Hello,
I run the Classic 150 on a 200AH LiFPO4 24v battery (with BMS)
I don't seem to find the rebulk voltage setting. Can anyone help me find it please ?
Also, is it possible to have disable the Absorb stage ?
I need BUlK charge to 28.8V -> stop charge -> rebulk to 28.8Vvolts when battery is at X volts.
(my float setting is already set to the minimum)
Hi lox,
ReBulk voltage setting should be in the Charge>Advanced menu, and is Defaulted to 8.0 V, IIRC.
You could set the Vabs to Vfloat.
But, seems to me, that ReBulk would normally be set BELOW Vflt.
Am not sure just what you are trying to do ... More Later, Vic
I just finished setting up my backup solar generator using LifePo4. I looked around on this board, and several other boards to try to come up with a charge profile that would work. There are many factors to consider depending on how the system is used. Please, others chime in here and give their advice, but I wanted to jump in on this thread as well. My system is a 12v system with a 500ah bank. I use the following settings:
Bulk to 3.5v 14 (3.5 cell)
Absorb to drop .05c (5A) 14 (3.5 cell) 3 Min (classic 150 minimum)
Float to maintain at 3.375 13.5 (3.375 cell)
Equalize (auto off) 14 (Time 00:00) (NO EQUALIZE)
Rebulk 13.2 (3.3)
Aux1 â€" SOC% Low (Auto) On 25% Off 35%
Battery Efficiency 99%
My system always has a small load from the BMS, and our green room (cannot afford to do an entire house, but we can run 1 room complete green for power emergencies. I have the AUX1 relay to trigger a small 10A backup charger when the battery bank is low to assist (via 12v SSR). I chose to not push the standard at 14.4v, as the charge loss is minimal, and better for the LifePo4 cells to not hit the top end, the curve is quite rapid after 14v, and the 3 min absorb will not affect the charge cutoff timing much. I too was not sure about the "Rebulk" setting, as I had not factored this in. Above is what I set it to initially, anyone else have a better recommendation? I chose 3.3v / cell as it is just above the nominal battery charge, and if the bank hits this charge it can safely top off the charge without losses (3.3 - 3.4 seems to be the cell resting voltage after full charge). Eventually I will be installing this solar generator on the 30A powermax auto transfer switch for essential house loads (refrigerator, freezer, furnace fan, sump pump) as a UPS of sorts. These loads can be limited or manually cycled for extended outages.
My house is also 85% powered by an Enphase solar array at 6.45kw. The backup generator is powered by 6 120w panels at 12v array. I chose the lower voltage array and generator setup for a working 12v array in case that system fails, I could still easily power 12v devices with emergency equipment. The classic may be happier with a higher voltage array, but this still works well so far, and will cross that bridge later if need be. Hope this helps.
>> Bill
Quote from: Vic on December 12, 2015, 12:24:27 PM
ReBulk voltage setting should be in the Charge>Advanced menu, and is Defaulted to 8.0 V, IIRC.
Thanks, how did I miss that ...
What about the ChgTime > Absorb setting ? I did put the minimum (3minutes) Does that mean that it will absorb for 3min max ?
Quote from: eyeinthesky on December 12, 2015, 12:34:20 PM
Float to maintain at 3.375 13.5 (3.375 cell)
Float for a Lifepo4 batterie ? Don't do that.
Hi lox,
If your Absorb time setting in the Charge menu has a Min & Max time setting, then, you should really try to Update your Firmware in the MNGP (assuming that the Classic is not a Lite), and also in the Classic, if that is at all possible.
Believe that Absorb Min/Max times were removed from the Firmware about 2.5 years ago, and there have been many improvements in the Firmware in the past few years.
OR, perhaps, am misreading what you are saying about having set a Min time ...
Vic
Are you top or bottom balancing your lithium pack ?
I asked Cniemand who is the lithium bottom balancing aficionado of this board to comment - he knows a lot about the Classic and setting it up for lithium.
Larry
Theres a few concepts you need to sort out, in regards to a charge controller. You cant actually stop a CC from charging for the simple reason that you need the CC to supply your loads once the battery is full. So float is a fact of life. The trick for the lithiums is to choose a float voltage that more or less reduces the charge current to 0. Various people here have reported doing this by trial and error, just keep reducing it in 0.1V increments until you get the result you want. Becasue LFP is so stiff, you will see an interesting thing happen when it goes to float, the batterys will discharge for a short period. Thats normal. Until the inside voltage comes into alignment with the new outside voltage. On lead this only lasts a couple seconds, but for lithium people report minutes to an hourr even.
Rebulk is almost never used. It really only applies if you charge early in the day, heavy loads kick in around noon, and you need to do a big boost in the pm. Generally the solar day is too short to fit all that, and we just live with a single cycle a day. Some go a step further and further limit the cycles to once every n days. Others even set float voltage to absorb voltage and just let the sun manage the absorb duration. The sun is a funny beast.
To recap float is nothing but a a lower charge voltage. Hence if you disable absorb altogether, banks still change on float mode if they are flat enough. Just a bit slower is all.
Quote from: zoneblue on December 13, 2015, 02:41:03 PM
Theres a few concepts you need to sort out, in regards to a charge controller. You cant actually stop a CC from charging for the simple reason that you need the CC to supply your loads once the battery is full. So float is a fact of life. The trick for the lithiums is to choose a float voltage that more or less reduces the charge current to 0. Various people here have reported doing this by trial and error, just keep reducing it in 0.1V increments until you get the result you want.
Thanks zoneblue for your input on this. I got concerned when it was stated to not use float on my battery bank, and I almost pulled my post not wanting to give out bad advice, as I am new to this board. I just gave the settings that seemed to work for me on my setup, but spent a lot of time close monitoring the system and batteries. The key is that I set my float charge at the battery charged resting voltage, not at a higher charging voltage. Basically the float voltage cuts down the array wattage, and only sustains the load, while not adding any amperage to the battery bank while the sun is still shining. I've confirmed this by monitoring the system status with the WBjr, MyMidnite2, local app, and many checks with the voltmeter, until I felt comfortable with the setup.
For my system, I have my battery bank top balanced, as it should never reach LVC with the small backup charger. Also, at the 14v (3.5v cell), I have also left some headroom at the top end as well. I also use the miniBMS on the bank for the extra added protection. The only downfall to that setup is the contactor uses more power than I want it too, and an SSR DC relay is not designed for current to move both directions.
Regarding rebulk, I don't see this ever happening with my system, unless I have a full grid outage, and the transfer switch load is being used, but the system will only act as a UPS for several hours this way, and the inverter will shut down before LVC, and the MiniBMS will also go into LVC as a fail-safe.
>> Bill
ZoneBlue is correct. Rebulk isn't really needed with Lithium setups. He also is quite correct about the Voltages having a moment or two after it freshly switches from ABSORB to float before the FLOAT cycle really figures itself out. If I am around when it switches to FLOAT what I will do is turn on the Kettle or space heater for a minute to remove that "surface charge" and then switch it off. The CC will keep up fine with the Float at that point. FLOAT cycle isn't detrimental to LFP batteries, you simply have to match a voltage that keeps them topped up without actually charging. The VOC on a LiFePo4 cell is ~3.34v. If your bank has small background loads throughout the day then by setting it slightly higher you can keep them charged and full by sundown.
I have a 48V nominal pack. We work based on voltages per cell. So whatever number you have, times the amount of cells by the individual voltage in order to get to your CC settings. I have my BULK set to 3.5v per cell. So, I have 16 cell = 55v. Absorb is also 55v or 3.5vpc.
FLOAT is interesting. In my setup I used to use the open circuit/non-load voltage of a charge cell at 3.34 or 53.4v. Problem is that Lithiums have a very linear discharge curve in the center. Very little voltage change. So, if you leave FLOAT to 3.34vpc you may actually be slowly taking aH out of the pack throughout the day. Leaving you... say 10aH lower than you would have been if you set it higher to compensate for loads as FLOAT is designed to do. Since changing it to 54v FLOAT or 3.375vpc I am always left with a full bank by sundown and yet is a safe enough voltage that I can be gone and it won't slowly creep to overcharge the bank. As the bank is being drawn throughout the night, when it goes through another set of cycles is brought to square one anyhow.
Now, this is of course based on the philosophy of Bottom Balancing. That is, I drew all my cells completely down to 2.75vpc and then charged them all up to 3.5vpc in series. Thus each cell functions as a 100aH battery rather than one being 103, another 110, and another at 109. You always want to operate lithiums within the middle ranges because on both sides of the charge/discharge curve it either plummets or shoots up drastically with very little useful energy in either end.
Also. If you have a Whizbang JR, use it for END AMPS! Otherwise you never are getting a proper charge. When I did not have one I had to set my CC to 8 EndAmps in order to compensate for background loads. If I set it for 5 amps (Lithiums are considered full at 0.05C Endamps) then with the background load of say 2-3 amps, the battery is then seeing an effective ENDAMP of 3!!! With the Whizbang JR it terminates correctly regardless if you are using a heavy load at that moment of switch such as a space heater or kettle.
I've done it this way since installing the system in june 2013. Over 885+ days. It has worked beautifully. It keeps the cell balances very tight amongst each other.
- Cloud
@ Bill / EyeInSky (Parsons Reference?? )
A number of people choose to Top Balance and then use BMS boards to shunt power around. For me... I do not trust all those boards to not fail after a number of years.
When you purchased your cells and were told that they were 200aH each... they actually very likely were not exactly that even coming from the same batch. One may have 200. Another 203. One more could have 215. All are considered 200aH.
The Chinese do this interesting thing where they understate their capacities. Giving you something better so that the cells end up lasting the cycle-life stated.
If you Top balance each cell to 3.5v you are correct. You are leaving headroom, which is great! You are keeping them from possible over-voltage. Although if there is a wild variance amongst your "200aH" cells then for that one 215aH cell to reach 3.5v... your other set have to shunt a bit of power to remain at 3.5v each. The problem I see with it is when you get around to discharging them. Even if all your cells are "fully charged" at 3.5v each you still have variances in capacity. They are not actually balanced at all. When you pull them all down you only have the capacity available of the cell with the least amount in it. Say the 200aH one. But if you continue to pull it down to a certain voltage then you run the risk of over-discharging that smaller cell because you have moved all the variances to the bottom of the pack. The only way to keep from doing that is to only discharge them to say 20-30%SOC.
You leave yourself having to actively monitor both ends of the spectrum. Should your BMS fail and you inadvertantly use more of the pack than you wanted (Could be away and those small loads from the BMS continue to pull things down) your larger cells will have juice in them still while the small ones are pushed into the dirt.
I prefer to bottom balance because I'd rather be in a situation that should the pack go into an "over-discharge" state... all the cells are within millivolts of each other at the bottom so they just bleed off to 100%DOD and do not murder each other. Leaving me to charge them back up again as by that point my inverter itself would have shut down and only have the CC remain to bring them back up the next day.
Occasionally bringing the pack to 100%DOD is ok. Actually CALB CA cells are rated at 2000 cycles @ 100%DOD. 80% gets you 3000. However, if you push a cell beyond to less than 2.5v you seriously compromised its purpose as a battery.
Just my 2 cents. :)
Cloud
Lastly. Forgot to add. I only use the REBULK mode should I get into a situation where I was running a large load like a space heater during the day while it was sunny but then without me paying attention a cloud layer passes over the array. If it cuts the power back so much that the heater places a draw on the bank itself to any substantial amount then It is needed to have a rebulk. Otherwise if you had the sun return... it would shoot back up to the float voltage and charge it back at a very slow rate. I.E. = Lower amps. If you hit a REBULK voltage then having it jump back to Absorb it could quickly ramp up the amps to put it back into the bank while the last bits of the sun was available.
Thanks for all those information.
I ended up setting it up this way:
28.8V Bulk (as adviced by the manufacturer)
5A Absorb EndAmps (100Ah pack)
27.2V Float Voltage (as my cells seem to be resting at 3.41V)
Then Rebulk at 26.6V. But with float keeping the voltage at 27.2V, rebulk can't happen ?
Quote from: Cniemand on December 13, 2015, 07:15:31 PM
The CC will keep up fine with the Float at that point. FLOAT cycle isn't detrimental to LFP batteries, you simply have to match a voltage that keeps them topped up without actually charging.
And as you hint at, with solar you also get this daily automatic total charge termination. Its called sunset!
Hydro would be another matter, requiring more thought.
But what is that Rebulk setting for then ?
Here is what the Midnite support told me:
QuoteSetting the FLOAT to a very low voltage like 10V would in effect disable it which is what most Lithium Battery users do.
The Classic has a “REBULK†feature you could set to 23.4V and when the battery hits that setting the charger would start charging again.
But if I do so, when absorb is done, my classic shows an error "Battery Over V" and when the rebulk voltage is reached, nothing happens ...
@ LOX
Rebulk is useful only in a rare circumstance with LFP. Say, you reach FLOAT. Your battery got to 5 ENDAMPS and switched over. Still plenty of sunlight left and you decide to run a heavy load. A cooker. A space heater. A gaming PC. Who knows. You pick. You go about your day not thinking about the solar input because your bank is full and FLOAT should be able to take care of the 15-30amp draw on the system.
However, a stream of clouds comes rolling in. They cut your incoming solar watts in half and your CC can no longer cover all FLOAT loads so it starts drawing down your bank. When those clouds blow by the sun will resume charging the pack. However, if it is in FLOAT it will only put in enough amps to push the voltage up to the FLOAT setpoint. Meaning = You would end up recharging the pack at a much lower rate as it wont allow the voltage to raise much. If you discharged your pack enough to reach the REBULK set point, it would place it back into a Full Cycle where it can give everything it can do based on the sun intensity and then taper back off to 5 END AMPS.
Besides that... REBULK really isn't much of a utility for LFP batteries.
@ ZONE
Exactly! That sunset is a failsafe of sorts. A cosmic reset. :) If we were running on hydro I probably wouldn't bother to have my FLOAT above the resting Cell voltage. LFP do not care whether they remain fully charged or at 50% all the time. With hydro always putting in something 24/7 It would make more sense to just leave FLOAT to 3.34vpc. With the occasional manually started REBUlK just to reset everything (SOC - WHizbang jr).
Also. If my system was one that sat unused for long periods of time I would either do a full battery disconnect to leave it or adjust the voltage setpoints lower than 3.34vpc so it never gradually creeps up as the CC would be a small draw on it anyway.
When you have access to the bank on a daily basis while utilizing the battery you do not come to a point where it ever starts to creep up in over-charging scenarios.
- Cloud
I didn't manage to get it rebulk by putting a high load that drains the voltage lower than the rebulk voltage. It only triggers the FLOAT MPPT mode. The voltage has to be lower than the rebulk voltage for a certain amount of time?
But all in all it seems all good with float set to 27.2. With that setting the current to the battery (measured by the Whiz) stays between 0.1A, 0 and -0.1A. I guess that means that my 27.2V float setting is all right ?
@ Lox
I am not sure why the quote above says that most LFP users disable FLOAT. I certainly don't If you did disable it then you would end up in a position where instead of having a 100%SOC bank at sunset, it would be a bit lower depending upon your chosen REBULK setting. I want my pack to be fully charged so overnight I have access to 80% of it. Not just 60-70%.
If your float is showing around 0amps on the local app from the whizbang and your voltages are not creeping up then i'd say you have it about right. I am curious though. Your voltages have it a 3.41v each. TO me that seems like a slight charge. the resting voltage not under load is 3.34v. I have my FLOAT at 3.375. You use a BMS? (I do not) Those little boards to each cell? Is it possible that they are actually shunting around some energy to dissipate and leave them at 3.41? I thought some of those boards start shunting at either 3.4 or 3.6 depending on the board you chose?
Quote from: lox on December 14, 2015, 05:00:52 PM
I didn't manage to get it rebulk by putting a high load that drains the voltage lower than the rebulk voltage. It only triggers the FLOAT MPPT mode. The voltage has to be lower than the rebulk voltage for a certain amount of time?
Was that voltage measured with the load still attached or without the load?
I am thinking that the load was attached, and if that is correct, the batteries real voltage ,without load , would be higher and when the charger kicked in the panels were able to meet both the load and also to raise the battery voltage, so, naturally there would not be a ReBulk , only Float or Float Mppt..
Battery voltages that we talk about are not under load voltages but, unloaded &
at rest voltages.
At rest means after 3 or more hours of no charging or load applied... ie a stable voltage
hth
I have a BMS. I bought my battery pack from a Chinese factory and they made the BMS. I have one board for each row of cells (2 rows of 8 cells in series, then the two rows are in parallel)
One of the equalizing PCB failed so I disconnected one row until I solve the issue with the manufacturer. Those cells are now resting at 3.41V ...
But yes I can confirm that setting my FLOAT voltage at 27.2V, the current to the battery, under normal load, is never more that 0.1A and never less than -0.1A once it stabilizes (takes a few minutes to stabilize) and SOC stays at 100%.
Thanks for your worthy advices buds.
I didn't plug the battery temp sensor. Should I on LFP batteries ?
About working without a BMS, do you have to manually re-balance your cells often ? How do you bottom balance them ? I mean, how to discharge them to the same voltage. Is any special equipment required ?
I am getting some LiFePo4 soon and I am also wondering about temperature compensation for them .
Larry
@ Larry and LOX
You do not need to Temperature compensate LFP batteries. I recently plugged in the BTS only for the purpose of having an ambient temperature reading on the Local App of sorts. It really doesn't modify the charging parameters at all.
@ LOX
Sorry to hear your PCB failed. Paying money for something... it should work. I hope they resolve the issue. One of my beliefs in regards to active BMS units is that they have a high probability of failure and when they do you take out half the pack.
"Bottom balancing" is pretty straight forward.
1. Pull each cell down to 2.75v (I used a giant resistor and heatsink in order to pull out energy at a 30amp rate for the 100aH cells)
2. After pulling a cell down to 2.75v the first time, let them recover over night.
3. They bounce back up in voltage. Pull them back down to 2.75v. let them rest for 30 minutes to recover.
4. repeat and let recover 30 minutes until they stabilize at around 2.75 (within a few millivolts of each other.)
5. When each cell is at 2.75 (within a few mV of each other. They are 100% discharged so balanced amongst each other) String the pack in series and charge up to 3.5v each ( 3.5v *16 cells = 56v)
6. When your CC hits the 5 ENDAMPS they are all full and balances with themselves. They may be off a few mV of each other. They settle down in rest.
I have a video explaining more about it and showing how I did it. Its connected to the little world icon under my screen name.
Rebulk on solar is pretty much pointless. On hydro it would make more sense, because there is no "day", and rebulk would allow you to create one. As to what would happen if you set Vfloat to 10V on a 24V bank, well id be surprised if that worked, really i would, as the controller needs a voltage to regulate the loads at. Sure couldnt do that at 10V. AND, theres been a thread here recently where the controller does some pretty interesting things if the battery is held at a higher voltage than the controller. The controller seems to dump power into the PV for a period until it drops to sleep mode. ON this matter i beleive that 2079 alters the timing of that so exactly what occurs would need to be compared to the actual firmware you are using. Bob might chime in and clarify there, but for now your approach to choosnig a float voltage appears to have done the job, and with Tcomp disabled you should be away laughing.
I ended up buying an awesome little and cheap BMS, it is called MiniBMS (http://cleanpowerauto.com/product-support/) (the HousePower version). It took me so long to find such a great product that I wanted to share it with you guys. That BMS really is cheap and very clever, which makes it safe and efficient.
So I tuned my Midnite Classic to an absorb setting of 28.5V so the top balancing of the cells is properly done. It has been running on my LiFEPO4 cells for months and not single issue so far (I produce/use between 3 and 4Kwh daily)