3 LiPo Setup, Can it Work?

[lsutiger607]

Ok, so I've got an idea for a build and before I start putting the work in, wanted to ask a question. Basically I want to build a twin engine pusher and have a specific battery setup in mind to give the stability/cg characteristics I want. The desired battery setup is as follows:

One 1500mah 3s LiPo feeding each motor independently (the C rating will be adequate for the motor). Now I want to put (for CG purposes and endurance) a single 5000mah 3s Lipo elsewhere in the airplane and use it to feed each of the two 1500mah packs in parallel.

I guess what I'm asking is can you parallel two 3s Lipos with different Amp ratings and/or with different C ratings? I guess you would need to make sure that the weakest amp draw battery cannot be less than the max demand of the motor. In other works, when paralleling LiPos, is the matching of the Voltage the only requirement? Can you parallel a single high capacity pack with two smaller capacity packs at the same time?

 

[Craftydan]

Electricly, yes. Practically, maybe.

The biggest problem you need to be careful about in paralleling batteries is back charging weaker cells, and usually it'll be worse at connection. In cases like these, a diode is your friend.

To prevent back charging, put a power diode, with a higher current rating than your battery (overkill, but you don't want it to fuse), on the hot feed line of each battery. Connect all grounds together. The diodes will cut off lower voltage packs, until the higher packs have drawn down, then throttle each pack to keep the voltage at the esc input constant.


On the practical side, you are adding weight (doesn't appear to be your concern), generating heat(about 10w), and you'll loose somewhere between 1 and 0.5 volts. On a 3s, it's liveable against brownouts, but you're rpm will drop. Your esc will need to be switched to nicad mode to prevent it's auto-protect from kicking in, and I'd recommend a charge/ballance alarm on *each* of the batteries - one goes off, it's time to land.

 

[lsutiger607]

Crafty Dan,

Thanks for all the info. Regarding your comment on "back charging weaker cells" what do you mean by weaker cells? Weaker voltage (they should all be properly charged 3s) or mah?

I'm adding this diagram to again illustrate what I'm trying to do.

 

[squishy]

The problem is not all cells have the exact capacity and health as the others, meaning the overall c rating may be 20c but one cell may be 18c after a few weeks of use. After this happens that cell will have to work harder to keep up with the rest of the system, it will empty 1st and the other cells rush in to fill the void, the results can be catastrophic...like exploding airplanes catastrophic...I wouldn't recommend this at all..there is always a simpler solution and the simpler solution is always the best one to choose..

 

[lsutiger607]

The problem is not all cells have the exact capacity and health as the others, meaning the overall c rating may be 20c but one cell may be 18c after a few weeks of use. After this happens that cell will have to work harder to keep up with the rest of the system, it will empty 1st and the other cells rush in to fill the void, the results can be catastrophic...like exploding airplanes catastrophic...I wouldn't recommend this at all..there is always a simpler solution and the simpler solution is always the best one to choose..

Not sure I follow this logic. By connecting the same C, same V batteries together in parallel, the system is sharing all available capacity right? In other words, one could look at a single 5000mah pack and say, 'I'ts one part 2000mah and one part 3000mah combined in one pack', just the same as connecting two individual packs tied to one system. The batteries will all be new before use and with a constant single amp draw from the motor (which is less Amps then the max capable of the weakest battery) shouldn't each battery be taxed the same? Perhaps I don't understand this stuff as well as I thought.

 

[xuzme720]

I think what is throwing you off is this: "In other words, one could look at a single 5000mah pack and say, 'I'ts one part 2000mah and one part 3000mah combined in one pack'"

That is not how the packs are made up at all. Within a single pack, all cells are matched and identical. With proper care and balance charging, they stay pretty much that way until they die/crash/puff, even as they age. That's not to say packs don't occasionally have a cell that goes bad. That is also a cause for alarm and one of the main reasons we have so many different measuring devices to go along with the computerized balance chargers. You definitely want to find out about an imbalance in your packs on the ground, or the whole model will end up all over the ground in a hurry!

What you are talking about, at least as far as running 2 packs together, in parallel (or series as is done also) is done quite often as capacity or amp draw is needed in limited space, but is introducing a much greater need for caution. It's imperative to be sure both packs are in fairly close health and age. Running 3 packs of differing capacities is asking for a David-like end result. Even with a diode in place, the lower pack/s will continue to drain until it's either permanently damaged or bursts into flame! This is why Dan said you need to alarm all the packs individually. It can be done I am sure, but Like squishy said, simpler is better.
Can you use 2 larger packs in parallel to achieve what you are looking for? From what you have posted, it looks like you need 8000mAh. Can you use 2 4000mAh? Or what about going to a higher cell count to boost efficiency? I guess we're just a little confused over your ultimate goal...at least I am.

 

[Tritium]

ONLY parallel Packs of like voltage and CAPACITY (MAh). Anything else WILL eventually cause a fire as squishy stated!

Thurmond

 

[vk2dxn]

The setup that you want to run is a bad idea, you are still limited by your smallest cell. You would be better off just running the larger 5A 3s

 

[lsutiger607]

......... Even with a diode in place, the lower pack/s will continue to drain until it's either permanently damaged or bursts into flame! This is why Dan said you need to alarm all the packs individually. It can be done I am sure, but Like squishy said, simpler is better.
Can you use 2 larger packs in parallel to achieve what you are looking for? From what you have posted, it looks like you need 8000mAh. Can you use 2 4000mAh? Or what about going to a higher cell count to boost efficiency? I guess we're just a little confused over your ultimate goal...at least I am.

Ok, I understand that bad cells or faults in the system are bad but that's true even if I had only one battery not to mention there's nothing I could do about it anyway. About this 'smaller battery being discharged first' thing......; Current to the motor is a function of amp/hour rating and C rating right? So if the bigger battery has a bigger A then the amperage coming from it would overpower the amperage from the smaller battery downstream in the wiring and exclusively provide the electricity to the motor. That's not to say that it would back charge the smaller one right? Just that both batteries are pumping out amperage but one is stronger than the other. When the bigger battery was drawn down to the amperage of the smaller battery then they would both discharge equally. Right?

Regarding 'simpler is better', well that's just ridiculous. If we did everything simple and easy we wouldn't learn anything. If I see one more video of a stock Bixler with an FPV camera on it I'll be certain then that all hope is lost for the sport.

As to why I'm doing this, well it delves into aircraft stability and performance which is a topic for another thread.

 

[vk2dxn]

No they will not discharge equally. Think of it like a big cup of water and a small cup of water that both have the same liquid (voltage) but different volume (current). They will distribute to the load 50/50 but the small cup will run out first. You will damage the smaller battery

 

[vk2dxn]

Btw I want to say that yes adding a diode will stop back feed but finding a diode with the peak current carrying capacity means one BIG diode. Trust me. I work with full wave bridge rectifiers of up to 6MW that supply 1500v DC for our local train network.

 

[Craftydan]

Btw I want to say that yes adding a diode will stop back feed but finding a diode with the peak current carrying capacity means one BIG diode. Trust me. I work with full wave bridge rectifiers of up to 6MW that supply 1500v DC for our local train network.

Hence the "Practically, maybe." The rectifier diodes needed aren't mega-rectifiers weighing many pounds, but they won't be tiny and may weigh an oz or few on their own. If we're talking large enough airframe he doesn't care splitting large packs, that weight may be pocket change. If it's not large enough . . .


Tiger,

Primarily I meant lower voltage by "weaker", but C*mAh capacity of an individual cell do come into play. Diode protecting each pack will regulate for both in a very simple way. Do be aware you need to oversize that diode for the supplied current-- a burst current from a starting motor might be well above the ESC's rating and be ok, but would fuse the diode leading to one or more disconnected packs.

If you haven't noticed you've hit a visceral fear of most E-pilots - those little freindly battery packs turning into flying firebombs. It's not an irrational fear. Care in connecting battery packs is one of the ways we keep both the fear and the *very real* danger at bay. You're proposing something that endangers this, but diode protecting your power system is safe.

For what it's worth, I agree with most of the detractors, but don't hesitate to let others experiment and learn safely for themselves -- warn of the pitfalls, hand you a flashlight and a map, and send you on your adventuring way. Most flyers resist the urge to split packs for safety and simplicity, but true safety and simplicity demand we buy RTF and BnF planes, with branded battery packs, and stay in an empty park field.

Bubble wrap safety and pre-built simplicity have their place -- Take care but have fun!

 

[Craftydan]

Tiger,

Also, to be clear -- put the diode on the power-system side of each battery connector (on each hot lead, pointing away from the connector, toward the ESC).

Don't modify your battery pack leads with diodes, or the new pack you buy next year might accidentally become the flying firebomb because you forgot to mod the leads.

 

[xuzme720]

Btw I want to say that yes adding a diode will stop back feed but finding a diode with the peak current carrying capacity means one BIG diode. Trust me. I work with full wave bridge rectifiers of up to 6MW that supply 1500v DC for our local train network.

I was thinking the opposite direction. Thanks for the catch!

I also work with moderate amperage D/C systems but much lower voltage. I do telecom work for primarily AT&T in the central offices and mostly deal with -48VDC systems but up to 1200A. Wet cell strings for the most part and a plethora of different rectifier types. Mostly plug-and-play so surely not as much fun as your stuff!

 

[xuzme720]

Regarding 'simpler is better', well that's just ridiculous. If we did everything simple and easy we wouldn't learn anything. If I see one more video of a stock Bixler with an FPV camera on it I'll be certain then that all hope is lost for the sport.

I only meant that in the regard of learn about the system before experimenting. I fully encourage innovation and experimentation but wanted you to understand what you were getting into. I'm sorry if you took that the wrong way, I should have been clearer in that regard.

 

[vk2dxn]

I was thinking the opposite direction. Thanks for the catch!

I also work with moderate amperage D/C systems but much lower voltage. I do telecom work for primarily AT&T in the central offices and mostly deal with -48VDC systems but up to 1200A. Wet cell strings for the most part and a plethora of different rectifier types. Mostly plug-and-play so surely not as much fun as your stuff!

Thumbs up my friend