells below a minimum voltage is extremely damaging.  It is good practice to avoid discharge below 50% capacity to maximize battery life. 

Also, unlikely other batteries chemistries including lead acid, lithium batteries have long life if not stored at 100% capacity.

High rate discharge and charging will also shorten battery life. High rate is typically defined as above 1C. (so a battery rated at 100 ah should not be discharge faster than 100 A)

 

[John61CT]

> LiFePo4 is much more tolerant of excessive charging voltage than other lithium chemistries.

It won't damage them short-term, but definitely shortens lifespan off the back end.

> I'm not aware of any data that suggests cycle life or capacity will be reduced by slightly elevated charging voltage, especially in solar cyclic use when the charging voltage will not be sustained for longer than the solar is at peak production.

Yes you are currently unaware, but if you follow the links I posted you can fix that.

And the source of the charge is irrelevant, the question is, is there enough amps available to get the bank to (user definition of) Full?

Unlike Lead, LFP will absorb all the current the SC can deliver, very common to be completely Full by noon or 1pm. At which point charging should be shut down, or diverted to carry loads only.

What do you think you're actually gaining by going over 13.8-13.9V ? Other than a tiny % of capacity.

And you're right, you should not let them get to that Full point if the bank is just going to sit there, only if Loads are going to start pulling SoC back down.

> Deeply discharging the cells substantially reduces cycle life and discharging the cells below a minimum voltage is extremely damaging.  It is good practice to avoid discharge below 50% capacity to maximize battery life. 

**Absolutely not**, this is leftover "old lead" thinking. At 7X the price of a lead bank, you'd be silly to no make use of what's available, the other factors here more than make up for any DoD/cycles effect.

To be clear, it is no problem to use 90+% of a LFP bank's rated AH capacity.


> High rate discharge and charging will also shorten battery life. High rate is typically defined as above 1C. (so a battery rated at 100 ah should not be discharge faster than 100 A)

Again, lead thinking. Even many quality AGM have no problem with higher rates than that.

But largely irrelevant in House bank usage anyway.

 

[ZacLee]

And the source of the charge is irrelevant, the question is, is there enough amps available to get the bank to (user definition of) Full?

**Absolutely not**, this is leftover "old lead" thinking. At 7X the price of a lead bank, you'd be silly to no make use of what's available, the other factors here more than make up for any DoD/cycles effect.

To be clear, it is no problem to use 90+% of a LFP bank's rated AH capacity.


> High rate discharge and charging will also shorten battery life. High rate is typically defined as above 1C. (so a battery rated at 100 ah should not be discharge faster than 100 A)

Again, lead thinking. Even many quality AGM have no problem with higher rates than that.

Solar cyclic use means charging ends when the solar panels are not longer in full sun so that is very different than when they are float charged continuously such as for backup use. In solar cyclic use, batteries would only be full charged (or overcharged) for a short time.

The cycle life of LiFePo4 cells (and nearly all battery chemistries) varies with the depth-of-discharge.   80% depth-of-discharge means far few cycles than at 50% depth-of-discharge.  For example, the 24V Telecel brand LiFePo4 batteries I use have this cycle life rating::

80% depth-of-discharge, 2000 cycles
70% depth-of-discharge, 3000 cycles
25% depth-of-discharge, 8000 cycles

High charge and discharge rates cause heating, gassing, and other undesirable effects which reduce both capacity and cycle life.  This is why batteries discharged at C/20 have a much high capacity (amp-hour or watt hour) than if discharged at C. In other words, a 100 amp-hour battery discharged at 5 A would yield more amp-hours than if it were discharged at 100 A.

 

[John61CT]

OMG, you don't need to cite Peukert, yes cycling the bank gently is a given for House usage, that's not an issue. Obviously, we're not talking backup usage, we've also agreed they should not sit anywhere close to full for any length of time.

And LFP should not be float charged, that's another holdover from lead thinking; just get them to the target voltage when you need to use them, stop charging and start discharging.

Those cycle numbers are true relative to each other, but gross underestimates, **if** you keep away from the voltage shoulders at both ends.

Out of hundreds of installs, no one is shallow cycling unless it suits them at the time, going right down to 10% (or even 5% when needed) is fine, no harm done at all.

Why spend $12,000 on a bank when one half the size at $6,000 is likely to last 15 maybe 20 years?

But sure, shoot for 25-30 if you have the money to burn, your setup your call.

 

[VanKitten]

Ok... I have spent a fair amount of time researching LiFePo4 over the last year. It is the reason I choose them. That being said...back there to topic

This MPPT is obviously not intended for lithium. Not acceptable for permanent usage.

So the questions are, do you think my assumptions about the behavior of this mppt is probably correct.... charging only half the time but charging....etc. if I depart with this it is de facto acceptance. Should I sit tight and wait of another...maybe it too will turn out to not be acceptable.

 

[John61CT]

Yes, I would not spend any time or energy on it, nor trust that seller anymore.

BTW "lithium" is meaningless here, it is an umbrella term, too general to say anything about charging profiles.

Your bank is specifically LFP (LiFePO4), correct?

That 16V controller is perfectly suited to a pack of NMC batteries, which is another type of "lithium" chemistry, of which there are dozens.

Most DIYers building their own "lithium" packs from 18650 cells use NMC, it is much more energy-dense per pound, but shorter lifespan and IMO dangerous.

____
The Victron MPPT line is very commonly sold by LFP specialist dealers in marine electric systems, specifically to be used with LFP.

But again, all you need for a controller to be suitable for LFP, is the ability to set a user-defined custom battery profile:

turn off temp compensation, set your own Voltage setpoints and minimum Absorb time.

Do you still have the Morningstar?

Was the only thing wrong with that their saying you shouldn't open (cut off) the output circuit with your BMS OVD relay?

What do you think about opening the input circuit instead?

 

[jimindenver]

Having the panel input cut off instead of the controller is genius. I wonder if you could use the load terminals to turn on a relay until you reached the voltage you want. Then by turning off the load terminals you would cut the power to the relay turning off the solar. Even if it would work it isn't for everyone, neither of my Morningstar controllers have load terminals. The Eco-worthy had them but I never used the feature.

I am just hoping y'all have it figured out by the time my Lifelines need to be replaced.

 

[VanKitten]

Yes, I agree that cutting off the solar panels would be a great idea...except...solar is only one of three possible chargers feeding the system. I need that relay on the power bar for the chargers (plural).

Yes, this system is LiFePo4.

I sent the Morningstar back. Just will not perform in my system.

So, I see that Victron only has 75amp/15amp

Why so low on the output side? I will see what they say about the proper support for my batteries.

 

[John61CT]

> Yes, I agree that cutting off the solar panels would be a great idea...except...solar is only one of three possible chargers feeding the system. I need that relay on the power bar for the chargers (plural).

No, one control mechanism - in this case OVD -can be set up to trigger multiple relays in different locations.

The field current on an alternator is another example.

Remember, we're not just talking about damage to the charge source. A charging current just cut off suddenly can also create a load dump, surge/spike damage to sensitive load devices.


> So, I see that Victron only has 75amp/15amp

That is just one very small model. For optimum shade handling, they are cheap enough to use 1:1 on each panel. But Victron makes a very full range of sizes.

They are fully customizable by the user, either via connected PC or from a smartphone if you also buy a BT dongle.

Therefore compatible with any chemistry.

 

[John61CT]

> I wonder if you could use the load terminals to turn on a relay until you reached the voltage you want. Then by turning off the load terminals you would cut the power to the relay turning off the solar.

SC load terminals are LVD, keep current flowing only as long as voltage is high, cut off when low. Could be adapted to for example kicking a genny **on** when needed.

Functionality here is the opposite, OVD sensor on charge buss that **stops** all charge sources when the target "definition of Full" voltage is reached.