BMS for DIY motorcycle starter battery.

I've seen no changes in the current offerings. There was a battery conference in Orlando a few weeks ago addressing safety etc.
The big players do all the UN 38.3 testing and that is all we have.. If there are application specific issues of vibration / shock ,then the user should do what he can to insulate and minimize it.

Lithiummoto said:

Well, if this thread is still active, and why would'nt it be ??
Let me just say, I've been making LFP starter batteries for 10 yrs now, so, what ever needs starting, I can do it.
Naturally, its going to cost you more than if u do it yourself, but, thats not the issue here as people didnt know where to get something made.
The link is to me; lithiummoto.com
I'm in florida but, can ship worldwide.
not selling, just telling.
contact me if you are in need.

Click to expand...

5 to 10 year life? I'd stick with AGM instead.

Have been used it for more than 10 months ago (LFP)
No any bms added, just simply active balancer 1.2a.
I use it on my 125cc motorcycle to replace the old lead acid battery 12v5a.
Since your the altenator voltage in standard and normal condition I believe every thing gonna be OK.
Te maximum alternator voltage is between 14.2v-14.4v

Vicriz said:

Have been used it for more than 10 months ago (LFP)
No any bms added, just simply active balancer 1.2a.
I use it on my 125cc motorcycle to replace the old lead acid battery 12v5a.
Since your the altenator voltage in standard and normal condition I believe every thing gonna be OK.
Te maximum alternator voltage is between 14.2v-14.4v



View attachment 93733

Click to expand...

That is very cool. Can you please share links to the cells and the balancer you used? How large of an engine would this start?
Thank you, Doug

Little bit of a odd thread.

First someone wants to see a CCA rating on an LFP data sheet, which is not a great sign for someone seeking to design a new high current application from the ground up. If you’re seriously considering this you really need to know how CCA is calculated and be able to get a comparable figure from a LFP voltage curve. If that’s not in your skill set then I would suggest sticking to projects where you’re mostly tweaking someone else’s design for the time being.

But the response… Yes, LFP will struggle to provide CCA level amps for any period of time. But anyone claiming to know about batteries should know that CCA is not a realistic draw for a starter and that the rating exists solely due to the quirks of LA chemistry. And they should also know that real cell voltage != output voltage under load.

Anyway, apologies for the negativity, so let’s try to calculate a realistic estimate of the C rate an LFP might need and be able to provide in this application.

So as stated a 260 draw on a 7ah battery is 37C. No one can deny that’s really sucking it down. But is it really implausible? Well the first data sheet I found rates an 8ah cell as being able to provide 240 Amps for 30 seconds. So not quite there, but a lot closer than some would have you believe.


But of course current isn’t really the thing - we want power. 260A @ 7.2 v is 1.87Kw. So let’s pull a random set of LFP discharge curves:


In the middle of the curve we drop from 3.2 to 3.0 volts when going from 1 to 6 c. So let’s just swag that at 30 C it would be output about 2.4 volts. In a 4S 7 ah system that gives us 9.6 volts at 210 amps and 2.1kw. Now there are a lot of assumptions and this is running right on the ragged edge, but so is that CCA rating.

But then someone did us the courtesy of posting actual data on the starter draw and found a 75A peak non-transient draw. Thanks cal! That 75 amps is about 11 C, which is very reasonable for a short period. And the LFP will provide that current at a higher voltage than LA.

So I can’t believe I’m saying this, but 7 ah LFP is probably enough for the application. And given how cheap the cells would be, how much risk is there to trying it really? It’s just 4 cheap cells, and there’s minimal risk it will damage the bike. I get that a lot of people think of LFP as some hallowed object that needs to be carefully guarded, and that makes sense with a 320 ah system that can burn your house down with a short, but that’s not the situation here.

This is getting long, so I’ll try to add some thoughts about implementation in another post.

So the BMS. I count 4 options for any starter battery system:

1. Spend more than the cells on a massive BMS rated for the peak current.
2. Get a underspeced BMS and hope your starter finishes before the disconnect timer (usually 5 to 10 s) AND you don’t fry your transistors.
3. Skip the BMS, triple check the charge system parameters and trust it to stay there. Use a balancer.
4. Use a small BMS for continuous loads (lights, electronics etc) and charging but route the starter solenoid around the BMS.

3 is probably what I would do with such cheap cells, but 4 is the right way and much more interesting. You will need to rewire the starter (solenoid) for starters. Your battery pack needs an extra lead that bypasses the BMS and connects to the solenoid main input. And the BMS needs to control the positive side of the battery, NOT the ground. If it uses the ground then the disconnect will just redirect the ground current through the chassis into the starter ground wire. Do that and you can use a tiny BMS while still preventing over charging and discharging in all cases. Even the starter won’t hit if the BMS is disconnected because the signal wire will be disconnected.

Now we get to temperature. From what I can gather, cold storage over winter without a tender is the primary killer of LA batteries on bikes. If that’s you then congratulations- LFP should solve the issue handily. It stores just fine cold with low self discharge rates. But if you live in a cold place and ride in the winter you want a BMS with temperature control. I’m not that scared of trickle charging LFP below freezing, but the motorcycle can likely charge over 1C which is bad news at low temps. Option 3 is no option in this case.

The other risk is vibration. I would build the biggest box that fits then stuff it full of foam. Maybe even wrap the cells in some kind of heavy material to further deaden the acceleration the cells will see.

But I don’t think LFP is truly right for this application. For a little more money you can get LTO and have the ultimate in charge/discharge rates with incredible service life, zero temperature risks and much better tolerance to over discharge. Which means you can skip the BMS and rewiring your starter, and if you leave your lights on it just needs a jump. Just slap a balancer on there and you’re golden.

Something like 5S or 6S of these:


Or take a risk on Ali:

With 40ah cells you can start even massive diesels with them. Anyone saying lithium as a whole isn’t suitable for starting applications is grossly uninformed.

Edit: just remembered that with a BMS the charge system could fry your electronics if the BMS suddenly disconnected and caused a voltage spike. I’m guessing it would be no problem on a motorcycle, but you might consider some kind of secondary system like a capacitor to absorb that spike. Or implement a 2 BMS system that isolates charging from the electronics. Mostly it’s just another reason to choose titanate.