In the Assembly Guide, it mentions for longer run times, hook two 6V’s in series instead of using one 12V.
Wouldn’t we get an ever better run time hooking two 12V in parallel? Or is that just adding too much mass to the rover?
- Robert
I think you are misreading it… the point isn’t the “two instead of one” part, it’s the “higher capacity” part.
I do not like putting NiMH batteries in parallel. Some say it’s ok, but I don’t think it’s safe.
viking_robert,
I agree with Robot Dude.
No two batteries will charge to the same final voltage and the peak charger will always key on the battery with the higher final end voltage.
Once the charger is removed, the battery with the higher final end voltage will attempt to equalize with the other battery.
This results in an unrestricted movement of current between the two batteries until the voltage across the two is equal.
Just a thought.
Regards,
TCIII
Parallel all the batteries you want, that’s not a problem…
Parallel isn’t safe, however I have wired my two 7.2v 2800 mAh batteries in series for 14.4v and have been using it this way for two years now.
Paralleling batteries IS safe so long as the cells are matched and the packs are of identical capacity/voltage and stay that way. As mentioned, the packs will attempt to balance each other. If you have a dead cell it won’t like this very much.
I’ve used paralleled high voltage NimH and NiCd packs for years and never had a problem. I’ve even knowingly used 2 identical packs with unpatched cells and never had problems. R/C Truck,boat and plane guys have been doing it for years.
Just be sure they are the same voltage and same capacity. If they are different packs, you will have big problems. And ensuring the cells are matched by custom building and testing them is even better.
This subject has been beat to death here before. All I’m saying is when there are no problems, it works well. But when there is a problem, such as if one cell shorts, a lot of current can be transfered in a very short time. 
But then you get pretty fireworks and wonderful colors of smoke! Where is the downside!
Hey, today IS the beginning of chinese New Years 
I’ve been thinking about this as I can see the need to extend the run time of my rover soon, what about a circuit that switches from one battery pack as it runs out to another ?
When everyone parallels batteries, why is it that no one puts diodes in to prevent the ‘bad cell syndrome’ scenario?
It is very much possible to combine DC power sources, safely. I know, my company does it all the time in their products (military).
There is an efficiency loss when adding the diodes in series, and also there is a voltage drop proportional with the current drawn. This has the added effect of making the effective voltage at the load bounce around between light and heavy loads. If you have an actual power supply, that is something line or generator powered, you can work around this by sensing the voltage after the isolation diode and monitoring output current so the effect is not so pronounced.
No fair! All my smoke is a boring shade of dark grey 
I am confused on what this affects?
Yes, I understand there is a small loss across the diode. I think the safety might be worth it, at least for me.
What would be affected by the voltage swing though? As far as I know, the only thing people put on batteries without regulation are motors. Motors can handle the swing quite well I would think (they already do, day in and day out).
Logic control shouldn’t be effected by it, as their circuits are almost always on voltage regulators.
Personally, I will probably be diode isolating my bots, as I am looking at LiPos* for power. So in my case, I already plan to have a 5V switch mode regulator for electronics, and 6V switch mode regulators if I need servos. The motors I could care less about, they already cause battery drops every time they pull current, even without diodes in.
*(3S - so 11.1V nominal - 2100mAh seems to be the ‘sweet spot’ for the moment, and because if I go LiPo I want longer run time then my 7.2V 3000mAh SubCs, I will be putting at least two 2100mAh LiPos in parallel.)
Really at the voltages and currents being used here it probably makes little to no difference. Your opion seemed relative to your working with power supply systems and I was mostly just attempting to illustrate that a power supply can actively compensate for the effect introduced by a series diode through regulation where as a battery system can not. In the simple case of just the motors being powered by the series diode isolation the effects will not be significant.
Why would you choose to wire two seperate 2S1P packs in parallel instead of just buying a 7.2V 4200mahr 2S2P pack wired correctly? This is a bad practice for Li-ion cells, although perhaps not as dangerous at 2S as anything larger. A 2S2P pack is normally wired as 2 series of 2 cells in parallel, not 2 series cell strings in parallel. Done properly the cells are (hopefully) matched and charge balanced before they are wired in parallel. Then each set of cells wired in parallel are wired in series. There are good reasons for doing it this way, a lot having to do with cell balance as the cells age. Advanced Li-ion controllers actively balance the cells in a string to the same voltage within a few mV. This is a large part of why you see 3S and larger packs having “taps”… so the chargers can both protect against cell over-voltage and actively balance the cells. Wiring these 2S1P packs as you describe prevents this from being done properly. Yes, diode isolating the packs prevents the peak instantaneous currents between them when you place them in parallel, however you need to make sure each pack alone is rated for the entire load current since only the pack with the highest charge level will be delivering current until it’s terminal voltage comes down to that of the other pack level.
This is the first time I’ve heard of these issues in the real world.
Ah, got it. Yes, you are right.
And I am not looking at 2S LiPos. I am either looking at 3S or 4S. If I only wanted 7.2V, I have left over stick packs from RC stuff. I need to run motors higher, which is why I am doing all the LiPo research. I have several reasons for not wanting a 2P or greater pack:
- I cannot monitor individual cells in a pack 2P or more (yes, I plan to have a microprocessor on every balance tap in the system for monitoring)
- Size and Weight: Smaller packs can be used individually on smaller bots (like hexes), and used in parallel on larger bots, whereas bigger packs can’t be used on the smaller bots.
- I would rather get a larger capacity series pack then anything wired in parallel at the pack level. (this is just personal bias)
I do plan to take the proper precautions with LiPos, and will be using a balance charger.
In addition, the current draw for small robots isn’t going to be that big. If I build a tank style robot with two PGHMs, the stall current is only about 7A. IIRC, in another thread, Jim said that 18 HS645 servos in a walker take about 8A. Even at 8A, in a 2100mAh pack, thats only a 4C discharge rate. All the LiPos I am looking at are rated at a minimum of 20C continuous discharge. So, I think I am well within safety margins.
I think, and this is just an opinion, that so long as you are using a good charger on cells with taps for protection/balance and your equipment is LiPo aware with a minimum cutoff voltage of around 2.9V/cell you should be ok. This is pretty similar to the R/C airplane approach to LiPo mangement… hopefully you will not have the potential for physical damage like they do though. 
This kind of stuff gets interesting when you start playing with designs that allow for hot-swapping power supplys… causing imbalances or a “glitch” with several hundreds of watts pulling from a power bus can get, um, well it’s not a good thing to do and some people get grumpy when it happens.
I seem to remember TI had some chips designed specifically to deal with these kind of issues but it’s been like 5 years anyway since I worked on this kind of thing so… 
Don’t want to go too far OT though.