KingArthur
Well-known member
- Joined
- Aug 12, 2016
- Messages
- 70
- Reaction score
- 0
BradKW said:
I won't be installing anything in my van until I'm fully convinced what I've created is sane and safe.
FLA vs SLA vs Li-Ion vs LiFePo4 and air conditioning
- Thread starter KingArthur
- Start date
Help Support Van Living Forum:
I won't be installing anything in my van until I'm fully convinced what I've created is sane and safe.
And getting even 10% of the AH you need is a huge time-consuming DIY project.
Get straight into the more interesting learning bits with a 4-pack of lower-AH prismatics, isn't too bad a cost, monitor what's going on, double-check your DMM, very unlikely you'll murder them.
And they'll make a nice portable power pack one day, maybe last for decades.
Get straight into the more interesting learning bits with a 4-pack of lower-AH prismatics, isn't too bad a cost, monitor what's going on, double-check your DMM, very unlikely you'll murder them.
And they'll make a nice portable power pack one day, maybe last for decades.
jonyjoe303
Well-known member
- Joined
- May 18, 2017
- Messages
- 678
- Reaction score
- 83
Since last year I have built several battery packs using the 18650 cells. You need to start small and slowly work up to bigger packs. Start with 9 cells.You can build a 3s 3p battery pack. This is small so you can use it to test your BMS and also soldering and how to connect the bms connectors. A 3s battery pack will be 11.1 volts, fully charge to 12.6 volts. Its at 0 percent at around 10 volts. I find it will run everything that ran on my agm,including inverters.
From 9 cells I build a 24 cell pack then moved up to 84 cells(44 amp hour). I then build another 84 cell(50 amp hour) pack and connected both packs together to make a 94 amp hour battery pack. I've been using this 94 amp hour lithium battery for the past year as a house battery.
I kept buying more 18650 batteries until I had nearly 500, to build an even bigger battery. I use those to build two smaller power packs one with 60 cells (37 amp hour) and another with 90 cells (65 amp hour).
Just as an example of some costs, the 94 ah lithium cost me less than 300 dollars (comparable 12 volt 100ah lithium is over 1299.00), the 37 amp hour cost me about 140 dollars ( a goal zero 35 ah lithium costs 700 dollars). Your main expense are the batteries, I have been able to buy 80 batteries for as low as 60 dollars, you can find some bargains. On ebay you can type "lot of 18650" and see what you find, I prefer to buy the ones that are tested and have the mah hour ratings written on them. Avoid the untested ones, if you buy 60, many will be low voltage or 0 voltage. Anything over 1800 mah are good, but I use batteries as low as 1200 mah on my 94 ah battery.
When you build them, put a 4 or 5 amp fuse on every cell, if one cell shorts out it blows the fuse. If the cell overheats it vents. As long as you do a good job on all your wiring, you minimize hazards. Test your BMS to make sure it will prevent the battery from being overcharged and undercharged, I use a smaller 9 cell pack to test all my BMS. Some bms wont shutdown the charging when the battery goes out of balance, that's why I always test them. Some 5 dollar bms work better than the more expensive bms.
On your battery pack, make sure you put a 3 in 1 lcd meter, this will tell you exactly what each row of batteries voltage is, if it goes out of balance you know right away.
Make sure all the batteries are at the same voltage when you start soldering them together. For connectors I been using the xt60 (can handle 60 amps) for more power output you can use the xt90. You can use batteries of different mah to build a battery pack, just as long as each row of batteries is the same Amp hour. If on a 3s battery row 1 is 8
amps, row 2 is 6 amps, row 3 is 9 amps, you have a 6 amp battery pack, the bms will shut off the battery at the lowest amp one. When you got all your batteries ready, go to the repakr website, enter the mah of all your batteries and it will tell you what row to put each battery in. The best charger to test and charge individual 18650 is the opus btc-3100, it can charge/discharge/ and check capacity of 4 batteries at a time charges/discharges at 1 amp each battery.
Don't be scare too much of the lithium, just don't use the ones that wont hold charge or heat up too much when charging. Namebrand cells are the safest to use. Just start small and work your way up. Lithium is the only realistic way you will ever run a air conditioner in your van, and building a 300/400 ah lithium is not unrealistic.
On youtube there's a video "lithium battery test versus gel " where they test a 100 ah lithium against a 100 ah gel and a 220 gel test in a worst case scenario where you quickly drain the battery like an air conditioner would. The lithium easily beat both of them, they were able to get 94 ah of usable power out of the lithium. The 100 gel had 31 ah of usable power and the 220 ah gel had 73 ah of usable power. Since I been using lithium, I would never go back to agm.
From 9 cells I build a 24 cell pack then moved up to 84 cells(44 amp hour). I then build another 84 cell(50 amp hour) pack and connected both packs together to make a 94 amp hour battery pack. I've been using this 94 amp hour lithium battery for the past year as a house battery.
I kept buying more 18650 batteries until I had nearly 500, to build an even bigger battery. I use those to build two smaller power packs one with 60 cells (37 amp hour) and another with 90 cells (65 amp hour).
Just as an example of some costs, the 94 ah lithium cost me less than 300 dollars (comparable 12 volt 100ah lithium is over 1299.00), the 37 amp hour cost me about 140 dollars ( a goal zero 35 ah lithium costs 700 dollars). Your main expense are the batteries, I have been able to buy 80 batteries for as low as 60 dollars, you can find some bargains. On ebay you can type "lot of 18650" and see what you find, I prefer to buy the ones that are tested and have the mah hour ratings written on them. Avoid the untested ones, if you buy 60, many will be low voltage or 0 voltage. Anything over 1800 mah are good, but I use batteries as low as 1200 mah on my 94 ah battery.
When you build them, put a 4 or 5 amp fuse on every cell, if one cell shorts out it blows the fuse. If the cell overheats it vents. As long as you do a good job on all your wiring, you minimize hazards. Test your BMS to make sure it will prevent the battery from being overcharged and undercharged, I use a smaller 9 cell pack to test all my BMS. Some bms wont shutdown the charging when the battery goes out of balance, that's why I always test them. Some 5 dollar bms work better than the more expensive bms.
On your battery pack, make sure you put a 3 in 1 lcd meter, this will tell you exactly what each row of batteries voltage is, if it goes out of balance you know right away.
Make sure all the batteries are at the same voltage when you start soldering them together. For connectors I been using the xt60 (can handle 60 amps) for more power output you can use the xt90. You can use batteries of different mah to build a battery pack, just as long as each row of batteries is the same Amp hour. If on a 3s battery row 1 is 8
amps, row 2 is 6 amps, row 3 is 9 amps, you have a 6 amp battery pack, the bms will shut off the battery at the lowest amp one. When you got all your batteries ready, go to the repakr website, enter the mah of all your batteries and it will tell you what row to put each battery in. The best charger to test and charge individual 18650 is the opus btc-3100, it can charge/discharge/ and check capacity of 4 batteries at a time charges/discharges at 1 amp each battery.
Don't be scare too much of the lithium, just don't use the ones that wont hold charge or heat up too much when charging. Namebrand cells are the safest to use. Just start small and work your way up. Lithium is the only realistic way you will ever run a air conditioner in your van, and building a 300/400 ah lithium is not unrealistic.
On youtube there's a video "lithium battery test versus gel " where they test a 100 ah lithium against a 100 ah gel and a 220 gel test in a worst case scenario where you quickly drain the battery like an air conditioner would. The lithium easily beat both of them, they were able to get 94 ah of usable power out of the lithium. The 100 gel had 31 ah of usable power and the 220 ah gel had 73 ah of usable power. Since I been using lithium, I would never go back to agm.
Building your own from cylindrical may be a nice hobby, if you're motivated and have a lot of time. Not something I would do nor recommend.
Yes getting started with LFP the normal way is very expensive up front compared to lead. Take care of them they may well be cheaper than "fancy lead" in the long run. Screw things up you lose a lot.
Yes getting started with LFP the normal way is very expensive up front compared to lead. Take care of them they may well be cheaper than "fancy lead" in the long run. Screw things up you lose a lot.
VanKitten
Well-known member
I have begun the project to build LiFePo4 battery bank. I and a couple of hobbyist have joined ranks and will do this.
The lithium that have caused the fires are not LiFePo4.
Anyway...everyone who has done this feels that the effort was more than worth the pay off. The biggest issues that people with LiFePo4 have experienced is that trying to drop them in "plug and play" has resulted in a greatly shortened life for the battery. Solving the BMS-Charge controller compatibility issue is the big hurdle Blue sea, outback, and the like are not building for this compatibly...yet.
I know I am on the leading edge of this...but, I have found the BMS and controller I will use. I'm waiting for the cells and my project partners to be ready to begin.
The lithium that have caused the fires are not LiFePo4.
Anyway...everyone who has done this feels that the effort was more than worth the pay off. The biggest issues that people with LiFePo4 have experienced is that trying to drop them in "plug and play" has resulted in a greatly shortened life for the battery. Solving the BMS-Charge controller compatibility issue is the big hurdle Blue sea, outback, and the like are not building for this compatibly...yet.
I know I am on the leading edge of this...but, I have found the BMS and controller I will use. I'm waiting for the cells and my project partners to be ready to begin.
It really makes you consider getting something like a used Nissan Leaf with its 24kW-hr battery and all of the charging systems, plus 12V at 120 amps available to use. They are around $8,000 with 3 years left on the battery warranty. I am not sure I can build a 24kW lithium bank for $8,000 much less get a car out of it too.
Did I miss something?
A small inverter generator would run that AC unit just fine..Many van expediters have this exact setup. It works well.
The main problem as I see it is that we simply dont have enough square feet on the roof of a typical van to provide the needed watt/hr of solar to recharge the battery bank from a large overnight drain.
We are all pretty much limited by the low efficiency of present-day solar panels and the cost for them, and the cost of high capacity batteries.
It's physics, man.
In your research be sure to look into electric APU units....they are used in the trucking industry and are made for commerical trucks, and will have a large alternator recharging the batteries during a 10-14 hour daytime driving shift, BUT...the overnight AC powered from batteries method is well known.
Seriously, you can buy a LOT of gasoline for a small genset compared with spending mega bucks on a roof covered with solar panels and 500 pounds of flammable batteries under your bed.
Or so it seems to me.
A small inverter generator would run that AC unit just fine..Many van expediters have this exact setup. It works well.
The main problem as I see it is that we simply dont have enough square feet on the roof of a typical van to provide the needed watt/hr of solar to recharge the battery bank from a large overnight drain.
We are all pretty much limited by the low efficiency of present-day solar panels and the cost for them, and the cost of high capacity batteries.
It's physics, man.
In your research be sure to look into electric APU units....they are used in the trucking industry and are made for commerical trucks, and will have a large alternator recharging the batteries during a 10-14 hour daytime driving shift, BUT...the overnight AC powered from batteries method is well known.
Seriously, you can buy a LOT of gasoline for a small genset compared with spending mega bucks on a roof covered with solar panels and 500 pounds of flammable batteries under your bed.
Or so it seems to me.
A gasoline generator can be just as deadly as batteries if used incorrectly.
And it is a constant drone of noise unlike solar which is very quiet.
I always hate being at a camp spot and someone running their generator well past the posted quiet hours.
And it is a constant drone of noise unlike solar which is very quiet.
I always hate being at a camp spot and someone running their generator well past the posted quiet hours.
IGBT said:
Yep, downsides to everything....including the 35 gallons of gasoline packed under the floor of our vans...
I didnt see where he mentioned staying in quiet campgrounds but that was my oversight if it was mentioned. But inverter generators are very quiet, and can be made almost silent if installed in a vented, plywood enclosure. The compressor on the AC unit would likely be as noisy.
I just dont ever camp anywhere that requires me to run AC all night long but I know that others do. If I did, I'd pay the extra few dollars for AC hookups.
AC hookups are not always available. On the property we bought here in eastern Washington, it probably would cost $3,000 to $5,000 to hook into the grid (several power poles needed to get the power from the road to our camp spot). You also then have the monthly fee, even when you are not using the power.
In the summer here it is going to be some days of 95 degrees and nights well over 80. I am trying to pre-plan our power usage and ability for this. I know we are going to want some A/C during the evening and early night. We for sure can run our Honda EU2000i since we are on 28 acres but it is still loud and you have to get up to turn it off or put more fuel in it. I am considering a bit more solar and a few more batteries to give us that extra 2000 watt-hr for running the air conditioner for about 3 hours past sun down (our little A/C uses about that amount in 3 hours.
In the summer here it is going to be some days of 95 degrees and nights well over 80. I am trying to pre-plan our power usage and ability for this. I know we are going to want some A/C during the evening and early night. We for sure can run our Honda EU2000i since we are on 28 acres but it is still loud and you have to get up to turn it off or put more fuel in it. I am considering a bit more solar and a few more batteries to give us that extra 2000 watt-hr for running the air conditioner for about 3 hours past sun down (our little A/C uses about that amount in 3 hours.
IGBT said:
Ok in your situation would it not be possible to set up bank of ground-mounted solar panels and a wind turbine, and a small shed with the batteries, controller, and inverter?
Another solution might be to put the generator 100 feet away with a really long, heavy duty shore-power cord?
28 acres is a fair amount of room.
KingArthur
Well-known member
- Joined
- Aug 12, 2016
- Messages
- 70
- Reaction score
- 0
Thanks for all the replies everyone! I haven't visited this post in awhile...
At this point I'm still leaning towards getting this AC to run on 18650's. Unfortunately I've got about 15 other projects going on at once, I just had to put 700 bucks into my van for inspection, so things are going slowly... but hey, maybe I'll get this worked out for summer 2018, and that's just fine with me. At this point I'm looking at it as more of a challenge and a labor of love. Whether I'm doing that because I'm crazy or because it justifies my choice, I don't know. Probably somewhere in the middle
I'm hoping to at least by the end of this year feel like I'm living comfortably in my van rather than being a mad scientist that sleeps inside a rattly moving unfinished science experiment. Such is life when you're an 'idea person' I suppose...
Anyway, I've started with the cells that I had lying around from a previous 18650 project. I've capacity tested about 300 of them and have ended up with around 170 cells that are above 1500 mAh, which is my arbitrary cutoff. My average capacity across the board is looking to be about 1850 mAh per cell at 3.7v, so I'm fairly pleased with that. I've bought some 18650 cell holders off amazon and built what I consider to be my first module in my pack.
So here are my thoughts so far... I think eventually when things start to really develop I'm going to make a new thread because this, even right now, is beginning to fall far outside the scope of the topic of this post so I don’t want to have it run off the rails too far, but since we're here now…
Firstly my plan is to put the cells in 4s, not 3s. The reason for this is because of the lower amount of capacity loss I will suffer from having the cells in this arrangement. I look at it this way. Fully discharged to safe discharge capacity, a single cell will be at 3v. In 3s, fully discharged the cells will be at 9v. This is 1.5v too low to run almost all inverters, and is not a healthy voltage to run most things that run at 12v. When I was experimenting with some throw-away marine batteries, I noticed that when they fell below 10.2ish volts or so, my compressor fridge would begin to ‘hiccup’, my LED lights would flicker and not shine the right color, my inverter of course wouldnt turn on, and my fan ran slow. I don’t think the electronics like running at a voltage below 11v. So if an 18650 cell has a useable voltage range of 3,3.7,4.2, that means that my useable range in a nominal 12v system per cell is going to be 3.6,3.7,4.2. I’m losing over half a volt of the batteries capacity, and that equates to about a 50% reduction in useable cell capacity depending on discharge characteristics. To my eyes that makes using 3s incredibly inefficient.
The problem with 4s is similar just in reverse. However, to make a long story short it isn’t as bad. In 4s, a pack would have voltages of 12,14.8,16.8. The only hurdle is that I would need to get an inverter that allows a higher and obscure input voltage upper limit over the normal 15v that it seems is almost the industry standard. I’ve only found one inverter that gets close. ( http://www.samlexamerica.com/products/ProductDetail.aspx?pid=503 )It allows an input voltage of 16.5v. As far as I’m concerned, I won’t be programming my charge controller to charge the cells to 4.2v anyway, because that results in capacity degradation. So at my upper limit of 4.1v per cell or maybe less, I will be taking a .1 or .2v capacity loss hit instead of a .6v capacity loss hit. I’ve made a small test pack out of 8 of these cells, and I’ve run all my current appliances off those cells at full voltage, and they seem to do just fine. Maybe I’ll burn some stuff out, maybe I won’t. So far I’m feeling lucky
So I’ve decided that at this point I’d like to have my packs arranged as 1s80p, then those individual packs in 4s4p. I’m planning on having 16 packs in total. That would be a battery of 1280 cells. At an average of 1850 mAh per cell, that would give me a bank with a nominal voltage of 14.8v and an amp hour rating of 592, according to my unverified math.
And I think I can even fit them under my floor!
This is the point where you call me crazy and go browse some other post….
I was originally thinking to put the packs in 4s20p, but that would essentially do nothing but increase complexity and increase my need to balance smaller groups of cells. As long as the cells are in parallel, they do not need to be balanced. Only cells in series need to be balanced, so it seems the simple solution to me is to put as few groups of cells in series as possible.
So there’s a lot of soldering to do…
I hate soldering. I’m so bad at it I don’t think I’d last a day trying to get these cells soldered to a point where I can actually trust the connections. I’ve also read that these batteries shouldn’t be soldered anyhow, but it seems that many people do it without issue, so perhaps that isn’t true. I’ve always been one to err on the side of caution however, so I’m going to spot weld mine with a battery spot welder and nickel strips. This however, leaves the problem of not being able to fuse the cells individually, which I think is absolutely vital especially since I will indeed literally be sleeping on top of them. I’ve thought of a possible solution but I have yet to try it out…
I’m thinking of taking the individual nickel strips and essentially just drilling/filing/cutting some of the strip away in order to make it thinner, and then testing that strip with a power supply to get to the exact width where the strip will break at 5-6A. After I find that exact width, I can stack all my needed strips in a jig of some kind and cut them all at the same time to the needed width, and then proceed to spot weld them as normal. Maybe someone who is crazy enough to care about this idea can tell me if they think it's a good one or not.
So now we get to the really, really, really interesting part.
Dun
Dun
DUNNNNN
The BMS.
Because I am a complete masochist and am utterly unaware of my own limitations, I have decided I want to program my own BMS. I’ve bought myself an Arduino and am currently learning how to program it, and I think without an incredible amount of effort I’m going to be able to write myself at least a rudimentary management program, especially with the help of the online Arduino community. I have a few reasons for choosing this route. Firstly, I want to understand what my BMS is doing completely. I don’t really trust what one of these ebay BMS’s is going to do for me… are the components accurate, are they rated for the voltage I want to use? Are they safe? Do they need additional components to work correctly? I want to understand this system from the ground up. There’s others I have seen online who have made battery management systems with Arduino, so I don’t think it’s terribly beyond me, but I guess we’ll find out eventually won’t we? I think my saving grace is that if I do this right, I’m only going to have to balance a pack of 4s instead of a pack with 16 groups of 4s. That will reduce the complexity of this project enormously.
The last hurdle would then be getting my solar controller to play nice with the BMS. It is a programmable solar controller, and lithium ion batteries are actually very very simple to charge (as in they don’t need to be floated, or bulked, or whatever else) so I don’t think this leg of the journey will be quite as rigorous as the rest of it.
So at this point these are my tentative plans, subject to change at any moment due to necessity or loss of interest. I really value all the input I’ve received so far, and I’ve read all the replies and I thank you for them with sincerity. Hopefully next time I post I have some good news for you all!
At this point I'm still leaning towards getting this AC to run on 18650's. Unfortunately I've got about 15 other projects going on at once, I just had to put 700 bucks into my van for inspection, so things are going slowly... but hey, maybe I'll get this worked out for summer 2018, and that's just fine with me. At this point I'm looking at it as more of a challenge and a labor of love. Whether I'm doing that because I'm crazy or because it justifies my choice, I don't know. Probably somewhere in the middle
I'm hoping to at least by the end of this year feel like I'm living comfortably in my van rather than being a mad scientist that sleeps inside a rattly moving unfinished science experiment. Such is life when you're an 'idea person' I suppose...Anyway, I've started with the cells that I had lying around from a previous 18650 project. I've capacity tested about 300 of them and have ended up with around 170 cells that are above 1500 mAh, which is my arbitrary cutoff. My average capacity across the board is looking to be about 1850 mAh per cell at 3.7v, so I'm fairly pleased with that. I've bought some 18650 cell holders off amazon and built what I consider to be my first module in my pack.
So here are my thoughts so far... I think eventually when things start to really develop I'm going to make a new thread because this, even right now, is beginning to fall far outside the scope of the topic of this post so I don’t want to have it run off the rails too far, but since we're here now…
Firstly my plan is to put the cells in 4s, not 3s. The reason for this is because of the lower amount of capacity loss I will suffer from having the cells in this arrangement. I look at it this way. Fully discharged to safe discharge capacity, a single cell will be at 3v. In 3s, fully discharged the cells will be at 9v. This is 1.5v too low to run almost all inverters, and is not a healthy voltage to run most things that run at 12v. When I was experimenting with some throw-away marine batteries, I noticed that when they fell below 10.2ish volts or so, my compressor fridge would begin to ‘hiccup’, my LED lights would flicker and not shine the right color, my inverter of course wouldnt turn on, and my fan ran slow. I don’t think the electronics like running at a voltage below 11v. So if an 18650 cell has a useable voltage range of 3,3.7,4.2, that means that my useable range in a nominal 12v system per cell is going to be 3.6,3.7,4.2. I’m losing over half a volt of the batteries capacity, and that equates to about a 50% reduction in useable cell capacity depending on discharge characteristics. To my eyes that makes using 3s incredibly inefficient.
The problem with 4s is similar just in reverse. However, to make a long story short it isn’t as bad. In 4s, a pack would have voltages of 12,14.8,16.8. The only hurdle is that I would need to get an inverter that allows a higher and obscure input voltage upper limit over the normal 15v that it seems is almost the industry standard. I’ve only found one inverter that gets close. ( http://www.samlexamerica.com/products/ProductDetail.aspx?pid=503 )It allows an input voltage of 16.5v. As far as I’m concerned, I won’t be programming my charge controller to charge the cells to 4.2v anyway, because that results in capacity degradation. So at my upper limit of 4.1v per cell or maybe less, I will be taking a .1 or .2v capacity loss hit instead of a .6v capacity loss hit. I’ve made a small test pack out of 8 of these cells, and I’ve run all my current appliances off those cells at full voltage, and they seem to do just fine. Maybe I’ll burn some stuff out, maybe I won’t. So far I’m feeling lucky
So I’ve decided that at this point I’d like to have my packs arranged as 1s80p, then those individual packs in 4s4p. I’m planning on having 16 packs in total. That would be a battery of 1280 cells. At an average of 1850 mAh per cell, that would give me a bank with a nominal voltage of 14.8v and an amp hour rating of 592, according to my unverified math.
And I think I can even fit them under my floor!
This is the point where you call me crazy and go browse some other post….
I was originally thinking to put the packs in 4s20p, but that would essentially do nothing but increase complexity and increase my need to balance smaller groups of cells. As long as the cells are in parallel, they do not need to be balanced. Only cells in series need to be balanced, so it seems the simple solution to me is to put as few groups of cells in series as possible.
So there’s a lot of soldering to do…
I hate soldering. I’m so bad at it I don’t think I’d last a day trying to get these cells soldered to a point where I can actually trust the connections. I’ve also read that these batteries shouldn’t be soldered anyhow, but it seems that many people do it without issue, so perhaps that isn’t true. I’ve always been one to err on the side of caution however, so I’m going to spot weld mine with a battery spot welder and nickel strips. This however, leaves the problem of not being able to fuse the cells individually, which I think is absolutely vital especially since I will indeed literally be sleeping on top of them. I’ve thought of a possible solution but I have yet to try it out…
I’m thinking of taking the individual nickel strips and essentially just drilling/filing/cutting some of the strip away in order to make it thinner, and then testing that strip with a power supply to get to the exact width where the strip will break at 5-6A. After I find that exact width, I can stack all my needed strips in a jig of some kind and cut them all at the same time to the needed width, and then proceed to spot weld them as normal. Maybe someone who is crazy enough to care about this idea can tell me if they think it's a good one or not.
So now we get to the really, really, really interesting part.
Dun
Dun
DUNNNNN
The BMS.
Because I am a complete masochist and am utterly unaware of my own limitations, I have decided I want to program my own BMS. I’ve bought myself an Arduino and am currently learning how to program it, and I think without an incredible amount of effort I’m going to be able to write myself at least a rudimentary management program, especially with the help of the online Arduino community. I have a few reasons for choosing this route. Firstly, I want to understand what my BMS is doing completely. I don’t really trust what one of these ebay BMS’s is going to do for me… are the components accurate, are they rated for the voltage I want to use? Are they safe? Do they need additional components to work correctly? I want to understand this system from the ground up. There’s others I have seen online who have made battery management systems with Arduino, so I don’t think it’s terribly beyond me, but I guess we’ll find out eventually won’t we? I think my saving grace is that if I do this right, I’m only going to have to balance a pack of 4s instead of a pack with 16 groups of 4s. That will reduce the complexity of this project enormously.
The last hurdle would then be getting my solar controller to play nice with the BMS. It is a programmable solar controller, and lithium ion batteries are actually very very simple to charge (as in they don’t need to be floated, or bulked, or whatever else) so I don’t think this leg of the journey will be quite as rigorous as the rest of it.
So at this point these are my tentative plans, subject to change at any moment due to necessity or loss of interest. I really value all the input I’ve received so far, and I’ve read all the replies and I thank you for them with sincerity. Hopefully next time I post I have some good news for you all!
KingArthur
Well-known member
- Joined
- Aug 12, 2016
- Messages
- 70
- Reaction score
- 0
This here is my first module...
KingArthur
Well-known member
- Joined
- Aug 12, 2016
- Messages
- 70
- Reaction score
- 0
XFILE36 said:
I don't think that's a really good price... You would need 4 of these wired in series in order to achieve a nominal 12v. You get cells from China close to $1 per amp hour. Personally I would never buy from one of these massively overpriced domestic suppliers. Others have a very different opinion on that, but personally I think it's just so insulting I don't really want to give them my money just because they got their cells from China and slapped some English stickers on them and charge twice as much.
Close to $1 per amp hour: https://www.alibaba.com/product-det...60026525142.html?spm=a2700.8443308.0.0.8vWgBx
The seller quoted me at 12 of these shipped for $3155. $725 for shipping. :-/
jonyjoe303
Well-known member
- Joined
- May 18, 2017
- Messages
- 678
- Reaction score
- 83
For fuses you will probably have to solder them on( at least on the plus side), the negative you can spot weld. You can spot weld a small nickle strip then solder the fuse too it, solder works extremely well on nickle strip.
I find that if you scuff up the cells with a metal file, it makes the solder stick better. Some use sandpaper but the file works better.
This is a 3s30p (65 ah) battery pack I made for a small handheld portable powerpak.
As far as charging them with solar, I do it with my ecoworthy 20 amp mppt/240 watt panel, If you set the absorb to 14.4 v, I get up to 15 amps, when set to absorb at 13 volts ( as low as the ecoworthy will go) I get about 5 amps. I have to manually set the absorb voltage depending if I want to fast charge or slow charge. After the voltage gets to about 12 volts, I slow charge it so it don't trip the BMS.
The only problem is when the BMS is tripped and shutsoff the charging, the battery will read 11.7 volts even though it won't accept any more amps. This causes the controller to keep trying to charge it, trying to force more than 15 volts into the pack, the way around this is to put a overvoltage relay between the the controller and battery, if it senses more than 15 volts it trips a larger 40 amp relay thats between the solar panel and controller. This stops all voltage. Once you go over 16 volts I find I start shorting out fans/etc. If you decide to use any off the shelf BMS, make sure you test them on a smaller battery pack, I find that some that I used won't shutoff voltage if the cells go off balance, and that is very bad. They all will prevent prevent the battery from getting drain too far down.
Right now I'm using the 65 ah battery pack as a house battery, I took the 94 ah lithium battery pack (the first one I built) off line until I install fuses on it (168 fuses). Even though I've been using the 94 ah battery for the past year with no problems, to play it safe I will add fuses. The newer powerpaks I built 37 ah and 65 ah all have fuses in them.
My 65 ah house battery built from an old 17 ah jump pack. I removed the lead acid battery and installed 90 x 18650 and some modifications.
I install one of these on all my powerpacks, it shows the voltage of each row of cells. You see at a glance if your pack is in balance.
I find that if you scuff up the cells with a metal file, it makes the solder stick better. Some use sandpaper but the file works better.
This is a 3s30p (65 ah) battery pack I made for a small handheld portable powerpak.
As far as charging them with solar, I do it with my ecoworthy 20 amp mppt/240 watt panel, If you set the absorb to 14.4 v, I get up to 15 amps, when set to absorb at 13 volts ( as low as the ecoworthy will go) I get about 5 amps. I have to manually set the absorb voltage depending if I want to fast charge or slow charge. After the voltage gets to about 12 volts, I slow charge it so it don't trip the BMS.
The only problem is when the BMS is tripped and shutsoff the charging, the battery will read 11.7 volts even though it won't accept any more amps. This causes the controller to keep trying to charge it, trying to force more than 15 volts into the pack, the way around this is to put a overvoltage relay between the the controller and battery, if it senses more than 15 volts it trips a larger 40 amp relay thats between the solar panel and controller. This stops all voltage. Once you go over 16 volts I find I start shorting out fans/etc. If you decide to use any off the shelf BMS, make sure you test them on a smaller battery pack, I find that some that I used won't shutoff voltage if the cells go off balance, and that is very bad. They all will prevent prevent the battery from getting drain too far down.
Right now I'm using the 65 ah battery pack as a house battery, I took the 94 ah lithium battery pack (the first one I built) off line until I install fuses on it (168 fuses). Even though I've been using the 94 ah battery for the past year with no problems, to play it safe I will add fuses. The newer powerpaks I built 37 ah and 65 ah all have fuses in them.
My 65 ah house battery built from an old 17 ah jump pack. I removed the lead acid battery and installed 90 x 18650 and some modifications.
I install one of these on all my powerpacks, it shows the voltage of each row of cells. You see at a glance if your pack is in balance.
