Mixing battery types in series

Joined
Mar 30, 2007
Messages
645
Location
San Diego, CA
Are there any complications or dangers involved when mixing battery types in series? Like lead-acid and lithium? Assume each pack is already wired for individual use. The only things I'm aware of is that the total current for the system will be determined by the lowest-current pack, and that each pack should probably be charged one at a time... correct?

EDIT: I thought of another complication. Obviously lead-acid won't last as long as lithium, so it would be nice if the bike would simply stop using the lead-acid batteries when they get too depleted and just use the lithium. But this won't happen automatically, right? Instead, it would keep attempting to draw power from the lead-acid batteries if the lithiums are still giving enough voltage, possibly permanently damaging the lead-acid batteries... right? So I guess a user-operated switch should be used to toggle the lead-acid on or off? Like in the attached picture. What's a good switch to use for this? I've never used one before.
 

Attachments

  • PbLi.png
    PbLi.png
    1.1 KB · Views: 3,456
Yep. This is what I did on my Currie. Adding batteries in series usually negates the controller's natural low-voltage cut-off protection. Hence it becomes very easy to over-discharge the cells of least capacity. Since the other type will be still be going strong, terminally fatal cell-reversal is a bigger risk than if the types were all matched in capacity and chemistry. To obviate the risk, the battery of least capacity can be monitored with a voltmeter on the handlebars, or some sort of low-voltage bypass like you mentioned can be implemented.

I wired 8 volts, 33ah in series with the 24 volt 22ah SLA's On my Currie. The SLA's drain well before the lithium -- a good thing as lithium cell reversal isn't pretty, or cheap, and I know when the SLA's are getting low by the bike's sluggish response that the extra 8 lithium volts won't fully make up for. Therefore I didn't need to add a voltmeter, a bypass or anything like that (since the bike won't run on just 8 volts).

Keep in mind that charging different types simultaneously requires more than one charger (obviously). So the chargers must be electrically isolated by design (like Soneils), or by lack of third ground prong on their wall-socket plug (most but not all other chargers).

PS: just saw your edit and am contemplating the schematic and switch question....
 
I'd be using a Soneil for the lead-acid and some other charger (obviously lithium-specific) for the lithium. Is it enough for just one of them to be isolated? If not, there's no harm in just charging one pack at a time, right?

Actually, since I'll be using that toggle switch anyway, I can completely isolate the battery packs from each other in which case it's a moot point. After further research, I'll use a double-pole changeover switch instead of a singlepole like in my picture.
 
The schematic is ambiguous, but looks to me like you've got a short across the lead battery when throwing the switch from lithium-only to lithium-lead series. If you or somebody else doesn't beat me to it I'll draw one.
 
Sorry. The 'pivot' or input of the switch is the lithium-ground wire, and it toggled between output-ground or lead-positive. However I'd actually use a doublepole as shown in the attached picture. Any risks?

If it's not obvious, there are two possible paths: blue or pink. Pink is lithium-only; blue is lead and lithium.
 
A slightly off topic, but I think a related, question comes to mind reading this thread.

What do you guys think about adding a new set of batteries in series with an older set of the same type?

EG: I have a 36V 12Ah NiMH pack with 340.9 total Ah @ 38 cycles (DB V1 info).

In looking to build a new rig (5304, 72V 12Ah NiMH) what are the pros and cons of getting one additional 36V pack and adding it to the existing pack, verses starting with 2 new 36V packs and selling off the old setup?

Greg
 
I'm not understanding your schematic. Here's how I'd wire it using a single SPDT switch or relay.
 

Attachments

  • SPDT-bypass.jpg
    SPDT-bypass.jpg
    20.1 KB · Views: 3,429
What do you guys think about adding a new set of batteries in series with an older set of the same type?

That's the way I'd do it, particularly with NiMH which doesn't suffer as much slow, progressive loss of capacity as lithium-cobalt. Nickel batteries do lose capacity, but die usually by shorting internally some time between 5 and 10 years of age. Your packs should be matched closely enough to where the slight difference would only become a problem if you were running them down farther than you ought to be anyway. So far as I know, to limit damage risk discharge should be stopped at or before 1.0 volts/cell on nickel batteries (NiMH & Nicad).
 
CGameProgrammer said:
That diagram makes no sense. You have the "A" path connecting the positive controller terminal with the ground lead-acid terminal.

Doesn't matter because the minus lead-acid terminal is connected to the + lithium terminal, and the circuit through the SLA is broken when the switch is on "A".

On "A" all power is supplied by the lithium battery only. On "B" power is supplied also by the SLA in series with the Lithium.
 
I believe that this is proberly the easiest way to go as the original picture was drawn it will isolate the two batteries for charging as well
 
CGameProgrammer said:
This is what my doublepole implementation was. Hopefully it's clearer. Are there any short-circuit risks with this?

Much clearer. I don't see any problems. Looks perfectly workable. As they say, more than one way to skin a cat, or skip a batt. :) I'd make sure the switch contacts are rated for at least 20 amps.
 
This thread ought to be in "Battery Technology".

:arrow: Whoever is moderating needs to move it.

I'd like to add "Sag Compensation" to my SLA battery by adding a few NiMh batteries in series... (this is already in the "Battery Technology" area) but not to trigger them until the SLA cells have already gotten into their voltage sag state later on in the ride. This way you "refresh" the voltage to what you started with. For those of us with gears that get used to shifting at certain speeds and on certain hills this would restore it's original performance. Just 1.2 volts is enough to compensate for the sag and it can be done with 4 NiMh 10 Ah cells hooked up in parallel. I was thinking I'd use Dean Connectors and have to pull over and reconnect it manually... unless there's a good way to do it that isn't expensive or complicated. ???
 
You could use a relay instead of a switch so you could control the relay from the handlebars. The relay could be automatic, but that would be complicated.

Rather than compensating for sag by switching batteries, you could just use a bigger battery full time and back off on the throttle a bit until you need the extra juice.
 
fechter said:
Rather than compensating for sag by switching batteries, you could just use a bigger battery full time and back off on the throttle a bit until you need the extra juice.

I was going to simply add the extra batteries in series when the voltage began to drop.

So you would go from:

12V 38Ah + 12V 38Ah + 12V 38Ah

...to:

12V 38Ah + 12V 38Ah + 12V 38Ah + (4)(NiMh 1.2 V 10Ah in parallel)
 
xyster said:
I'd make sure the switch contacts are rated for at least 20 amps.
Is that enough? The system uses 35-40 amps sustained at what will be 96V. Would something rated for 20 amps (and I assume at least 110V) be enough?
 
CGameProgrammer said:
xyster said:
I'd make sure the switch contacts are rated for at least 20 amps.
Is that enough? The system uses 35-40 amps sustained at what will be 96V. Would something rated for 20 amps (and I assume at least 110V) be enough?

Yes because assumedly you'd not be switching under full load -- which is what the amp rating basically refers to. For almost a year now I've used a $1 20 amp SPST from Home Depot on my 35amp 80 volt bike without problem.
 
Back
Top