Before I connect these batteries in parallel…

[justlooking808]

:wink:

oh, forgot to add, check the BMS if it has one to see what max amps you can charge at as some OEM can only take 5amps max and can prevent charging if it's too high a current.

Yeah our batteries are different in the sense mine has separate wires for charging and discharge. So the way I have it now, example my bike is charging at the moment and only one battery is being charged through the BMS while the other pack is being charged through the discharge wires. That’s why I’m thinking it might be safer to parallel connect the charger wire so it does go through bms on both.

But yeah I know one pack can handle 8 amps as I’ve done it about 5 to 10 times while needing to fast charge on some dirtbike tracks. That was when I just had the one single battery. But now I feel like my normal charging should be 8amps since I have two packs and and the current would be divided, but I’m not sure hopefully someone can chime in..

did you confirm this by checking the bms?

most times there is a separate charge(dc connection) and discharge port (xt60), but it's soldered on the same line going into the BMS. having separate ports usually cost more and in most cases are only added for custom DIY batteries or custom orders. but only way to know for sure is if you built it your self or you open it up.

Well when I was setting up my region several months back I talked with the seller and they confirmed that I would have no BMS protection using the region back through the discharge port. So therefore they’re not connected on the same line. When my battery is fully charged I’m just careful to not press the E break :lol:
But let’s be honest our battery is only fully charged until you full throttle for more than just a few seconds especially at 7+ kilowatts

so then yah, if the charge port is seperate, from my understanding, it's only one way in. so should be fine to connect it in parallel and charge it parallel. unless the charge port has no protection (which most do) then i would be worried.

 

[docw009]

Eastwood gets more range from two packs in parallel than from running them separately and others wonder why . It's because we have half the discharge rate. Here's the discharge curve of my favorite cell, the Panasonic GA. It's rated at 3400 maH, but that is at 2 amp and taking it to 2.5V. Run it at 4A, and it hits the controller's 3 volt LVC at 3050 mah. Kick it up to 10A, and you get 2700 mah.

If you have packs of high capacity or poking along at low current, the effect will be less since you wont be pushing high discharge rates in either case,

 

[TDB]

There is a 100% chance that you will get more range (more energy out of the 2 packs) if you drain one battery pack, disconnect it, and then connect the second pack and drain that.

In my experiences, the opposite has always been true.

For me, paralleled packs of same cell type and voltage, regardless of differences in age and capacity and cell quality, give greater range (more total Wh) than any of them by themselves.

I suppose if you used packs so large that they were capable of far more current output than the system would ever draw even at peak, and so had no voltage sag and no internal losses to heat, then you would get about the same range / Wh whether paralleled or not.

But I haven't had a case yet where running the packs empty one after the other ever got more range / Wh than running them paralleled. (again, assuming same cell type (chemistry) and pack voltage (number of series cells).

Ditto.

Li-ion cells at a lower discharge rate put out more total energy. Every cell test supports this.

Cells will have a longer useful life if the same Amps are spread across twice the number of cells.

And cells will have a longer useful life if either pack is not being fully discharged on shorter trips.

 

[Eastwood]

Eastwood wonders why he gets more range from two packs in parallel than from running them separately. It because we have half the discharge rate. Here's the discharge curve of my favorite cell, the Panasonic GA. It's rated at 3400 maH, but that is at 2 amp and taking it to 2.5V. Run it at 4A, and it hits the controller's 3 volt LVC at 3050 mah. Kick it up to 10A, and you get 2700 mah.

If you have packs of high capacity or poking along at low current, the effect will be less since you wont be pushing high discharge rates in either case,

GA_curves.jpg

Well I wasn’t wondering why I get more range from two packs in parallel, someone else brought that up haha
But thanks for the input tho!

 

[Chalo]

Well while everyone is arguing can someone please try to answer this question? :lol:

So is it safe to parallel the charger ports just like I did the discharge BUT while leaving the discharge connected?? I want to connect them so that I can charge at 8 A so each pack would be getting 4 A each.

The cheap dumb BMSes I use can discharge through the charge port. I've even screwed up and plugged my controller into the charge connector before (they're both Anderson PP45), and the bike got to work just like usual, surprisingly without apparent damage to the BMS.

That's a good reason to use a different connector for charge and discharge if your BMS separates the two.

If you have both charge and discharge ports paralleled, then no single BMS will be able to cut off its pack until the other BMS agrees with it.

 

[amberwolf]

But that's not relevant to the question, because differnet voltage cells can't be paralleled

Take cell A with 4V and 50mOhm resistance. Then take cell B with 4V and 60mOhm internal resistance. Now parallel them. Now connect them to a load. What voltage is cell A putting out? What voltage is cell B putting out? Are they the same or different? I think you just paralleled cells with different voltages.

Nope. If you parallel them they will then have "the same" voltatge, regardless of what they started at, differing only if and when there is current flow thru the resistance of the paralleling interconnects themselves, causing a voltage drop.

That drop will be so tiny it is essentially zero, unless you have a very poor or inadequate interconnection (in which case you have much bigger problems with your battery setup to worry about than anything discussed in this thread).

If you don't parallel the cells, then there will be differing voltage drops under the same load for different internal resistance of the cells.

If paralleled, the cell with lower resistance will take more of the load than the cell with higher resistance, and the current thru each cell will be proportional to each other's resistance and the load, and the resulting voltage drop across each cell will be the same (assuming near-zero resistance of interconnects).

 

[amberwolf]

So is it safe to parallel the charger ports just like I did the discharge BUT while leaving the discharge connected?? I want to connect them so that I can charge at 8 A so each pack would be getting 4 A each. I mean I’ve charged the first pack at 8 A a few times but I certainly don’t wanna make a habit of putting 8 A through the first pack so it seems like it would be better for the longevity of the battery to parallel the charger ports, so that I can charge at 8 A without all that current going through the first pack.

If you parallel charge ports and also (separately) parallel the discharge ports, for the most common cheap BMS designs using separate sets of FETs for charge vs discharge, then neither port can stop the battery from overcharging or overdischarging, because current can flow backwards out of one battery and into the other uncontrollably.

Somewhere around ES in a "parallelling batteries" thread I have a detailed explanation of it, but essentially the FETS act as shortable diodes in the BMS. If they are turned off, the diode prevents current flow into the port but allows flow out of it, for the charge port, and prevents current flow out of the port but allows flow into it, for the discharge port.

This only matters if one pack has a problem of some kind relative ot the other--if they are both equally healthy then it doesn't make any practical difference, but the BMS is there specifically for the case of there being a problem and protecting the cells against damage from that problem.


There are other potential issues but you'd have to find my old post because it's too much for me to type up at the moment.

 

[amberwolf]

That's the key. Connect two batteries of the same capacity in parallel that have different internal resistances, and the battery with the lower resistance will be providing more current. So they will be discharging at different rates. But since the batteries are parallel, the battery that is discharged more, will have a lower voltage, meaning the battery with the higher voltage (the one with more charge left) will have to constantly charge the other battery. Taking energy out of a battery is not 100% efficient. Putting energy into a battery is not 100% efficient. Hence less total capacity than if the batteries were not paralleled.

That is backwards, because if they are paralleled there is less load on each cell (presuming the same total load), and each one has less Peukert loss (even though this is far less significant with Lithium than with others like Lead chemistries) than if you put the same load on just one cell or pack.

There is less voltage drop per cell, and you also get more total power (watts) out of the paralleled cells than you can out of the nonparalleled version, given the same current draw. Whether you get this in practical reality on an ebike depends on your controller's behavior and the loading you put on it. With a static load in a bench test you could see this, however.


You're welcome to setup a video of a well-instrumented bench test to show all the current flows and disprove this, if you like.

Or setup a bike on an indoor dynamometer with the same trip characteristics for three "rides", and then run a test with first one of two packs, then the other, and then both in parallel, and measure what you get out of the system each time.

Either of these will show you what is really happening is not what you think is happening.

 

[Eastwood]

If you have both charge and discharge ports paralleled, then no single BMS will be able to cut off its pack until the other BMS agrees with it.

neither port can stop the battery from overcharging or overdischarging, because current can flow backwards out of one battery and into the other uncontrollably.

Yeah this makes sense, I figured I would have no BMS overcharge protection.

That being said you guys think it would still be more practical to parallel the charger ports as well just so there’s more even current flow going through the packs while charging? I mean if I’m charging at 3 or 4amps it doesn’t matter if it’s all going through one pack, but I want to be able to charge at 8 A. So seems it would be better for the batteries to have parallel charger port as well So the current flow from charging would go in the packs evenly and not all the way through one and then back out the discharge into the second battery. And again I know the overcharge protection is already out of the equation.

Sidenote on my adjustable charger I have the voltage pot set slightly less than 84volts which seems to be a good thing now that I don’t have overcharge protection.

 

[docw009]

Eastwood gets more range from two packs in parallel than from running them separately. and others say no way? It's because we have half the discharge rate. Here's the discharge curve of my favorite cell, the Panasonic GA. It's rated at 3400 maH, but that is at 2 amp and taking it to 2.5V. Run it at 4A, and it hits the controller's 3 volt LVC at 3050 mah. Kick it up to 10A, and you get 2700 mah.

If you have packs of high capacity or poking along at low current, the effect will be less since you wont be pushing high discharge rates in either case,

GA_curves.jpg

Well I wasn’t wondering why I get more range from two packs in parallel, someone else brought that up haha
But thanks for the input tho!

Sorry. fixed it,

 

[amberwolf]

I figured I would have no BMS overcharge protection.

I wouldn't recommend that; the BMS is there to prevent pack problems. If you want to do that, I'd recommend manually monitoring the cells in each pack for such problems during charge/etc., then manually shut off charging as needed to let the BMS fix things if it can. If you have cells that you fully trust without a BMS, and you know they're all good quality well-matched cells, and the charger is a trustable quality unit, then it should be ok without any monitoring (like I do with my EIG packs, no BMS and just bulk charger), but even good ones change with age , and eventually need monitoring.

That being said you guys think it would still be more practical to parallel the charger ports as well just so there’s more even current flow going through the packs while charging? I mean if I’m charging at 3 or 4amps it doesn’t matter if it’s all going through one pack, but I want to be able to charge at 8 A. So seems it would be better for the batteries to have parallel charger port as well So the current flow from charging would go in the packs evenly and not all the way through one and then back out the discharge into the second battery.

But...that latter *will* happen. Every time one of the BMSs shuts off it's charger port due to HVC or other limits, then the current does flow thru the other pack and then back into that one. (unless you unparallel the discharge ports).



And you don't necessarily have that 8A split between packs. If the BMS does shut off one pack due to a high cell so it can balance that one down (and the discharge ports are not paralleled) then the full current will be going thru the other one. That's probably not going to happen while at high charge currents, it usually goes down a lot before reaching that stage, but if there is anything wrong with one of the cells (like as the packs age) then this can happen, so you end up with full high current into just one pack.


If that's more than the BMS charge FETs can take without overheating, the port could fail; sometimes they fail shorted, which means they can never shut off charge at all, and that is usually a "silent failure" so you don't even know it's happened until you start having problems with the pack over time. If they fail open then the pack just won't charge, and it will be obvious.

If the current is more thant he cells are meant to take, then they'll heat up during charge, and they will age faster than normal, though unless it's WAY higher than they can handle it's unlikely to cause anything dramatic to happen immediately.



If you have no BMS in the way, or the discharge ports are also paralleled, then yeah, it'll be distributed based on pack internal resistances, whcih change as charge state changes. But this way you have nothing to prevent overcharge of cells, in the various failures that can happen (such as the charger not obeying it's voltage limit and shutting off, and continuing to pour current into the packs until something gives way).

If the BMS on both packs shuts off the charger port, then they're safe, because nothing flows in at all, but if for whatever reason only one does, then you still get flow back thru the discharge port while the charger still pours current into the still-on charger port of the other one.

If you set it up so that you can unparallel the discharge ports (just unplug one) while charging ports are paralleled, and the same with charge ports while discharging paralleled, then you don't have the risks of both being paralleled.

If your BMS uses a single port for charge and discharge, the FETs are setup in a different way, and these problems don't exist there, either.

 

[Eastwood]

If you set it up so that you can unparallel the discharge ports (just unplug one) while charging ports are paralleled, and the same with charge ports while discharging paralleled, then you don't have the risks of both being paralleled.

Thank you very much for all this detailed information, this helps tremendously!
I will take your advice and disconnect the paralleled discharge wires while charging. I’ll make a parallel connection for the charger ports that are only connected while charging.

Now when I charge the batteries through the charger ports since they Will be paralleled
do I ever need to check each batteries voltage before reconnecting The discharge wires? Or could I assume that they are the same voltage since they were both connected to the charger in parallel? maybe one pack could be turned off by the BMS for something random while the other pack remains charging. Or maybe Should put some type of voltmeter on each pack so I can visually see it, I don’t know if that’s possible on the charge or discharge wires since there’s so much current flow. I mean I can manually check with a voltmeter, I would just like to make it more practical

 

[99t4]

do I ever need to check each batteries voltage before reconnecting The discharge wires? Or could I assume that they are the same voltage since they were both connected to the charger in parallel

I vote for yes, as the consequences of blindly connecting the batteries could be very unpleasant if there was a wide variance in voltage.

Or could I assume that they are the same voltage since they were both connected to the charger in parallel

You easily thought up a few not uncommon scenarios that could cause unequal voltages; there are more.

 

[Eastwood]

I vote for yes, as the consequences of blindly connecting the batteries could be very unpleasant if there was a wide variance in voltage.

Yeah I’ll go this route. I’ll put a voltage displays connected to each battery so I can be 100% sure. Or actually I might put the voltage displays on the Y connector for my parallel connection. Well see..

Out of curiosity what do you think the maximum different voltage could be when connecting two packs before there’s a problem? Trust me I won’t be testing this :lol:

I will make sure the voltage is identical before connecting. Even pass the decimal :lol:

 

[spinningmagnets]

Between a 4.0V pack and a 4.1V pack, no resistor is needed, they will equalize with virtually no heat in the cells or connectors....

 

[Eastwood]

Between a 4.0V pack and a 4.1V pack, no resistor is needed, they will equalize with virtually no heat in the cells or connectors....

OK good to know, thanks! I’ll always make sure they are exactly equal to leave no room for error :wink: