Cruisers Forum
 


Join CruisersForum Today

Reply
 
Thread Tools Rate Thread Display Modes
Old 03-10-2014, 11:53   #1
Marine Service Provider

Join Date: May 2011
Location: PDX
Boat: Gulfstar 50
Posts: 896
Charging schemes for LiFePO4

I’ve got a set of 4 Winston 700 ah LiFePO4 batteries that I am using for my house bank. I thought to start this thread so as to create a forum for discussion of charging strategies. But first if you have not read the thread “LiFePO4 Batteries: Discussion Thread for Those Using Them as House Banks” you should do so. There is a wealth of information there.


Real world or theory are welcome.
__________________

__________________
evm1024 is offline   Reply With Quote
Old 03-10-2014, 11:54   #2
Marine Service Provider

Join Date: May 2011
Location: PDX
Boat: Gulfstar 50
Posts: 896
Re: Charging schemes for LiFePO4 -Xantrex SW3000

To start with I have a Xantrex SW3000 inverter/charger. This is an older unit part number 815-3000. It does not have programmable voltages and only has Flooded, AGM and Gel profiles. There is an OEM Preset which is could be set at a service center for a large order of inverter/chargers and does not help us at all.

The follow on machine is the SW3012 (815-3012) which does have a user programmable battery setting which would allow setting voltages that would work better with LiFePO4 batteries. Sadly, I do not have that model so I cannot speak about how well it would work.
Back to the SW3000. First off you really need to have the SCP (system control panel) to control your SW3000 to have any chance of charging LiFePO4 without significant risk of an expensive failure. And you also need an accurate way of measuring the pack voltage at the pack. I’ve been using a calibrated Fluke meter clipped to the pack terminals. My Victron BMV-602S which measures voltage at the Battery disconnect switch (about 5’ of 2/0 cable from the pack) shows a voltage drop in that 2/0 cable of over 16 mV at high charge rates. (I need to move it closer to the pack).
In real world terms it does not matter which battery type you select. The absorption voltage is 14.4 (14.3 AGM) volt for all battery types and is too high for LiFePO4 in my mind. And the SW3000 is capable of charging at 150 amps. In my setup I see a max charge rate of around 135 amps when the pack is 50% discharged. At a 14.4 charge voltage and up to 150 amps you cannot turn your back on it. The upper knee is quite abrupt at this voltage and current.
The charger runs in constant current mode for all phases while charging a LiFePO4 bank. What to do? Well of course the answer is to dump the SW3000 and get another charging solution. In the meantime here is what I do.
First, remember that this charger cannot be allowed to charge unattended so don’t even think of letting it have its way with your house bank. It is only for charging your bank at the dock or from a generator under adult supervision.
My first rule is that the charger is disconnected from the battery (at the inverter/charger isolation switch) when not being used. When disconnected from the battery the SCP will not be energized.
I set the battery type to AGM with a 14.3 volt absorption voltage, and select 2 stage charging. Also, there is a parameter “Force Charge” that you should set to off. If this is on (which is the default) anytime the unit sees AC it will automatically turn on the charger. You want to manually turn on the charger.
The single most important parameter that we can adjust in the SW3000 is the “Max Chg Rate” parameter. This parameter allows us to set the maximum current that the charger will supply. The parameter is defined as Battery AH setting / 5 * Max Chg Rate. So in the case of my bank I’ve set the battery capacity parameter to 700 and with Max Chg Rate set to 75% we get 700/5*.75 = 105 amps. With a max Chg Rate of 12% we get 700/5*.12 = 16.8 amps.
Initially my pack was top balanced and fully charged on the bench then installed in the boat. I consider a pack voltage of 13.9 volts and a charge rate of less than 20 amps to be fully charged. My Victron battery monitor was manually set to 100% SOC at that time. At the dock I typically have all charging sources disconnected.
I do have to resist the temptation to keep the pack fully charged. This is something left over from years of lead acid…. Using the battery monitor I tend to charge the bank when it gets down to a 75% or 60% SOC. Still pretty high but it does reduce the charge time.
Remember this is at the dock but it could be using a generator as well. First I hook up my Fluke to the pack. Moving the battery monitor may give me good results. I’ll report back on that later once I have more info. Then I turn on the battery to inverter switch and wait for the SCP to boot up. Once all is ready I set the max charge rate to 100% and enable the charger. The battery monitor shows 130 amps. I keep a close eye on the pack voltage and when it reaches 13.8 volts I change the max charge rate to 50%. The pack voltage drops to 13.65 or so at 70 amps charge. I monitor the pack and when it again reaches 13.8 I set the max charge rate to 14%. At 14% the pack voltage again drops and the charge current is about 20 amps. Continuing to monitor the pack voltage I stop charging when the pack voltage reaches 13.8 volts and call it full. Sometimes I let it go to 13.9 volts. If need be I reset the SOC on the BMV-602S to 100%.
Then I disconnect the charger/inverter from the battery (with the disconnect switch). No chance of the charger coming on and trashing my pack. The AC side is still connected and shore power is passed through the SW3000’s transfer switch to the boats AC panel.
When away from the shore power I turn on the inverter/charger to battery switch when I want to use the inverter and back off again when I’m done with the inverter.

In short the SW3000 is not to be trusted with your LiFePO4 bank. You can use one to charge your bank but you must monitor it closely. Don’t even step into the head while charging. And swap it out for something more suitable when you get a chance
__________________

__________________
evm1024 is offline   Reply With Quote
Old 03-10-2014, 15:15   #3
Jd1
Registered User

Join Date: Aug 2011
Location: Victoria, BC
Boat: Catalina 36 MKII
Posts: 1,096
Re: Charging schemes for LiFePO4

I was always under the impression that the charge rate doesn't matter for lithiums (well, within reason). My plan is to leave my alternator set as is (let's say 14.4V) and monitor both pack voltage as well as cell voltages. Any single cell voltage reaching 3.6V will trip the charge current relay and disconnect the charge line to the lithium pack. The same sort of arrangement will happen on the load side with the load being dropped when a single cell reaches whatever voltage (yet to be determined).
Related question - I will be ordering a House Power BMS - should I use 3.6V monitoring boards on the cells or 3.8V? I am leaning towards 3.6V but am not certain so thought I would bounce that off you guys.
__________________
Jd1 is offline   Reply With Quote
Old 03-10-2014, 15:55   #4
Marine Service Provider

Join Date: May 2011
Location: PDX
Boat: Gulfstar 50
Posts: 896
Re: Charging schemes for LiFePO4

Quote:
Originally Posted by Jd1 View Post
I was always under the impression that the charge rate doesn't matter for lithiums (well, within reason). My plan is to leave my alternator set as is (let's say 14.4V) and monitor both pack voltage as well as cell voltages. Any single cell voltage reaching 3.6V will trip the charge current relay and disconnect the charge line to the lithium pack. The same sort of arrangement will happen on the load side with the load being dropped when a single cell reaches whatever voltage (yet to be determined).
Related question - I will be ordering a House Power BMS - should I use 3.6V monitoring boards on the cells or 3.8V? I am leaning towards 3.6V but am not certain so thought I would bounce that off you guys.
You can charge at higher rates and the voltages will be higher-sooner during charge and the knees reached sooner and much more abruptly.

See the graph here: http://www.cruisersforum.com/forums/...1&d=1359229048

The Clean power folks recommend 3.8 volts for ThunderSky and for the 700 ah use size "E".

Do you have a smart regulator like the MC-612 or MC-614? MaineSail has his MC-614 parameters in the house bank thread. Perhaps we can get him to post them here with a writeup of why he selected each value. Be nice to have them in this thread so that they are easy to find and in case there are any changes.

My next "long" post will be my MC-612 charging parameters and the reasons I selected them.
__________________
evm1024 is offline   Reply With Quote
Old 03-10-2014, 18:00   #5
Jd1
Registered User

Join Date: Aug 2011
Location: Victoria, BC
Boat: Catalina 36 MKII
Posts: 1,096
Re: Charging schemes for LiFePO4

Quote:
Originally Posted by evm1024 View Post
You can charge at higher rates and the voltages will be higher-sooner during charge and the knees reached sooner and much more abruptly.

The Clean power folks recommend 3.8 volts for ThunderSky and for the 700 ah use size "E".

Do you have a smart regulator like the MC-612 or MC-614? MaineSail has his MC-614 parameters in the house bank thread. Perhaps we can get him to post them here with a writeup of why he selected each value. Be nice to have them in this thread so that they are easy to find and in case there are any changes.

My next "long" post will be my MC-612 charging parameters and the reasons I selected them.
Since the monitoring will be done via automated equipment, I see no need to sneak up to the final charged status slowly although in reality it will be slow anyway because I will charge from the alternator at (only) 160 amps.
I do realize that Clean Power recommends 3.8V but that is a pack voltage of 15.2V which seems excessive. The 3.6V boards would have 14.4 V pack voltage which seems more appropriate. Thoughts ???
I have the MC614 regulator but again I am thinking of not changing any of the parameters because the alternator is also used for the lead acid bank. I am thinking that both lithiums and lead acid are charged by the alternator but the lithium has the charge relay that drops out when the bank is full. On the discharge side I have a switch that draws power from one or the other set of batteries but not both.
I am looking forward to see what charge parameters you will be using.
__________________
Jd1 is offline   Reply With Quote
Old 03-10-2014, 22:31   #6
Marine Service Provider

Join Date: May 2011
Location: PDX
Boat: Gulfstar 50
Posts: 896
Re: Charging schemes for LiFePO4

Quote:
Originally Posted by Jd1 View Post
Since the monitoring will be done via automated equipment, I see no need to sneak up to the final charged status slowly although in reality it will be slow anyway because I will charge from the alternator at (only) 160 amps.
I do realize that Clean Power recommends 3.8V but that is a pack voltage of 15.2V which seems excessive. The 3.6V boards would have 14.4 V pack voltage which seems more appropriate. Thoughts ???
I have the MC614 regulator but again I am thinking of not changing any of the parameters because the alternator is also used for the lead acid bank. I am thinking that both lithiums and lead acid are charged by the alternator but the lithium has the charge relay that drops out when the bank is full. On the discharge side I have a switch that draws power from one or the other set of batteries but not both.
I am looking forward to see what charge parameters you will be using.
Actually it does not represent a pack voltage of 15.2 volts.

The per cell monitors are there to catch an out of balance cell that has hit the knee before the others and will go beyond safe limits before the pack monitors trigger.

The House power BMS board implements the pack level HVC = 3.6V per cell , LVC = 2.9V per cell and assumes that the pack is balanced.

At least that is how I read it.

Mixed battery configuration.... I'm still working on that one.
__________________
evm1024 is offline   Reply With Quote
Old 03-10-2014, 22:48   #7
Jd1
Registered User

Join Date: Aug 2011
Location: Victoria, BC
Boat: Catalina 36 MKII
Posts: 1,096
Re: Charging schemes for LiFePO4

Quote:
Originally Posted by evm1024 View Post
Actually it does not represent a pack voltage of 15.2 volts.

The per cell monitors are there to catch an out of balance cell that has hit the knee before the others and will go beyond safe limits before the pack monitors trigger.

The House power BMS board implements the pack level HVC = 3.6V per cell , LVC = 2.9V per cell and assumes that the pack is balanced.

At least that is how I read it.

Mixed battery configuration.... I'm still working on that one.
I am confused .... isn't it 3.6 V if the 3.6V daughterboard is used and 3.8V if that daughterboard is used ????
I don't know if they also monitor overall pack voltage but I guess that is really irrelevant in the big scheme of things.
__________________
Jd1 is offline   Reply With Quote
Old 03-10-2014, 23:22   #8
Marine Service Provider

Join Date: May 2011
Location: PDX
Boat: Gulfstar 50
Posts: 896
Re: Charging schemes for LiFePO4

Quote:
Originally Posted by Jd1 View Post
I am confused .... isn't it 3.6 V if the 3.6V daughterboard is used and 3.8V if that daughterboard is used ????
I don't know if they also monitor overall pack voltage but I guess that is really irrelevant in the big scheme of things.

Just went to the bench and measured the levels on the cell boards.

2.8 volts cell loop goes from open to closed
3.55 volts Shunting resistors activated.
3.8 volts cell loop goes from closed to open

2.8 * 4 = 11.2 volts
3.55 * 4 = 14.2 volts (active balancing starts)
3.8 * 4 = 15.2

So if any cell goes below 2. or over 3.8 the pack is disconnected (open cell loop)

The House BMS only sees pack voltage and from the manual:

HVC turn on – 3.6V per cell
HVC turn off – 3.45V per cell
LVC turn on – 2.9V per cell
LVC turn off – 3.1V per cell

But remember that the house bms does not monitor cell voltages. they only report it as cell voltages because you can get a 12 volt or 24 volt versions of the board.
__________________
evm1024 is offline   Reply With Quote
Old 03-10-2014, 23:39   #9
Jd1
Registered User

Join Date: Aug 2011
Location: Victoria, BC
Boat: Catalina 36 MKII
Posts: 1,096
Re: Charging schemes for LiFePO4

Quote:
Originally Posted by evm1024 View Post
Just went to the bench and measured the levels on the cell boards.

2.8 volts cell loop goes from open to closed
3.55 volts Shunting resistors activated.
3.8 volts cell loop goes from closed to open

2.8 * 4 = 11.2 volts
3.55 * 4 = 14.2 volts (active balancing starts)
3.8 * 4 = 15.2

So if any cell goes below 2. or over 3.8 the pack is disconnected (open cell loop)

The House BMS only sees pack voltage and from the manual:

HVC turn on – 3.6V per cell
HVC turn off – 3.45V per cell
LVC turn on – 2.9V per cell
LVC turn off – 3.1V per cell
Just like it should if you have 3.8V daughter boards ... if you had 3.6V boards than cell balancing would still turn on at 3.55V but HVC would cut out at 3.6V.

Quote:

But remember that the house bms does not monitor cell voltages. they only report it as cell voltages because you can get a 12 volt or 24 volt versions of the board.
Um.... I assume you meant to say that the house BMS does not monitor 'PACK' voltage.

I am starting to think that maybe I should just use the Cellog8's for monitoring and auto shutdown rather than bothering with the house power bms system. I am not sure what the bms circuit gives me (besides maybe some automated balancing) that the cellog8 will not give me. I will have to play with the cellog8 to see how it reacts to failure events.
__________________
Jd1 is offline   Reply With Quote
Old 04-10-2014, 09:19   #10
Marine Service Provider

Join Date: May 2011
Location: PDX
Boat: Gulfstar 50
Posts: 896
Re: Charging schemes for LiFePO4

The House power bms only measures the pack voltage and activates HVC,LVC accordingly.

The cell boards measure the cell voltages and only open or close the cell loop (no data is sent.)

The House BMS also looks at the cell loop which is either on or off and sets the disconnect relay. I assume that there is a relay driver transistor on the house bms board.

At least that is my understanding. I'm no expert on these boards

Using an alternate board to control charging is reasonable if the house power BMS does not give the voltages that you would like to see. You could use the House BMS HVC,LVC as warning indicators and as pack disconnect.
__________________
evm1024 is offline   Reply With Quote
Old 04-10-2014, 11:35   #11
Jd1
Registered User

Join Date: Aug 2011
Location: Victoria, BC
Boat: Catalina 36 MKII
Posts: 1,096
Re: Charging schemes for LiFePO4

Quote:
Originally Posted by evm1024 View Post
The House power bms only measures the pack voltage and activates HVC,LVC accordingly.

The cell boards measure the cell voltages and only open or close the cell loop (no data is sent.)

The House BMS also looks at the cell loop which is either on or off and sets the disconnect relay. I assume that there is a relay driver transistor on the house bms board.

At least that is my understanding. I'm no expert on these boards

Using an alternate board to control charging is reasonable if the house power BMS does not give the voltages that you would like to see. You could use the House BMS HVC,LVC as warning indicators and as pack disconnect.
I think we need to agree to disagree. In order to see HVC and LVC, the measuring has to happen on the cell boards as that is the only way the house power system can monitor the cell voltage. I don't think we have enough detail to know for sure but monitoring the pack voltage is useless since you could have some low voltage cells and some high voltage cells giving a proper overall pack voltage while the cells are being fried.
If it was simply a matter of interrupting the loop, there would also be no way to determine if the event was a high voltage or a low voltage error (you could guess based on pack voltage but that would be poor design)
__________________
Jd1 is offline   Reply With Quote
Old 04-10-2014, 11:52   #12
Registered User
 
nimblemotors's Avatar

Join Date: Apr 2011
Location: Sacramento, California
Boat: Solar 40ft Cat :)
Posts: 1,557
Re: Charging schemes for LiFePO4

If you have balanced cells, there is no problem using a 14.4v charger
to charge a '12v' LiFePO4 pack. That is 3.6v per cell.
You don't get any damage to the cells until around 4.2v.

The key phrase is having balanced cells, which you should check capacity of the cells and balance them before you connect them in series into a pack.
If a charger is constant voltage at 14.4v it will 'cutoff' itself, as the pack will just not take any more charge when full.

You can really overthink and overengineer it if you want, but it really isn't necessary.
__________________
JackB
MiniMPPT Solar Controller
nimblemotors is offline   Reply With Quote
Old 06-10-2014, 14:23   #13
Jd1
Registered User

Join Date: Aug 2011
Location: Victoria, BC
Boat: Catalina 36 MKII
Posts: 1,096
Re: Charging schemes for LiFePO4

I need to revise my earlier post about the House Power BMS system. I was under the impression that the system actually measures voltages in the cell level boards and sends this info back to the main board to allow correct triggering of HVC and LVC. I now have to agree with EVM1024 - the head board determines which alarm to trigger based on pack voltage only and does not care what happens at the cell level. The cell boards trigger the main contactor based on either a low voltage or a high voltage on a cell. The cell level board does not care what the event is, it just drops the main contactor.
The main board voltage levels that trigger an HVC or LVC do so earlier than the cell boards and under normal operating procedures this would be sufficient.
I am not sure I like this arrangement but I can't, at the moment, think of a situation where this method of protection would not be good enough. It certainly is a cost effective design.
__________________
Jd1 is offline   Reply With Quote
Old 06-10-2014, 15:02   #14
Marine Service Provider
 
Maine Sail's Avatar

Join Date: Jan 2007
Location: Maine
Boat: CS-36T - Cupecoy
Posts: 3,061
Re: Charging schemes for LiFePO4

Quote:
Originally Posted by evm1024 View Post
Just went to the bench and measured the levels on the cell boards.

2.8 volts cell loop goes from open to closed
3.55 volts Shunting resistors activated.
3.8 volts cell loop goes from closed to open

2.8 * 4 = 11.2 volts
3.55 * 4 = 14.2 volts (active balancing starts)
3.8 * 4 = 15.2

So if any cell goes below 2. or over 3.8 the pack is disconnected (open cell loop)

The House BMS only sees pack voltage and from the manual:

HVC turn on – 3.6V per cell
HVC turn off – 3.45V per cell
LVC turn on – 2.9V per cell
LVC turn off – 3.1V per cell

But remember that the house bms does not monitor cell voltages. they only report it as cell voltages because you can get a 12 volt or 24 volt versions of the board.

Dimitri has made some recent changes to the HPBMS..

#1 It is now HVC at 14.4V not 14.2V like earlier models. He can do a custom one at 14.2V if you want. I know he just recently did a 14.2V for a guy I know. (mine is 14.2V)

#2 HVC now has some hysteresis programed into it. Older models did not have any hysteresis.


The 3.6V cell boards IIRC begin shunting at 3.55V. The 3.8V cell boards begin shunting at 3.65V IIRC (I use 3.8V boards because I prefer to do any balancing myself.)

*HVC is pack level (this is a warning level on the HPBMS)
*LVC is pack level (this is a warning level on the HPBMS)
*Shunting is cell level
*Pack relay/main contactor is cell level (this is emergency level on HPBMS)

If you stay out of the HPBMS's voltage ranges you will have a nice insurance policy...
__________________
Marine How To Articles
Maine Sail is offline   Reply With Quote
Old 06-10-2014, 15:04   #15
Marine Service Provider
 
Maine Sail's Avatar

Join Date: Jan 2007
Location: Maine
Boat: CS-36T - Cupecoy
Posts: 3,061
Re: Charging schemes for LiFePO4

Quote:
Originally Posted by Jd1 View Post
I have the MC614 regulator but again I am thinking of not changing any of the parameters because the alternator is also used for the lead acid bank. I am thinking that both lithiums and lead acid are charged by the alternator but the lithium has the charge relay that drops out when the bank is full..
__________________

__________________
Marine How To Articles
Maine Sail is offline   Reply With Quote
Reply

Tags
charging, lifepo4

Thread Tools
Display Modes Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
LiFePO4 Batteries - Okay Tear Me Apart ;-) jallum Electrical: Batteries, Generators & Solar 395 03-12-2015 13:19
charging batts with genset/charger, wet, lifepo4 witzgall Electrical: Batteries, Generators & Solar 6 20-10-2013 17:27
Advice w/ Changing Color Schemes BigSteels Powered Boats 6 10-04-2013 15:32
Advice w/ Changing Color Schemes BigSteels Construction, Maintenance & Refit 3 10-04-2013 07:31



Copyright 2002- Social Knowledge, LLC All Rights Reserved.

All times are GMT -7. The time now is 08:03.


Google+
Powered by vBulletin® Version 3.8.8 Beta 1
Copyright ©2000 - 2017, vBulletin Solutions, Inc.
Social Knowledge Networks
Powered by vBulletin® Version 3.8.8 Beta 1
Copyright ©2000 - 2017, vBulletin Solutions, Inc.

ShowCase vBulletin Plugins by Drive Thru Online, Inc.