Selecting BMS

G

Gregory Charles

New member
Joined
Aug 31, 2018
Messages
11
Guys I know this is some pretty basicquestions but I'd sure appreciate a hand. I'm pitting together two battery packs to go into my electric golf cart. My wet cell deep cycles have reached the end of useful life and I refuse to replace them again.

I've pretty well figured out the battery pack configuration (13s using 18650s) to get my 48 volt system. I've found a 20 amp charger made for this configuration and now I'm down to finding the right BMS for each pack. As this is a golf cart the initial amp draw (smoking tires) is going to be high. Is this a concern when selecting a BMS or do I only need to concern myself with charging amp ratings?

Second, as stated there will be two packs in parallel to provide duration and distance traveled. Can I charge them in parallel with the same charger? Each pack having it's own BMS.

Any and all help / information is most welcome.
 
1) Why not go with 14s, as standard 48V is with lithium? What is the voltage cap on your charger? If it only goes to 54V, then I can see an issue there. You could use a different charger, but I suppose you want to re-use as much as you can of existing components.
If you re-use that charger, keep in mind you will need to manually disconnect it as it will probably try to trickle charge the battery after it's completely charged.
You state that you "found" a 20A charger, does this mean you haven't got this yet? If not, then I'd recommend the 14s and go with a charger that can handle the 57.4V cap (assuming 4.1V per cell max)

2) Why have two separate banks? Why not just make 1 large one, or is this for the event of one side going down, you can still limp back? Nothing wrong with this, just curious ;)

3) You can charge both with the same charger, that's fine. There can be two ways to do this:
- Each would have its own BMS configuration. Absolute Pos/Neg of each string would be connected in parallel, and then the charger connects to this
- Each string would have its packs connected in parallel and one 1 BMS would be needed. This does have the added complexity of running more wires/connections, but you get away with 1 BMS
 
Korishan said:
Why not go with 14s, as standard 48V is with lithium?

I have seen it done both ways whenbuilding them for golf cart use.Have no idea why that is to be honest. Just a note, all I have seen are DIY conversions. None of the big manufactures have gone this way yet. (If they have I can't afforda new cart anyway.)



What is the voltage cap on your charger? If it only goes to 54V, then I can see an issue there. You could use a different charger, but I suppose you want to re-use as much as you can of existing components.

I would love to use the existing charger to keep costs down but,,, the charger I have is pretty much a huge transformer / rectifier,external stand alone, itweighs a bunchand the output voltage is not what I'd call stable and controlled. I'm hoping to mount the new charger on the cart so I can charge anywhere by simply plugging it in.



You state that you "found" a 20A charger, does this mean you haven't got this yet? If not, then I'd recommend the 14s and go with a charger that can handle the 57.4V cap (assuming 4.1V per cell max)

Thanks, I am more than open to suggestions. No, I have not ordered it yet. I'll look this option over carefully.



Why have two separate banks? Why not just make 1 large one, or is this for the event of one side going down, you can still limp back? Nothing wrong with this, just curious ;)

I am simply looking at the available space and what would fit easiest. I was also looking at ease of lifting them in and out should I need to. With that said, there is no reason it can't be one but it's going to be BIG and Heavy. I was planning to go with 13s11p x two battery packs. (Roughly 1 kWh per battery pack.)

3) You can charge both with the same charger, that's fine. There can be two ways to do this:
- Each would have its own BMS configuration. Absolute Pos/Neg of each string would be connected in parallel, and then the charger connects to this
- Each string would have its packs connected in parallel and one 1 BMS would be needed. This does have the added complexity of running more wires/connections, but you get away with 1 BMS.

This entire BMS thing is what has me a bit confused to be honest. Obviously it has to handle all the current (amps) of the charger. What about when using the cart? In testing (and confirmed with factory tech) the speed controller and motor can have an inrush amperage in the 350 to 400 amps range if you just stomp the pedal.It falls off very quickly (half a second) once moving. FYI:That's with a supposed "soft start" speed controller. (Existing fully charged lead acid batteries it will turn the tires over.)

So the question for me is, does the BMS have to handle all that amperage when the cart is in use? All the schematics I have seen suggests all the power goes through the BMS or am I missing something? If so, I haven't seena BMSthat looks like it can handle that much power.

Thanks in advance for all the help!!! I clearly need it.
Click to expand...
 
I'd be nice to use the existing charger, I agree. You might be able to smooth the output worth some capacitors. But that might not be enough if the unit fluctuated to much. Doable, but maybe not worth it.
By your answer, I'm guessing you plan on making one large pack. Personally I'd go one of two g ways: common parallel packs that are then connected in series (14 packs of 80p for example) or have many system voltage packs that are connected in parallel in the cart.
The first will give you easy packs to handle and easy hookup. The second will give you more flexibility to add/remove packs without needing to disconnect a lot connectors. But, it also has a more complex bms setup. I think barium can handle this type of setup, but not sure.
In either setup, charge/discharge would go through a shunt that is monitored by the bms. This is so you can get an accurate reading of SoC. Something that's very important in a golf cart. Don't wanna get stranded away from base ;)
So you need to make sure the shunt can handle the peak amp load.
 
Korishan said:
I'd be nice to use the existing charger, I agree. You might be able to smooth the output worth some capacitors. But that might not be enough if the unit fluctuated to much. Doable, but maybe not worth it.

The existing charger is about as barbaric as they come. No electronics to control voltage that I can see. I guess that isn't a huge issue with lead acid but fear it would be with the new set up.

By your answer, I'm guessing you plan on making one large pack. Personally I'd go one of two g ways: common parallel packs that are then connected in series (14 packs of 80p for example) or have many system voltage packs that are connected in parallel in the cart.

Yes, I'm thinking two battery packs connected in parallel. That should get me roughly the same kWh as the existing system.


So you need to make sure the shunt can handle the peak amp load.

So,,, when I look at BMS units I need to find one that would handle 200 Amps in the "discharge" rating? (200 amps per battery pack to supply in-rush current to the motor.) And yes, I will be monitoring voltage and have an audible alarm as I get close to the discharged state. (3 volt per cell)

Not certain they make anything like that... (200 amp) if not, do I need to break it down to four battery packs so each supplies 100 Amps? See, told you I was confused on this...
Click to expand...
 
Gregory: I did not bother to read your answers since you added them inside the quotes. Please correct that if you want input :)

1. I would also go 14s but if 13 suits better go with it.
2. The BMS need to cope with max current. Get a BMS with charge and discharge port separated. They suits E-vehicles better in my world. If you get a chinese one make sure it can do 2x your max nominal current! And that is per BMS so if you got 2 packs i would still have 2X current per BMS = 4x in total current.

Not sure how your layout is but I would also go 1 pack instead of 2 if its possible. Not really needed as such though.

3 Yes you can charge them in parallel but once again why not build 1 pack? :) They are basically one pack with 2 BMS systems on it.
 
Daromer... Thanks for the reply.

2. The BMS need to cope with max current.

Ok, that can hit nearly 400 Amps of in-rush to the speed controller / motor.

why not build 1 pack?

You said double the max amps so that is 800. (WOW) I don't think any BMS does that. The biggest "discharge" rating for a 14S configuration I've found is 150 amps. Doesn't that pretty well force me into multiple battery packs (4) in parallel so I can charge them without having to mess with separating them electrically.


Note: The speed controller will limit current to 400 Amps.
 
I have seen several that does 1000A and up. But yes it cost of course!
You can always bypass the BMS for disconnection part and only use the BMS for balancing if you dont want to pay that much or find someting capable of 20kW...

For a larger BMS that does 20kW for disconnection they generally start at 500USD and up towards 2000USD easy. How much you want to pay for a decent BMS?
 
The other reason for doubling the inrush capability on the chinese units is because they several overrate their units. If it's rated at 20A, then it's most likely gonna peak at 10A, or 200A is going to peak at 100A. That's the reason for doubling the output current based on "their" ratings.
 
To all that have helped, THANKS!

I am beginning to see why people that have done this (converted to lithium) have opted to disconnect the BMS when not charging the batteries. $500+ for just the BMS is not going to happen. That puts a completely different slant to my building plans and a lot of re-thinking ahead. In fact, I may take a pass on the entire idea at this point.
 
I have a question to add to this topic as I am involved a lot with building batteries for mobile electric devices ...which of course a golf cart is. Why a BMS at all? I have read all the arguments against them and as I understand it, the best BMS is a human and a good balance charger. I have not done anything as large as a golf cart yet, but this approach in everything I have built batteries for ...as well as others that I have worked with for mobile devices like electric bikes ....seems to work quite well. I am always open to counter arguments though to expand my knowledge. Isn't a bad BMS one of the reasons we can pull good cells out of laptop and power tool batteries?
 
most cases it's not the bms that fails in those laptop/powertool packs. it's usually either they sit for along time (which with the bms uses to power itself and monitor) and gets drained. Or, a cell goes bad pulling down the whole system.

a human is probably the worse monitor you could have. mainly because you have to rely on memory, accuracy, frequency, availability, and many other facets to make sure the system works properly.

a good bms will automate all the testing, balancing, monitoring, notifications, etc making your life so much easier and more productive.

Not sure what material you've read, but this forum is mostly advocates for bms, even with new cells. Pretty much the only generally accepted reason to not use it is on new cells as they are very close to identical, and should theoretically degrade at the same rate. It might take 100's, if not 1000's of cycles before they start showing signs of drift.
 
Headrc said:
I have a question to add to this topic as I am involved a lot with building batteries for mobile electric devices ...which of course a golf cart is. Why a BMS at all? I have read all the arguments against them and as I understand it, the best BMS is a human and a good balance charger.

The problem for me is that the biggest balance charger I've seen is 5 amp or less. It would take forever to charge packs this large. Even the 20 amp charger I'm looking at will probably take 8+ hours to recharge them. I can live with that (overnight charge) but much longer than that and it becomes a problem. FYI: I do not golf at all, I use the cart daily on my farm.

After looking at what others have done it seems then use a gang plug to connect both the charger and BMS at charge time and take it out of the circuit during use. That may be the way I need to go.
Click to expand...
 
@Gregory Charles: Please put your replies "outside" of the quote section. We can't tell where Headrc's comment ends and yours begins
 
Well Jehu is a big one on no BMS from what I can tell. But there is alsoa major forum of power wheelchair owners (very applicable to mobile power applications IMO) that also is very anti-BMS. Although I am not a user of that kind of mobile device, the guys there are obviously very dependent on their chairs to conduct their daily lives safely. Also, I have had at least one person I sold batteries to for an Ebike ...and he is not using a BMS either and seems to be very happy with the choice of no BMS. I am also happy going this way so far for what are much smaller projects at this time ....with just a voltageindicator telling me when the voltage is getting low I just plug in a a balance charger and let it do its job. But again, I am always open tocounter opinions to doing things better.
 
Ok, firstly, Jehu and the others are probably using new cells. The wheelchair guys are most likely using new cells. I say Jehu as well because he can buy them in large enough quantities to get a discount. And, to be honest, had Jaron gone to Jehu about Batrium instead of Pete, he'd be raving about BMS units.
EBike users and other "high" drain applications tend to not use a bms because most of the bms units that are portable like that, are all encompassing. Meaning the FETs on the bms control the flow of power through the unit.
To get a bms that has an external shunt, like the one the Batrium has, uses quite a bit more space and is not as tidy and easy to pack away into a small space. Those systems also don't control the power through the shunt, they just monitor it.

And, using a balance charger, you are basically doing what a bms does in the first place. It keeps the cells in balance. So, in effect, you are using a cheap/limited bms unit, it's just not integrated and doesn't monitor the cells during cell/pack usage.
 
Right regarding just using a balance charger is essentially using it like a BMS. Which is why I asked the question in the first place.To my recollection Jehu is using used cells in his E vehicle and in fact went through the exercise of going back and looking at cells after a year of use with no BMS and showed that the cells were still good and well balanced. So I just visited his channel again to see if his thoughts had changed. This was what I found:


Of course as he admits in the video .... he is not the end all authority on this and willing to hear counter arguments.

Next I looked at Pete's channel regarding the Batrium BMS ...this is gonna take a little more time to review. Looks like he has gone the BMS route ...but my first take is this adds another level of complexity though

Yes most ofthe wheelchair guys are using new cells ....and for that matter so am I for most of my applications except for small things like power tools. But I have found some of the wheelchair guys using used cells .....but they are really big on Lifepo4 cells and very anti- Lithium Ion for their application. And with the used cells after testing etc. they also seem to be anti-bms.

So it still seems to me that if the user whose application is mobile applications educates themselves on battery management ....and how batteries work ...then no BMS and using a balance charger as a poor man's BMSis a way that many go and look to be successful with their application.

But the learning curve on this is on-going for meso I will certainly consider all the opinionshere that are pro BMS and constantly question how I approach my builds. Always great discussion and communication here.
 
The other thing to note with Jehu's bus is that he did bottom balancing. Which does make a difference. He's not at as much risk of overcharging his cells that way. But that's because of the way he uses his cells.

LiFePO4 cells make a HUGE difference on needing a bms. Not saying you "don't" need one with them, but they drastically reduce the risks involved. They also can handle over/under-charge a lot better than LiIons can.

As far as the learning curve, I think I'm only maybe 1/6th the way up the base of the hill :p
 
You must log in or register to reply here.
Back
Top