Lithium batteries and bench power supply

bbbuddy

We've decided to buy lithium cells- 48 of 72ah cells for a 432ah 24v system, and those who have them already have detailed here and elsewhere how to top balance them before installation with a bench power supply.  I know nothing about power supplies so I can't tell if one I buy is "good enough"...

Those of you familiar with the workings of these, can you tell me if this one will be accurate enough to do the job?  I picked this one because it is a 30 amp, which should mean I can run it at at least 20 amps continuous right?  We will only have daylight hours to do the balancing and I don't want it to take days and days..So if the battery cells come in at 30-40% charge, it won't take forever to bring them up like it would with the 5 amp cheaper ones. I plan to hook them up in groups of 16 paralleled to do the balancing.
https://www.amazon.com/gp/product/B01KPBAN6O/ref=ox_sc_act_title_1?ie=UTF8&psc=1&smid=AMH4W1K8OCGMX

At this time this is the only reason we need a power supply, but will use it in the future to check the balance periodically...

jonr

Note that it will only produce 100 watts into 3.5V batteries.   But hopefully balance will already be close.

https://www.orionbms.com/general/pre-balancing-cells/

bbbuddy

Umm, what am I missing? It says it's up to 15v at 30 amps, doesn't it? Or is this not how they work?

Raj174

Are these cells to be used in a solar power system? Have you considered 8 400AH cells? Maintaining cell balance would be easier with fewer cells in a single string, in my opinion.

bbbuddy

Hi Rick, yes I had actually contacted this company regarding the 400ah cells, and they suggested these instead for multiple reasons, including the fact that we wouldn't be "down" if a cell went wonky.  Plus these New Energy cells were put through a very difficult test regime by a fortune 500 company along with 20 others , and these and one other were the only ones who aced the test.

We plan on sets of 6 paralleled, and those 8 sets then in series.  For warranty purposes we have to have at least a battery balancer, they have one that is $21 per series connection and actively balances the bank. A 6 amp system that they haven't had any problems with...whether we actually hook it up or not is still in question, but we do have to have one...

The company is Electric Car Parts in Utah.  Your posts about the lithium batteries and your experience with them are among those  that convinced me this is the way to go for our future.  

bbbuddy

http://www.electriccarpartscompany.com/New-Energy-72Ah-Aluminum-Encased-Battery


Can anyone tell me if this power supply will do the job?
https://www.amazon.com/gp/product/B01KPBAN6O/ref=ox_sc_act_title_1?ie=UTF8&psc=1&smid=AMH4W1K8OCGMX

Raj174

bbbuddy,
Yep, I'm familiar with the cell. I have no experience with lab grade power supplies, but the reviews state that it is really a 20 amp continuous supply. It sounds like that would do the job. I used a 10 amp 3.65 volt Intelligent battery charger, purchased along with my batteries. Did they offer to balance them and box the cells with a BMS? Just asking because balancing 48 cells will take some time, as they will have to be topped off one at a time. My experience has been that cells in series stay well balanced at low charge and discharge rates, but the voltage of parallel strings will tend to drift apart. That is why parallel battery strings are generally avoided in solar power systems if possible. However, if you have a BMS that can maintain balanced strings as well as cells then that might do the trick.

Rick   

bbbuddy

Ah, that's the first I've read that the cells in parallel will drift. 

I'll just have to plan on unconnecting them periodically to check.  Hope to get an idea of how fast it happens pretty quickly..
The balancing system is an active balance system, one per string, no "smart monitoring " with computer readouts...No lvd or hvd, since my two Classic 150s and the Magnum have been operating fine for several years and I have more trust in them than getting new stuff to complicate things...

Plus the amps are such that we probably will never use more than 50% total.  Even our cloudy days usually give us enough to refill batteries.  The really dark days are probably less than 6 per year.  The gen can refill what we use in an hour for those days.

No I didn't ask for them to be balanced and pre wired, I'm not even sure I'll hook the balancer up.  I am getting batteries that are from the same lot and in production order.  They say they perform a check on all of them before sending out.

I did ask about some of the connections I have seen on the internet, where the cells are each connected in parallel and in series at the same time, and was told that it is a waste of time and money to do that.  Hope this info is correct.  One .pdf I read said the smaller cylindrical batteries should be hooked up in a lattice network like that...

karrak

bbbuddy said:

We've decided to buy lithium cells- 48 of 72ah cells for a 432ah 24v system, and those who have them already have detailed here and elsewhere how to top balance them before installation with a bench power supply.  I know nothing about power supplies so I can't tell if one I buy is "good enough"...

Those of you familiar with the workings of these, can you tell me if this one will be accurate enough to do the job?  I picked this one because it is a 30 amp, which should mean I can run it at at least 20 amps continuous right?  We will only have daylight hours to do the balancing and I don't want it to take days and days..So if the battery cells come in at 30-40% charge, it won't take forever to bring them up like it would with the 5 amp cheaper ones. I plan to hook them up in groups of 16 paralleled to do the balancing.
https://www.amazon.com/gp/product/B01KPBAN6O/ref=ox_sc_act_title_1?ie=UTF8&psc=1&smid=AMH4W1K8OCGMX

At this time this is the only reason we need a power supply, but will use it in the future to check the balance periodically...

If you are going to balance the cells by putting them all in parallel and then charging at 30A it could take around 86 (48*72*0.5/30)hours to charge assuming the cells are around 50% full when they are shipped to you.

You could buy something like this Turnigy charger https://hobbyking.com/en_us/turnigy-reaktor-300w-20a-ac-dc-synchronous-balance-charger-discharger-us-plug.html. Set the bank up as an 8p6s battery with the balance leads connected to th supply and charge to 3.4 volts/cell (20.4V). When it has charge to this point put all the cells in parallel and charge to 3.6 volts. This will cut the time down to ~15 (6*72*.5/15) hours and is safe because the charger will not let any of the cell voltages go outside the safe operating range if you have the balance leads attached. You could then put all the cells in parallel and charge to 3.6V.

If the cells are wired up correctly (see this article for more information http://www.smartgauge.co.uk/batt_con.html) I can't see any reason why having cells in parallel will make any difference to how well the battery will stay in balance. If anything I would have thought that having cells in parallel will average out any differences between cells.

I only manually balance my battery. When it was new I had to adjust the balance fairly regularly. Now it is nearly four years old I am only needing to balance it once or twice a year.

If you have to have some sort of battery balancing and monitoring for your warranty I would suggest you look at this http://123smartbms.com/ as it has both the balancing and monitoring functions. I don't think that having just balancing boards without any cell monitoring is a good idea. If one of the cell balancers malfunctions it can damage your battery without you knowing about it.

Simon

jonr

Note that balancing cells does not require fully charging them.   All you have to do is bring them up to the voltage of the highest one.  Cells/strings in parallel stay balanced, cells in series can drift.

westbranch

I just reviewed the specs...  is that 4 -6 % / 30 Days  Self Discharge normal for LiFePO4 cells?  Seems pretty high to me.  At that rate over winter, say 4 months shutdown, it could be as low as ~ 75% SoC?

Dave Angelini

LG Chems LiFePO4 state operating -10 to 45 C.  However, best warranty and lowest self discharge are 15 to 30 C. I can't find that spec westbranch but I know I saw it.....Storage temp -30 to 60 C but definitely lower warranty and higher discharge in this range. 
Keep them the temp that humans like

bbbuddy

Thanks for your replies!
I have to purchase at least a balancer for warranty, but that doesn't mean I have to hook it up...  

We have purchased the batteries and a 24 volt 40 amp lithium charger.   We'll charge them in the house on a table and periodically check the cells.  We can group the cells by voltage when we start, then watch them as the voltage goes up.  Wait an hour, check the individual cells, charge some more.  Should be able to get them up to around 3.4 and then bring the highest ones down to the lower ones, then charge some more...
My goal would be to get them all to 3.55 together and rest them. If most are there but a few aren't then bleed off voltage until they all match the lowest ones.

After we hook them up to the Midnight Classics we will keep the voltage low for a few days then check the balance and see what they are doing...

You guys here that already have been discussing them have been a goldmine of info and I appreciate it greatly!

I'm going to have to re-read the Trimetric directions so I can reprogram it...  Will 98% work for peukert do you think?

karrak

I would be very careful charging the cells in series without any individual cell monitoring. The cell voltages can go from 3.4v to > 3.6V in a matter of minutes depending on the charge rate.
Due to the hysteresis (difference) in cell voltage depending on whether they are being charged, resting, or being discharged I am not sure that your method will be the best to balance the battery. I think it would be better to get all the cells to roughly 3.4V then put then all in parallel and take them to 3.6V and let the charge voltage taper down to less than C/100.

Have you already bought the Trimetric? I have logged the overall coulomb (current) efficiency of my battery around 99.5%

Simon

jonr

> The cell voltages can go from 3.4v to > 3.6V in a matter of minutes

Another good reason not to charge a series string all the way to 3.6V/cell.

bbbuddy

Ok, that's why I was going to stop at 3.4 but I could stop earlier...Then parallel.

Yes I've had the Trimetric for years. That's why I need to reread the directions to reset it.  It's been performing perfectly. Coulomb at 99.5?
Wow.  

bbbuddy

So I guess I need to buy a power supply after all...to charge to 3.6 accurately.

PNjunction

Your choice of supply shows that you need a bit more info.  You don't just slap a battery onto it and start twisting knobs.

A) Most variable lab supplies are not designed or built to charge batteries - so try to get one that is designed to do that.  The ham-radio type supply is not expecting a voltage at it's output terminals from a battery, and can blow the crowbar circuits.  Poof - there goes your new ham radio type supply.

Mastech does have a line of variable supplies that will accommodate battery charging - the "EX" line I believe.  Check with the manufacturer and read the model description.

C) Those that can do so *safely* with their own supplies usually follow this setup:  BE SAFE AND NOT STUPID.  NO Watches, rings, and any other of the usual safety mechanisms need to be in place.  A buddy or secondary person should be available.

1) With NO BATTERY attached, turn both the voltage and current down to zero.  Turn on supply with no shorted clamps.  Supplies not designed for charging may go poof at this point if attached to battery, so don't attach to battery yet!

2) Use an *accurate* voltmeter, and measure the voltage at the clamps.  Display is not accurate enough for LFP.

3) Turn the voltage to the CV value (ie, 3.6v in this case for a single cell) *as measured on the accurate voltmeter* at the clamps.

4) If there is no voltage reading despite turning the knob, take the current just *slightly* off the zero setting to get an indication of voltage.

5) Readjust voltage to your desired CV value as measured at the clamps.

6) Attach clamps to battery cell.  Mind your polarities!!  Just in case, GOGGLES or FACESHIELD should be used.

7) Adjust current to desired value.

8) DONT TWEAK THE VOLTAGE KNOB .  Voltage will read low until the battery itself comes up to the CV value, as is normal with a cc/cv routine. If you want to watch the voltage, do so now with the voltmeter on the battery terminals!  Supply display is not totally accurate.

9) Once the battery reaches the CV voltage, watch the voltage hold itself, and the current fall away.  Typically, a full charge includes not only reaching CV (3.6v here), but also a C/20 drop in current.  For a 72ah cell, that would mean you are done when current has dropped to 3.6 Amps.

That being said, wouldn't it be easier to just invest in a "single cell" charger (for LIFEPO4 voltages!)

http://www.batteryspace.com/Smart-Charger-20A-for-3.2V-1cells-LiFePO4-Battery-Pack.aspx

There are similar items around.  Make *sure* it is for LFP, (3.2v nominal, and 3.6/3.7v CV) and not other lithium chemistries with different voltages.  No need to exceed 0.3C of the cell.  Don't choose one much smaller than 0.1C current rating of the cell.

karrak

bbbuddy said:We have purchased the batteries and a 24 volt 40 amp lithium charger.  

What is the make and model of the charger? If it is a real lithium charger I would have thought it would have some sort of individual cell monitoring.

Simon

bbbuddy

PNJunction, you're right, of course, I've never used a power supply.  I HAD read enough to know to turn it on before attaching a load, lol.
Thanks for the post about how to use it though!  Your post convinced me to go with a regular lithium charger...
That charger you linked to is the one I would buy, I've looked at it before, but is out of stock...

Simon the charger is this one: http://www.electriccarpartscompany.com/12kW-1200-Watts-Model-L-Lithium-or-Lead-Acid-Intelligent-Battery-Charger.  Set at the factory for either lithium or lead acid.  

My plan (please shoot it full of holes if needed) is to hook up 8 cells in series and charge them while monitoring with a multimeter.  The charger goes to 3.65. I would stop as soon as the first cell gets to 3.65.  Then I would hook up the next 8 cells and while they are charging the first 8 will be connected to the active cell balancer(6 amp balancer).  The balancer will bring down the high cell...
Keep doing this 4 more times until all 48 are charged.

Assemble them in the 6p8s form and attach the balancers for a day, so each parallel set can balance with the others, then replace my lead acid batteries with the new lifepo4 and reset my two midnight classics and the Trimetric.  The classic will be set to 28v for absorb and lower than that for float.

I also will set the lvd on the Magnum pae to 24.4 in the beginning.  So no chance of a problem on the low end...

How does that sound?

mike95490

Before charging or testing batteries, it's been advised to bring them all into balance together.   Check each one with a volt meter, make sure none are too far off from the average, and then hook them all in parallel and let sit over night,   When you start charging the next day, ALL cells will be starting from the same voltage/SoC  and capacity differences will appear as you approach charge completion,

Raj174

Mike is correct, each battery needs to be charged individually to 3.65V and then wire all the cells together in parallel for about 8 hours. This will ensure all the cells are balanced when the pack is assembled.
I used #10 bare copper ground wire to parallel my cells. Just be careful not to short anything. Here's a pic. 



 

karrak

bbbuddy said:

PNJunction, you're right, of course, I've never used a power supply.  I HAD read enough to know to turn it on before attaching a load, lol.
Thanks for the post about how to use it though!  Your post convinced me to go with a regular lithium charger...
That charger you linked to is the one I would buy, I've looked at it before, but is out of stock...

Simon the charger is this one: http://www.electriccarpartscompany.com/12kW-1200-Watts-Model-L-Lithium-or-Lead-Acid-Intelligent-Battery-Charger.  Set at the factory for either lithium or lead acid.  

I couldn't find any technical information on this supply. Does it charge to 29.2V (3.65V/cell) then stop charging or hold the voltage or drop down to a lower float voltage? Does it monitor the individual cell voltages?

Unless there is something special about the lithium supply, if it were a choice of this one or the variable one you first mentioned I personally would have chosen the variable supply.

Modern switch mode variable supplies like the one you mentioned in your first post should not blow up if you connect a battery up to them regardless of what voltage they are outputting if the battery voltage is less than the maximum output from the supply and you hook the battery up to the supply correctly. There is a good chance that the supply will be damaged if you hook the battery negative to the positive output from the supply and the battery positive to the negative or the battery voltage is greater than the maximum output from the supply.

Having said that, it is good practice to set the output from the supply just below the battery voltage before you connect it up. If the displayed reading on the supply is inaccurate you can use an accurate voltage meter to check and adjust the output.

My plan (please shoot it full of holes if needed) is to hook up 8 cells in series and charge them while monitoring with a multimeter.  The charger goes to 3.65. I would stop as soon as the first cell gets to 3.65.  Then I would hook up the next 8 cells and while they are charging the first 8 will be connected to the active cell balancer(6 amp balancer).  The balancer will bring down the high cell...
Keep doing this 4 more times until all 48 are charged.

Assemble them in the 6p8s form and attach the balancers for a day, so each parallel set can balance with the others, then replace my lead acid batteries with the new lifepo4 and reset my two midnight classics and the Trimetric.  The classic will be set to 28v for absorb and lower than that for float.

I also will set the lvd on the Magnum pae to 24.4 in the beginning.  So no chance of a problem on the low end...

How does that sound?

Not sure if you have balanced your battery yet but hope that everything went well if you did.

I am still concerned that you intend to charge up to 29.2V? with this supply with the only protection against any particular cell going above its safe operating voltage is you with your multimeter.

With the supply you have purchased and your 6A active cell balancers I would hook the battery up in your final 6p8s configuration with all the cell balancers connected and let it charge up while monitoring it. If any of the cells get above 3.7V halt the charging and let the balancers balance the battery. When the cell voltages have all balanced start the charger again.If you are going to do this I would recommend you purchase a Cellog8 and set the alarm to go off at a voltage of 3.45V.

28V is fine for your absorb voltage, could go lower to 27.6V (3.45V/cell) with an absorb time of 1/2 an hour or  a termination current of 1.4A. This will charge the battery to around 99% full. Setting the float voltage to 26.8V (3.35V/cell) will keep the battery >95% full while the sun is still shining.

If you are charging the battery but not using it you should not hold it at 99% full.

There is some good information on balancing in parallel here http://liionbms.com/php/wp_parallel_balance.php

Simon

jonr

I agree.  There is little to be lost and a lot to be gained by being conservative with the max voltage.

bbbuddy

Well we're operational! Sorry for abandoning my thread, I got sidetracked when the batteries arrived. We set them up on a table in the living room and first I took the voltage of all the cells. They varied by .125, the lowest was 3.331 the highest was 3.456. Much higher voltage than I expected since they are supposed to be shipped well below full...

Then I paralleled them and had to wait a bit for the charger to come, because the one they sold me wasn't in stock. I paid a bit more for one in stock: http://www.electriccarpartscompany.com/24V-256V-292V-23A-br-Lithium-Battery-Intelligent-Charger-br-LiFePO4-LFP-Intelligent-Charger-br-CE-Certified-br-USA-Stock_p_206.html

Turns out they shipped without any way to hook them up, so we again had to wait for connectors to ship...

The balancers were not the ones I asked for, had to send them back.... yikes, by this time I was getting ticked....The only ones I wanted are out of stock, so I won't get them for several weeks...

Finally the lifepo4 intelligent charger connectors arrived. I had the batteries in paralleled blocks of 6, and then in series. Double check the voltage, hook up the charger, keep taking the voltage of the paralleled blocks...nothing out of the ordinary. But then, a big surprise, the charger went to float after only putting 10 amps into the battery block. I kept checking voltages, and they varied by .102 right after, and this went down to .064 after 5 hours.

So, yesterday we took out our lead acid batteries and installed the lifepo4 in a big cooler. The reason for this is because they are out in a small "power house" for now and I wanted to control the temperature. They are on a plywood platform inside the cooler with a 23watt incandescent shop light underneath...so I can put it on a timer to be on for maybe 10 minutes every hour...but last night at midnight it was 62 inside the cooler, and it got down to 27, but inside the cooler it was 51 this morning. So I guess I don't need to worry about them freezing, plus in summer I can open the cooler at night then close it during the day too keep the battery temps down (it gets cool at night here even on really hot days because we're a mile high)

The Midnite Classics showed 1.4kwh this morning when they went to float, so that's all we used last night...14% of the capacity...I turned the inverter off when I went to bed...because I've always done that. Tonight it stays on - obviously we have plenty of power.

I have checked the voltages several times today and they are all within a few hundredths of each other.

I set the Classics to absorb at 27.6, to absorb no more than an hour, and the float is 27.3 for now...

We sure don't have to worry about them getting to the bottom of the curve.

westbranch

BBB, could you please add your battery bank specs to your signature line..? thanks, it helps to quickly  compare capacity with use.

the 1.4kWh was for the fridge, Inverter, and ????
Sounds like the fridge is the major load, very similar to our per day us of ~1.2 kWh...

2 Classics? how big is your array/s?

mike95490

LFP batteries should not need more than a couple minutes at your Absorb voltage.  Once the bank gets to your charge voltage, you should not let it sit there charging, you should back it off to the Float voltage.. Or you run a real risk of over charging the cells. Lithium does not need an absorb period like lead acid,  Leaving them at the higher absorb voltage can quickly overcharge them.
 But, hey, I don't own LFP batteries so maybe you have a better plan than I do.

bbbuddy

I don't know what happened to my sig line, it used to be there...I'll have to re do it...We have 3100 watts solar and a Magnum 4024PAE inverter. Two classics because it's more solar than one can handle and I like having backups just in case anyway...

Mike, 27.6 is 3.45 per battery, which is not fully charged. They would "eventually" get full at 3.45 from what I understand, but not in an hour, plus I have things running all the time anyway.
I can back off on the absorb time if everyone thinks it's necessary...

Tonight I ran the dryer late, and cooked dinner in an electric pressure cooker, etc and got the batteries down to 75% according to the Trimetric, and I checked the batteries, they all read within a few milliamps of each other! So far, so good.

mike95490

if you had run a 48V system you could use only 1 classic.

bbbuddy

Yes, but as I said, having a backup helps me rest easy...We are fairly remote, and sometimes the 600 foot wide dry river bed we have to drive through to get here runs 2 feet deep for days, and is unmanageable for weeks because it is sticky red clay when wet. BUT we have tons of privacy!

I prefer to have backup, and am giving serious thought to getting another inverter, not because I need it, but just to have "in case".

I haven't had any problems with our magnum, but it was built before the company was sold, and now I have seen multiple complaints about the new ones. Guess their quality went downhill? I may just buy a cheaper inverter to have on hand, rather than incorporate one now.

Dave Angelini

I don't think the concept of buying a cheap inverter would would make me rest easy