Lithium charge cutout with Classic

Started by bigal, September 11, 2018, 06:11:08 PM

Previous topic - Next topic

bigal

#15
I've done more reading from group of early lifepo4 users in Australia. To prolong cell life, they finally adopted a charging plan that shuts off charge at preset voltage near 3.45 volts (HVD).  When cells drop to 3.3, bulk charging is resumed.

No system that used a form of float voltage was successful LONG TERM. These batteries are not tolerant of being kept at or near 100% charge . It shortens the number of charge cycles.

This brings me back to my original premise of disconnecting the Classic at a preset voltage, and reconnecting at a lower voltage.

This could occur multiple times a day.

I'd like the Midnite Brains to comment on the wisdom of the using a Classic in this manner!! Am I going to have problems?  How could I safely do this??
27-235 watt Canadian panels, 2 CLASSICS,  Magnum Inverter,  Forklift Battery,  powers entire house easily.
Lots of sun here, recharge by 1 pm normally. 2 Fujitsu mini splits for summer AC.  13 solar thermal panels for heat.

FNG

Quote from: dgd on September 13, 2018, 12:11:52 AM
Quote from: bigal on September 11, 2018, 06:11:08 PM

Simple question, if I use a solenoid disconnect, will Classic tolerate a quick disconnect without damage? Charging current would be low, perhaps 10amps@48v

I have had battery bank disconnected from a Classic when in bulkmppt charging. No problems and the classic  continued to power an AC inverter. On another system when a battery breaker tripped the Classic again suffered no ill effects and with nowhere for output current to flow to the PV input current just dropped to near zero.
Dgd

Correct, The classic couldnt care if the battery goes away it will just set in absorb or float and power itself from the pv. if a sweep is forced it will of course then drop out do to 0 amps during the sweep but it will not hurt anything at all

australsolarier

bigal, it seems to me to you are still a bit confused.  HVD disconnects the battery via a breaker or relay. (to protect the battery from being overcharged.

in the midnite classic you can set the max voltage that it ever permits. and low. it has nothing to do with bulk and float settings. or HVD or LVD.

my bulk setting is 56V. end amp 8A. so in the morning it charges to at max 56V and then tapers out. after end amp kicks in, the midnite switches to float and keeps it there abouts at 53.6V.

you will have to adjust the voltage that the midnite shows with a good quality volt meter. (in the menu, they have excellent manuals) and then it might  still be out a tenth of a volt or so. that does not seem much for lead acid batteries, but it is much for lithium. in my case to absorb at 56, one midnite classic is programmed to 56.2V,, the float for 53.6, is programmed 53.7. it then DOES float and absorb in the demanded voltages) (after the voltages have been adjusted.)

generally you should not have a low voltage disconnect. i am not usually going underneath 60% SOC. but i have tested down to 10% with both systems. the large 400ah system did trigger a programmable disconnect. now,  a low voltage disconnect should only happen when there is none or little solar input anyway. it is the last of the defenses, before the battery will be destroyed. in my opinion, a disconnect should not harm the midnite classic. there are 10s of thousands of midnite classics out there and it is the least of the problems or questions thrown up. similar say you have a 10kw inverter running at full capacity and turn off a switch to disconnect a 10kw load. should just turn off without problem.


Ron Swanson

Are you doing all this with the BTS disconnected or what?

Aravilla

This thread is a bit old now but I thought I'd throw a couple of comments in as the OP seemed to be slightly misaligned on either terminology or strategy. Most importantly - disconnects are not operational, they are emergency. And in an emergency, if you lose a solar controller vs burn the whole place down, take the dead solar controller 100% of the time. That said:

1) HV/LV/HC disconnects are just that - disconnect actions. Throw a switch, the connection is broken, hard and without remorse. If one is planning to use this strategy for anything other than "holy @*#&, the power @*#& has really hit the shittin' fan" type disconnect, that's insane. It's potentially dangerous, operationally disruptive, and is gonna be expensive. With this type of disconnect, it's programatically planning for the scenario where all other failsafes and settings have been violated. Operational vs emergency. But...

2) On that note, I would never want the controller itself determining when a battery bank HV/LV/HC disconnection is to take place. The controller should operate within the configuration given to it which prevents the need for emergency disconnects. No knock on the Midnite product line, however they lack a few important things that are vital in determining when a disconnect is required, and furthermore, the product is designed for a single (non-related) purpose.
2a) First, the controller is only responsible for the charging (and possibly coordinating/monitoring) aspect of the battery lifecycle. Asking a solar controller to perform a HVD/LVD is, in two words, operationally wrong. Your BMS is responsible for this.
2b) Secondly, in the HVD/LVD scenario, the Midnite products lack a voltage sense feature, meaning whatever voltage the controller is seeing is not actual battery voltage. The resolution the controllers have is not sufficient for LFP management when they are at or near 100% SoC - this requires cell level monitoring and very accurate voltages typically to two decimal places. And you cannot be 100% sure that the solar controller has the entire battery picture especially with other sources of charging or load. The BMS is installed at the cell level and sees all this.

3) Back to the BMS concept - the BMS is what should determine when a disconnect is required, and then trigger action(s). This can (and should) be very quick. There are other steps that can be taken in the pre-disconnect process to either eliminate the need or gracefully take things offline. In the case of alternators, for example, a pre-process that helps prevent the destruction of the alternator (field disconnect will save it from killing diodes) is a nice to have. Same with a solar controller in a HV/HC disconnect scenario being told to turn off gracefully and possible mitigate the issue. But, if the secondary process either doesn't eliminate the problem or complete in a reasonable timeframe, the hard disconnect needs to be executed. This is why we typically wire into charge and load bus bars. Each bus is configured with a disconnect that is programmatically controlled. The nice thing with dual buses is that just because charging was disconnected, doesn't mean the load side stops functioning. Highly useful in a boat house bank scenario, for example.

Properly setup equipment should rarely present a problem (and that's what the disconnect is for) and I can't say I've heard of a Midnite controller, for example, programmed to 14.5v running away to 15.0v on its own. In fact, when we charge our LFP bank, we can watch the bulk voltage stay very steady throughout the day and climb slowly...until it doesn't...right around 99%. At this point the controller jumps to absorb quickly, and amps back-off rapidly and in less than 1 minute, the controller will move from bulk to float, with voltage never going higher than we've configured. Should it do that, the BMS takes over and would disconnect the whole battery.

One thing owners strive for is the notion of "getting everything possible out of the setup". I've had conversations with folks when setting up systems that goes something along the lines of "but the batteries never get to 100%, only 99%, what's wrong?". They feel like they're being ripped off that last 1% or something. But in reality, it's an installer protecting your assets and giving some wiggle room in case "the worst" should happen.

Anyway, hope this helped someone.

australsolarier

aravilla,

well said.
and yes it would be nice if the midnite classics would have voltage sensors at the battery terminals. this is particularly beneficial for lithium batteries.