Battery ah question

[USAEBIKECO]

This is a question for someone who has better knowledge than me.

I have two batteries. One is a 36 volt 10 ah , and the other is 36 volt 12 ah. If I tried to connect these in parallel, what would the outcome be? thanks

 

[Drunkskunk]

effectivly a 20AH battery, as the battery's AH is limited by the lowest capacity cell. I.E. The 10 amp cells.

effectivly tyou'll lose 2 AH of capacity from the 12ah, because the 10AH will run down first and fall to there cut off voltage while the other battery still has capacity. The danger here is a volt meter hooked to both batterieas will show the average voltage, which will be higher than the lower voltage of the 10AH cells, so you need to moniter it's voltage directly.

 

[dnmun]

no, no, he meant parallel, not series. so the capacity would add together and it would put out the full 22Ah at 36V, and you could combine them without the diode since they will exhaust at the same rate, but you need to maintain a BMS on both packs in case one cell goes dead, you need to know that if you aren't using the diode.

 

[dogman dan]

Youbetcha, the one with more internal resistance will start providing less of the amps, and both will end up running down at a similar time. It could be a problem though, if one was really a lot smaller than the other, and then started pulling a lot of current to recharge itself. Example, a 4 ah lifepo4 could start to charge itself at an amp rate that was not good for it. It won't matter if one battery is a different chemistry from the other either. I had great results with 8 ah nicad and 12 ah sla together paralell.

 

[USAEBIKECO]

what about a dewalt 36 volt battery 2.2 ah combined with a 12 volt 12ah battery... whould that work

 

[dnmun]

if you combine different chemistries in parallel, they will have different resting voltages so the low potential will draw current from the hotter pack, so you need combiner schottky diodes then.

 

[dogman dan]

We're still taking paralell here? 36v paralelled with 12v won't work, they both have to be the same approximate voltage. In series, it will work, to make a 48v battery, but the capacity, as stated before about series, would be the smallest one, so 2.2 ah. When the 36v pack is drained, it would either shut down using its bms, or if no bms, it would kill itself trying to discharge itself while the 12v batttery still ran.

 

[Drunkskunk]

no, no, he meant parallel, not series. so the capacity would add together and it would put out the full 22Ah at 36V, and you could combine them without the diode since they will exhaust at the same rate, but you need to maintain a BMS on both packs in case one cell goes dead, you need to know that if you aren't using the diode.

Yeah, Parrallel, and yeah, it would make a 22AH battery technicly, but you could only use it to 20AH since the 10AH cells would drain at the same rate as the 12AH cells, and run out of power 2AH ahead of the 12AH cells, meaning you would only have 20AH capacity.

 

[vanilla ice]

Really, I thought it'd be 22ah total. As the 10 runs down in voltage, the 12 should take up more of the load right?

 

[rhitee05]

Yeah, it would be pretty close to 22 AH effective, but the C-rating would drop over the last few AH as the larger battery takes up more load. If both batteries were 1C, you'd have a 22 A battery at full charge, but only a 12 A battery at the end of discharge when the 10 AH is pretty much done.

 

[DahonElectric]

This is a question for someone who has better knowledge than me.

I have two batteries. One is a 36 volt 10 ah , and the other is 36 volt 12 ah. If I tried to connect these in parallel, what would the outcome be? thanks

Overheating perhaps leading to an explosion if you're not careful depending on which battery combination, discharge rates and chemistry.

While theoretically, you would have 22ah, the true capacity is somewhat less. Ideally, you should connect 2 batteries in parallel with similar chemistry, same discharge rates (same C rates) and balanced so their open circuit voltage should be roughly the same and they drain simultaneously together. What you are proposing here is in fact an imbalance. With an imbalance parallel circuit, 1/3 of the current drain comes from the weak battery (10ah), while 2/3 drain from the stronger 12ah. So for a 10amp load, 3.33A will come from the 10ah battery, whereas 6.66A will come from the 12ah battery relatively speaking. If the current drain is high and consistent as in the case of an ebike, sooner or later the 36v 10ah battery may drop its open terminal voltage lower than the 36v 12ah (assuming both have exactly the same C rates, chemistry), and if that is the case, the stronger battery will recharge the weaker battery. The energy stored in the stronger battery is simply transferred with some loss from resistive (cabling) and heat to the weaker battery and that energy from the weaker battery is then used to continually power the motor. As far as the load is concerned, it is seeing a 22ah battery pack, but the actual capacity will be close, probably a bit more than 20% of full capacity (since quick charge can only achieve 80% of expect capacity in a short time).

You can actually do this as an experiment. Try paralleling a 5V power supply to a 1.2v NiMH battery and you'll see that, the voltage drops from 5v to somewhere near 3 or 2.5v on the terminal. If you leave it long enough, it starts to heat up the battery or may cause an explosion. So don't leave it too long!

But here's the problem. Not all batteries are made alike. Different chemistry have differing discharge characteristics and rates. Unless you know for sure the battery paralleled together are similar in almost everything, you're really playing with Russian Roulette. The problem is compounded by the fact that some batteries do not like to be recharged without a proper charger -- Lithium Ion or some specialized chemistry.

If you still do want to mix and match batteries, I say choose the first pack to have a higher C rate, higher capacity, so the open terminal voltage of the battery won't drop too much during load. So the 12ah should ideally have a 4 to 5C rate. The second extended capacity pack should be something that can take the abuse, like a NiCad, NiMH or even a SLA with a lower C rate. What you don't want is to have your 1C Chinese Lithium pack be recharged by a stronger high discharge rated pack just because the Lithium pack may drop the voltage more dramatically during extended loading by the motor, while the high discharge rate pack may drop less.

DE.

 

[DahonElectric]

Yeah, it would be pretty close to 22 AH effective, but the C-rating would drop over the last few AH as the larger battery takes up more load. If both batteries were 1C, you'd have a 22 A battery at full charge, but only a 12 A battery at the end of discharge when the 10 AH is pretty much done.

2 decades ago, I was an avid R/C guy. One problem with R/C cars in the past was with the Tamiya Nicads of the 80s (6V 5 cell and 7.2v 6 cell) is that, there was no place to recharge these batteries once depleted when you're in the field.
The trick we used then was to recharge these batteries up to about 80% with a 12V SLA automotive battery. You would hook the RC battery up to the auto battery in parallel with a special cable with one end laminated with plastic so you can use your hand, grab the line to feel the heat. When it's hot, it's charging. When it's warm, charging is almost complete, though not fully charged. But it's good enough to play.

I think the same thing will happen with the 12ah battery against the 10ah. If the 10ah depletes first, the open terminal voltage will be lower than the 12ah battery, which means that the strong battery will start charging the weak battery in the attempt to balance the circuit. As far as the motor is concerned, it is still seeing just one battery. I know Nicads can take the abuse as I've grew up around them. I'm not sure about those fancy LIFEPO4 or then LiMn. Can they be recharged without a proper charger? What would the BMS do if a recharge is attempted on the battery terminals?

DE.

 

[Anonymous]

.

 

[dogman dan]

The best way would be with the diodes, for sure. They would be mandatory with anything using a bms I think. Mostly to protect the bms though, which might not like a lot of amps going through it backwards if that should happen.

To really know what is going to happen, you'd have to run actual real world tests, of course. What occurs in reality will depend on lots of things, such as the quality of the batteries, how worn out they are, wire sizes and length, connector quality, etc.

What would be expected to happen with two chemistries , or even two different batteries of similar chemistry and similar size is the one with the highest internal resistance will provide less current than the one with a lower IR. So the one that gets more discharged is likely to have a higher IR at that point, and start providing less. This is why two slightly different size batteries will tend to end up fully discharged at nearly the same time. As with any battery, a shallower discharge cycle will help.

In another case, with a very big battery and a very small one, the small one will deplete so fast, that it will then tend to pull charge from the big one.

 

[dnmun]

alotta misinformation collected here. if you have 10Ah and a 12Ah SLA set in parallel, the voltage on each of the 12V SLA will drop by the same amount. therefore the capacity of each battery is reduced by the same percentage. the 12Ah battery would discharge 12Ah and the 10Ah battery would discharge 10Ah and they would be at the same voltage.

if you have identical batteries in parallel which use a BMS, then it is safe to use them without a combining schottky diode. they will draw down at the same rate, so neither would discharge into the other.

however, if one battery had a cell failure which caused the total battery voltage to drop below it's adjacent pairing, then current would flow from the good battery to the bad battery in an attempt to keep the voltage identical.

but one would expect a failure of one cell in the battery to cause the BMS to shut down the output FETs and this would prevent the discharge of the paired battery into the dead cell.

it would depend on the mental acuity of the rider to realize that one battery BMS in the pair was shut down but otherwise i would expect it to be safe.

different chemistries combined in parallel do need schottky diodes to keep them isolated, and different chemistries in series also present a problem because of limits on the discharge capabilities of individual chemistries. lithium ion and lifepo4 cells can discharge at 3 to 4 times the rate of nicad and SLA. so same chemistry and caapacity is needed in series.

obviously it is is simpler and safer to have only one battery, or if in series, have the same chemistry and Ah capacity, and for batteries in parallel, one should try to always use identical chemistries. just for simplicity too.

but i have 2 nicads in parallel without a diode. one is 500mV below the other so when i am not using them then i isolate them by removing the fuse in the line between them, and replace it when the packs are charged up and ready to ride. but they really should have a schottky, on the lower voltage pack at a minimum.

but i expect eventually i have to find the bad cell in there to be safe.

 

[rhitee05]

I think the same thing will happen with the 12ah battery against the 10ah. If the 10ah depletes first, the open terminal voltage will be lower than the 12ah battery, which means that the strong battery will start charging the weak battery in the attempt to balance the circuit.

You're mostly correct. The 10 AH batttery won't actually be discharged lower, then recharged from the larger 12 AH battery. Since they're in parallel, the voltages have to be the same and the currents will vary as needed. As the smaller battery nears discharge, the current will move away from a ~50/50 split to more 40/60, 30/70, etc. However, another way you could look at this is that both batteries are still supplying the same amount of current and there's an additional current flowing from the 12 AH battery into the 10 AH that's charging it as you said above. The currents add (called "superposition" in EE terminology) so you get I/2+Ic for the larger battery (same direction) and I/2-Ic for the smaller one (opposing directions), which means the net current is an unequal split as I said above.

Two different (and both correct) ways to look at the same problem!

As others have said, this could get you into trouble if the two batteries are different chemistries with different discharge curves. Probably safest to use Schottky diodes to keep them separate unless you're really sure of what you're doing. In fact, you really should know what you're doing regardless!

 

[Drunkskunk]

ignore what I said, I got no idea what I was thinking.

 

[DahonElectric]

Alright, I guess I didn't make myself clear in the past post as I was babbling what would happen if you are in a real world.

To be blunt in theory,

No battery in parallel can charge another regardless of their Ah rating unless their voltages are different, which was why I didn't suggest a Schottky diode in my post. Parallel batteries supplement amperage Ah and not voltage. This means that a 36 volt battery on the left and a 36 volt battery on the right will only supply 36 volts total; but their amp ratings, if for example were 10Ah and 12Ah, would total up to 22Ah. The ebike is completely oblivious to the Ah rating of the battery. All it wants is its minimum voltage draw to function normally. However, with a 10Ah and 12Ah, the current draw respectively if they are both rated as 1C, will be 45% for the 10Ah and 55% for the 12Ah. 10/22 = 45% and 12/22 = 55%

As current is used up at those ratios within the batteries, voltages will fall and fall identically/equally. Once a LVC minimum voltage (V) is reached, the Cycle Analyst or a BMS will shut off to protect both batteries. Like dnnum had alluded to, if any battery is drawn down too far regardless of its configuration (single, double, parallel, series) can reverse polarity and create a "runaway" battery. Your device protects against this and so do some smart batteries with a BMS internal circuitry of their own.

Ok, now we're back into reality..

The biggest problem with batteries occurs when mixing brands or types and buying batteries from questionable sources (concerning only on price rather than quality). As I had posted in my past message, the actual open voltage from one battery during the load can drop further due to quality of the cell,defective cell or cells and that was my concern. When this occurs, the stronger battery will attempt to recharge the weaker battery. And the reason for this concern is the fact that I do work in the electronics industry and have seen mixed batteries parallel together that never could charge another actually did and leak or explode. The main culprit? As we had discovered, people went cheap and bought Chinese made eBay batteries and mix them with good well made ones to save costs. I think a lot of people who had posted earlier are reflecting on their experiences in real life, though may not jive with the actual theory and I do agree that you should know something about this before attempting it or simply have 2 separate batteries and connect them individually.

Does this help?

DE.

 

[Anonymous]

.

 

[rhitee05]

DE,

You're right, the big danger is one battery pumping amps into another if something goes horribly awry. Depending on chemistry, that could cause anywhere from a dead battery to self-immolation. Even when using well-known batteries, "name brand" if you will, there's still the slight chance of something going wrong, more so if abused.

I imagine that's the main advantage of the parallel-first pack building philosophy. If you have several cells wired in parallel, they'll tend to protect each other and make it much harder for one cell to go bad. If fact, you'd probably have to kill the entire parallel string at once. When you build a single series string from these parallel "super-cells" then there's no danger of self-charge like you're speaking of. Seems to me that it'd be good practice to always use schottky's when joining series strings in parallel, unless they're very well-matched and carefully monitored.

 

[317537]

The 10ah would drain faster and the potential difference between the 10ah and the 12ah would make it so the 12 ah would drain into the 10ah. Internal resistance would should normally be calculated in parallel however this is in 2d theory . Current pathways from the 10ah through the 12ah bridge to load and the 12ah to the 10ah would cross and divide and loss of energy is imminent. It not an idea topology. The voltage difference between 12 ah and 10 ah could get at least .5 volts which in essence from a single load point creates and small short.

Diagonal terminal linkage is one solution, meaning you should hook the - load (neg) lug on the negative of the 10ah batt terminal and the + load lug onto the 12ah positive batt terminal and create equal length bridging for parallel connections, and do it in this order, 10ah battery ground terminal and 12ah positive terminal, as I am told electrons are the slowest moving component of electricity and move from + to- and to the negative terminal of the 10ah batt should be the shortest route, energy makes it way from - to + and travels through resistances more freely. The diode thing in there which in my view would be better to avoid as you get closer to your LVC and/or waisted energy on the V drop over a diod is evident, you could try it.

In a perfect battery series setup one should place lower capacities (runt cells) on the negative side of the series and go up.

In reality by mixing the ah capacity of the two batts in parallel is creating a series type of behaviour when one batt voltage becomes lower than the other but the voltages of each batt oppose internally against each other instead of accrue.

Suggestion in series setups of swapping the order of cells after 6 moths is well noted in the past, primarily because the positive side starts to lose capacity first. This may not be so apparent at first on batteries with lower internal resistance but this is observed to greater a extent in SLA batts.

Again I haven’t researched my post and I make stuff up as I go, so take everything I may say as the opposite of actually goes on.

Edit:
On further review of this thread even though the voltages on both appear batts appear to be equal there are is IMO complexities at work when all resistance are considered. In parallel it should by theory equal exactly 22 ah, I theorise more closely to 21.8 ah and the .2 ah loss is not much concern.

I do see batteries in parallel show slight voltage variation during high load with SLA, indicating one battery will drain into another and losses should be expected.

Other than all this over .2 ah you gain more by using this configuration than lose. Charging would be the main issue where the smaller battery drains into the larger battery and because the charging cycle is much slower than the drain cycle the smaller battery stays in full charge mode for longer than it should.

A question arises. Do these problems present to the dry dielectric as they do with water based batteries? . I estimate 5 times less of a problem.

Remember the life span is longer on lithium tech so these problems may not be so evident over shorter time frames
.
I believe the same things present on drain as it does with charging but instead you cant charge properly which over time leads to the larger not working to its full potential


As advised maybe one should adhere to old school ideals when dealing with expensive chemistry just to be safe. Do not mix different capacity batteries.

 

[DahonElectric]

So I don't need a shottky diode to join serial strings of tool batteries in parallel? I don't trust the minor variances between these batteries, they have differences in voltages and capacities, and I don't like the idea of any of these serial strings being able to push current through the others, especially when some have reached max depth of discharge. Is it totally uneccessary? It seems to me that a $3 part is worth it to make the current flow in one direction towards the controller, considering the cost of all these batteries. Nobody knows exactly how the BMS behave either, it's basically a black box protection circuit, we know approximately the max amp cutoff and lvc cutoff but we don't know much else about how the circuit will behave when exposed to other batteries, at least with these tool batteries.

Tool batteries are obviously designed for use with tools in mind and usually, the marketing people and the engineering people will come together and fight over what's right and safe and what's more cost efficient. Most likely, price is the determining factor with minimal safety factor built in for liability sakes and so, they will make something work fine as long as it works with the intended tools. They didn't design the cells to work safely with ebikes, but we use them anyhow because they are low cost. I plan to use 2 Canadian Tire 20V Yardworks battery to experiment with as a 40V 6Ah short range grocery getter battery. If you plan to string a series of batteries in parallel and you don't know the quality of the cells and batteries you're using, what's wrong with putting in a $3 Schottky diode? Consider that as cheap insurance alright? You might or might not get a dead cell during the middle of a deep discharge from the motor load, but at least you know you have a diode. The nominal voltage drop across the Schottky diode is minimal anyhow, so if you have a programmed LVC on your Cycle Analyst, it will not hamper range performance whatsoever.

The reason why I put up the theory portion is to make people understand what a parallel circuit does and does not. The thread were filled with people with good observations of their own battery system with limited electronics knowledge, in reflection no doubt to the variability of the battery quality they are using. I said earlier that I am not an expert with ALL recharge-ability of various Lithium Ion chemistry, but my years of experience working in the industry I'm in as we deal with Lithium Ion batteries of our own that when our customers start mixing and matching our battery with something they thought was made by us over eBay, but turned out to be fake, the result is mostly fatal.

DE.

 

[Anonymous]

.

 

[DahonElectric]

Heh thanks Nuts&Volts for sharing the schematic and info. I'm now off to CT to check out these Yardworks batt -- 2 Yardworks is what I wanted for the 9C.