Lithium Starter Batteries

How do you store and manage your electricity?
Post Reply
wedge
Noobie
Posts: 4
Joined: Mon, 26 Dec 2011, 22:24

Lithium Starter Batteries

Post by wedge » Wed, 02 Jan 2013, 03:02

I have been trying to read as nuch about lithium batteries as possible, and I see they are starting to be used as starter batteries for ICE cars and motorcycles (mainly by racers).

However, their use in this application seems to contradict normal EV practices. For example check out Braille GU1R

This is a LiFePO4 battery. The manufacturer claims full charge voltage is 13.2v. Fair enough, that's very conservative, only 3.3v per cell. However, the maximum output from most alternators is 14V, some even as high as 15V and Braille claims their batteries are compatible with standard charging systems. Is it possible they have some circuitry inside that disconnects the battery when the alternator output goes over 13.2V?? Seems unlikely to me as most alternators put out 13.8-14v at idle, so the battery would never get charged. How do they stop the battery from being over-charged?

Other interesting points:
- No mention of low voltage cutoff on their website.
- Zero mention of cell balancing. No balancing terminals as far as I can see.
- The lithium chargers on their site seem quite crude, and they quote a charger output of 13.8-14.4v!!!

It has me intrigued, I'm wondering how the companies making these batteries are not going broke from warranty replacements.

User avatar
acmotor
Senior Member
Posts: 3591
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

Lithium Starter Batteries

Post by acmotor » Wed, 02 Jan 2013, 08:18

Hi wedge,
It is good that you do your research.
As you have seen, there are numerous varieties of 'lithium' cells, even variations within the LiFePO4 family.

The LiFEPO4 cells have typically 2.5V per cell min operating voltage, nominal fully charged at rest of 3.2vpc and max voltage up to around 4.25vpc. This max voltage varies between manufacturers and chemistry fine tunes but the 3.2V nominal is standard.
Once a cell is over about 3.4vpc it is fully charged and the voltage then up to max represents almost no further energy stored and in fact this upper voltage is considered an area to be avoided, at least regularly, for cell longevity.

When you put 4 cells in series (preferably of the same Ah capacity) then you get a '12V' battery similar to a lead acid battery often used in ICE vehicles, and indeed, as it happens, in most EVs as the auxiliary battery.... there is a reason for this e.g. in iMiEV and Leaf. see below.

This 4 cell battery can operate quite well without any battery management system, at least under user care.
It is desirable to start with all cells in a battery fully charged individually if there is no battery (cell) voltage management system to help reach ballanced charge.

                      charging               total        discharging           total
min absolute                                         2.5 +2.5 +2.5 +2.5   10.0
min worst case                                      2.5 +3.2 +3.2 +3.2   12.1
nominal     3.2 +3.2 +3.2 +3.2    12.8    3.2 +3.2 +3.2 +3.2   12.8
max worst 4.25+3.2 +3.2 +3.2   13.85
max abs    4.25+4.25+4.25+4.25 17.0   

(hope the layout reads OK)

Battery at typical lead acid float voltage of 13.8V
          3.45+3.45+3.45+3.45=13.8
LiFEPO4 is quite happy to sit at this voltage fully charged

You can see that if battery voltage is between 10 and a bit over 14V then all is well. Provided of course that no cells are faulty.
Even if one cell went short circuit (common failure mode) then the battery will just supply 9.6V nominal but not be in trouble with a 13.8V float voltage as this will supply around 4.6vpc that will probably harm but not 'ignite' battery.

Once a LiFEPO4 battery of 4 cells has been top balanced (balanced at full charge) it will return to this balanced condition quite well even after being discharged until one cell has reached 2.5vpc and then recharged again.

A number of suppliers currently sell '12V' lithium batteries with no BMS and this works quite well. Still, at least a top balancing BMS would be useful and can be of a very simple resistor ladder or resistor/zener type.
Resistor ladder e.g. 330ohm resistor across each cell. This causes 10mA discharge but will equalise cells over time when charging.
Zener 3.3V and 100ohm resistor across each cell will also work.
But still, the 12V battery can work without any of this.

I would prefer a full BMS of course for maximum battery life.

So charging off an alternator is OK in voltage and as alternators are current limited in design so battery charge current is fine.

The problem comes when the lithium battery is flat (discharged) as you point out. Some battery packs e.g. ebikes have cut off systems but they deal with relatvely low currents. An ICE battery will need to supply many hundreds of amps and the switching silicon and BMS would add to losses and the cost of the battery so is not commonly fitted.
The user must stop using the battery when its output voltage drops.
Some ICE ECUs will stop you cranking if battery voltage is low anyway.

In general use, with an ICE that is working ! the battery does not use more than a few % capacity to start the motor so it all works. Don't try to start after leaving the headlights on though !

The reason for Lead Acid 12V aux battery in iMiEV and Leaf etc is probably that it can be flattened by lights and accessories etc with less risk of damage than a lithium battery run flat. Also that it can then be jump started without danger of charging a suspect lithium battery and perhaps another point that it can be swapped at the local servo if required.
True, a battery management system could cut off a lithium battery but then the more familiar run down to dull but still working lights may be prefered to an abrupt shutoff for safety reasons.

Does any of that help ? Image

edit: one other point is that a shorted lithium battery may not be a good idea so a fuseable link in the starter motor line may be advisable.
Last edited by acmotor on Tue, 01 Jan 2013, 21:24, edited 1 time in total.
iMiEV MY12     102,980km in pure Electric and loving it !

wedge
Noobie
Posts: 4
Joined: Mon, 26 Dec 2011, 22:24

Lithium Starter Batteries

Post by wedge » Thu, 03 Jan 2013, 02:46

Thanks acmotor that does help. I don't understand why Braille states their full charge voltage is 13.2V, when this will be exceeded in virtually all ICE cars?

A lot of newer ICE cars have electricsl systems designed for calcium lead acid batteries, which need a charge voltage of 14.7 volts, so 3.68 volts per cell assuming all cells are balanced. I also thought the float charge (constantly maintained) voltage for LiFePO4 should be lower than the regular charge voltage, but I can't remember where I read that.

User avatar
zeva
Senior Member
Posts: 421
Joined: Sat, 15 Dec 2007, 05:09
Real Name: Ian Hooper
Location: Australia
MSN: sigmunky@hotmail.com
Contact:

Lithium Starter Batteries

Post by zeva » Thu, 03 Jan 2013, 03:40

13.2V does sound a bit low for LiFePO4s (3.3V/cell isn't fully charged).

FWIW personally I'm in favour of bottom balancing when one does a 4-cell pack without BMS. The reason for this is that overdischarging is one of the most common reasons for damaging lithium cells, and occurs if one cell goes flat before the others but the others keep current flowing through it - driving it negative, and quickly doing damage. If they're top balanced, one tends to go flat first while the others still have significant voltage and current keeps flowing.

But if you bottom balance, they all go flat at the same time and none can be forced negative. I had actually had this happen in my RX7 EV just the other day - the little lithium 12V (4x LFPs) I built as the aux battery had gone flat and was down to 2V, but all individual cells were still above zero. (It is not great for the cells' cycle life to be discharged so deep, but is not instantly damaging like being forced negative.)

At the high end it is likely that you'll get one cell hit full first and have its voltage significantly higher than the others, but assuming capacity is *reasonably* well matched it'll be something like three cells at 3.35V and one at 3.75V. AFAIK damage doesn't occur with LiFePO4s until >4.2V, so there is a reasonable safety margin.

That said, bottom balancing may not be safe if the charge voltage is much above 14V, as one cell could easily exceed 4.2V and be damaged. So in that case top balancing with a LV cutout device may be a better option.
Ian Hooper
--
"Never doubt that the work of a small group of thoughtful, committed citizens can change the world." - Margaret Mead
http://www.zeva.com.au

User avatar
acmotor
Senior Member
Posts: 3591
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

Lithium Starter Batteries

Post by acmotor » Thu, 03 Jan 2013, 16:21

13.6 to 13.8V would be the typical Lead Acid float though often ECUs take the voltage to 14+V / cold weather after ICE start to push some charge back into battery.
This may be battery kind but doesn't do much for the vehicle lighting for instance which will often be rated around 13.2V (13.8 - wiring losses). In fact that is where the 13.2 number may have slipped into the battery spec. Just checking some headlight globe packets they state 13.2V

wedge, remember that the normal ICE is used less than 5% (and a taxi up to 50%) of the day so the lithium battery is not being held at >3.2Vpc for much time.

Ian, RX7 aux battery total discharge I guess backs up iMiEV and Leaf decision to stay lead acid for the aux battery. The real risk for user is that one cell may suffer a catastrophic and dangerous failure on attempted recharge. I bet you paid plenty of attention when recharging those cells. Not every user is that savvy.

You'll see that from the table in my post that one cell going below 2.5V already produces a battery voltage below 12V and the battery voltage will go below 10V before a single cell is reversed. This is under light load say <1A. Under heavy loads, all cells are below 3.2V already thus making LV detection even easier.
Thus no balancing is actually required if the load is intelligent or user takes care not to discharge too deeply.

For instance 12V to 230V inverters shut down at around 10.5V.
ICE vehicles may not be that smart but then the instance of flat battery by the user is rare nowadays. (less likely than a flat tyre)
Maybe too much music in the carpark ?
If there is a flat battery from battery fault then it is history anyway. Though a DIY 4 cell may be repairable.

Bottom balancing is simply not user friendly to maintain and makes the dangerous assumption that all cells are at the same temperature and age the same so that once the bottom balance is set up that it will continue.
I can see the thinking for bottom balance to allow for over flattening a lithium battery without reversing a cell.(hardly a good practice) But as you point out, that only proves you need a LV cutout.

Image
iMiEV MY12     102,980km in pure Electric and loving it !

Rusdy
Groupie
Posts: 169
Joined: Mon, 09 Jun 2014, 16:45
Real Name: Rusdy
Location: Australia
Contact:

Re: Lithium Starter Batteries

Post by Rusdy » Fri, 23 Mar 2018, 09:44

I replaced the starter battery of my Corolla (then sold to a friend of mine about the same time) 3-ish years ago with 40Ah battery pack from EV power. At the time, it was awesome, obviously being an LFP with such a high cranking amp.

In the last couple of weeks, however, a friend of mine told me the car struggled to start. True enough, when the starter motor was cranking, the battery pack voltage dropped to 7-ish Volt. I've replaced it back with good 'ol lead acid and back to normal.

In the last couple of days I've run some test to the battery as-is (not charged to full 14.4V):

Battery pack voltages prior test (as-found), with pack voltage = cell1 + cell2 + cell3 + cell4:
13.74 = 3.43 + 3.44 + 3.44 + 3.43

Then I discharged it with roughly 22Ah worth of light load (3A in 3 different stages, and then 1A in final stages), and final pack voltage:
11.94 = 2.92 + 3.13 + 3.13 + 2.76

I checked the battery internal resistance in various stages of discharge, and found pretty damn consistent, i.e.:
Cell 1 internal resistance = 10 milli-ohm;
Cell 2 internal resistance = 9 milli-ohm;
Cell 3 internal resistance = 9 milli-ohm;
Cell 4 internal resistance = 24 milli-ohm.

In summary, the battery capacity has at least 60% of the original, with one of the cell's internal resistance shoot through the roof. Cell-4 is the one that faced the engine. I wonder whether that was a coincidence, or the radiant heat from the engine killed it.

Looks like I underestimated the heat inside the engine bay. I was so confident back then that the LFP would survived the heat and last longer than the car. I was wrong, sad :cry: .

antiscab
Senior Member
Posts: 2208
Joined: Mon, 26 Nov 2007, 05:39
Real Name: Matthew Lacey
Location: Perth, WA

Re: Lithium Starter Batteries

Post by antiscab » Fri, 23 Mar 2018, 10:24

I'd love to take that battery off your hands if you're willing to part with it. There's a certain prelude conversion that would benefit from a light starting lighting battery.
Matt
2011 Blade Electron mk6
2007 vectrix - 145'000km
1998 Prius - needs batt
1999 Prius - needs batt

User avatar
Richo
Senior Member
Posts: 3249
Joined: Mon, 16 Jun 2008, 00:19
Real Name: Richard
Location: Perth, WA

Re: Lithium Starter Batteries

Post by Richo » Fri, 23 Mar 2018, 12:37

I thought the prismatics were to be charged 4.1-4.2V.
So the battery never got a full charge then had to put out high amps.
Quite likely the extra heat caused the closest cell to wear faster.

Perhaps move onto Cylindricals(headway) or LTO's.
Help prevent road rage - get outta my way! Blasphemy is a swear word. Magnetic North is a south Pole.

Rusdy
Groupie
Posts: 169
Joined: Mon, 09 Jun 2014, 16:45
Real Name: Rusdy
Location: Australia
Contact:

Re: Lithium Starter Batteries

Post by Rusdy » Mon, 26 Mar 2018, 11:36

Hi @antiscab, check your private messages.

Hi @Richo, looks like the chemistry is it (instead of the physical whether they are rolled or pouched). I ran a full charge regime just for curiosity and the voltage ramped very quickly when it hits 3.5-ish Volt (per cell). Once all cells reached 3.6 Volts, it couldn't absorbed more. I cranked up the CV supply and the battery pack simply followed the supply (with the cell balancer started to conduct).

What I've found interesting is the internal resistance is pretty much flat across charge state (+/- 10% with my bum calibrated instrumentation and method). Googling for whitepaper experiments also shown relatively flat internal resistance. So, for starter battery purpose, LFP is a perfect battery! If it can withstand the heat stress of course ... :oops:

User avatar
mikedufty
Senior Member
Posts: 494
Joined: Tue, 02 Dec 2008, 00:15
Real Name: Michael Dufty
Location: Perth, Western Australia
Contact:

Re: Lithium Starter Batteries

Post by mikedufty » Mon, 26 Mar 2018, 12:23

I put a commercial LFP replacement starter battery in my motorbike quite some time ago. The original actually died in a couple of years, but I did get a warranty replacement and the new one seems to be going great. Sits behind the airbox under the seat, probably better for heat stress than a car engine bay.

User avatar
Richo
Senior Member
Posts: 3249
Joined: Mon, 16 Jun 2008, 00:19
Real Name: Richard
Location: Perth, WA

Re: Lithium Starter Batteries

Post by Richo » Mon, 26 Mar 2018, 12:55

But the prismatics havs sloshy fluid and the pouch/cylindrical are solid.
I would expect the pouch/cylindrical to always fair better in the automotive environment.

I had been thinking of getting 10/12 LTO batteries to replace my starter battery.
$144.
Not that much difference in price to a SLA.
Even if only lasted the typical SLA 2-3 years I'm no worse off.
Help prevent road rage - get outta my way! Blasphemy is a swear word. Magnetic North is a south Pole.

Rusdy
Groupie
Posts: 169
Joined: Mon, 09 Jun 2014, 16:45
Real Name: Rusdy
Location: Australia
Contact:

Re: Lithium Starter Batteries

Post by Rusdy » Mon, 26 Mar 2018, 13:53

Richo wrote:
Mon, 26 Mar 2018, 12:55
But the prismatics havs sloshy fluid and the pouch/cylindrical are solid.
I would expect the pouch/cylindrical to always fair better in the automotive environment.
I thought pouch is the solid one (i.e. polymer elctrolyte), where as prismatic and cylinder are 'sloshy' one (i.e. solvent electrolyte)*? I also thought the heat degradation is a function of electrodes chemistry (not the electrolyte)? Ugh, my head spins when it comes the physic of lithium batteries! :geek:
Richo wrote:
Mon, 26 Mar 2018, 12:55
I had been thinking of getting 10/12 LTO batteries to replace my starter battery.
$144.
Small Ah I presume?

If only I can find a good reliable interface between the good 'ol battery terminal style to a lithium replacement. I had to re-crimp the end terminals of my previous Corolla. I definitely don't want to do that with the Leaf (the battery terminal is a beast with a built-in fuse).

User avatar
Richo
Senior Member
Posts: 3249
Joined: Mon, 16 Jun 2008, 00:19
Real Name: Richard
Location: Perth, WA

Re: Lithium Starter Batteries

Post by Richo » Tue, 27 Mar 2018, 13:01

The LTO's are 2.4V 2.9Ah 60C peak.

So 10 would be 12.0V/13.5V 5.8Ah
And 12 is 14.4V/16.2V 5.8Ah.

12 would be the obvious choice.

https://www.ebay.com/itm/2-4V-2-9Ah-Tit ... 2658741688
Help prevent road rage - get outta my way! Blasphemy is a swear word. Magnetic North is a south Pole.

Post Reply