Dane.Kouttron

[10.23.19]  Lithium Titanate Module Conversion &  6C charging on a Bike

I purchased two large lithium titante modules from a mitsubushi japanese vehicle. 

What?
 Out
 Cl Time for Paint Brushless Motor Mounting First Test Testing Notes Conclusion Image Directory

Some Project Background
Some points of note about LiTi, its an oddball chemistry that, at the time of this writing, is not widely used. 
Why LiTi ? The real goal here is having 6C charge on an electric bicycle. What is 6C charging? Excellent question. 'C' is a normalization factor, the C-Rate is simply the capacity removed or added in one hour. A 1 'C' discharge completley discharges the cell in 1 hour. Whats excellent about this is it irelevant to the actual capacity of the module, as a 1C discharge for a 1Ah or a 100Ah battry is exactly the same, the discharge occurs over the course of 1 hour. This applies to charge and discharge. 
Whats so great about 6-C charging?
Well there buckaroo banzai, speed is excellent. For this project i'm aiming for 6C as it equates to '10 minutes'. A 10 minute recharge for a long range electric bike would be quite legendary, as at the moment conventional Li-ion is limited to 1C charge, or fully charge in an hour.
Whast going to charge this monster at 6C?
Thats an excellent question.


A Quick look at the existing modules

High charge-rate plotting.
Lets do some battery  maths!
Each submodule is a 12 cell array, with a total of 24 cells. The cell voltage ranges from 1.8 to 2.8v, and battery management wise I can only really take in 16 cells.
That gives us a pack with the following characteristics: [Charged] = 44.8v, [Discharged] 28.8v. Thats a fairly large voltage swing. Each cell is 20Ah nominally so that puts us at  36.8v (nominal voltage) * 20ah (cell capacity) = 736watt-hours.
Charger plotting:
To charge the 16 cell pack at 1C wouldnt be to terrible, as <1kw charging is feasible with modified LED power supplies. Meanwell makes a range of LED  supplies that covertyly work fantastic as a constant current power supply thats IP67 rated.
To charge at 6C, we need at least  6x that charge rate, or 4.4kw. Now things are getting interesting.
In the states, outlets are 110v and generally peak at 1.2kw, so the desire to push to > 4kw is not going to happen off standard mains.
Enter J1772, or level 2 charging.
Level 2 charging is not terrific for automobiles. Generally speaking, level 2 EVSE's are 220v AC at 30A max, or ~7kw max. Charging a full electric car at 7kw is a slow process, nominally a '60 kwh' smaller electric car would take 8.5 hours to recharge. Thats forever. For an electric bike however, level-2 is perfect.
What is level 2 charging?
There are a number of level-2 evse's available on the side of the road, but they are all nominaly the same. Internally they are a cell modem, a mains connection, a giant relay / contactor and some very light microcontroller hardware displaying some details about the charge rate or status of charger. There is no actual 'charging' happening in these EVSE's though, they are effectivley a complicated relay.
How to get to 6 C charging.
There are a few options here, nominally we're looking at a 50v supply that can output 100A in a constant current mode. Thats not an impossible feat, when we're talking about a lab bench supply. When we transistion over to 'fitting on an electric bike' thats where things become interesting. Generally speaking the smaller the supply the more its thermal density, and the higher the thermal density the increased propensity to use forced air. I think its unavoidable to not use forced air in this application, but i can be sneaky and only use it on an external facing location (on the outside portion of a heatsink) not internal to a supply.
So what options are available for off the shelf.
Cosel makes a 500w supply with an external voltage adjust, which could be used as part of a constant current feedback loop. The small cosel supplies could be stacked up to work in parallel to get to a higher output power, however that puts us in the 'fairly large kludge' category. 8 to 10 of these in some kind of thermal distributing case could get unwieldy quickly. Finally isolation would be really appreciated, The external heatsink should very much be grounded and not lightly electrocuting the user.
What other things should i design around
If this were a consumer device, i'd probably spend a lot of time worrying about power factor. Power factor is how the device under test looks to the power line. An inductive load has a power factor <1 and a capacitive load has a power factor > 1. Its unclear if advanced chargers will see an out of spec power factor and discontinue charging, so, some effort should be made to ensure compatibility.
User Interface
I'd like to see whats going on during a high-power charge, and have the ability to curtail the charge rate, either manually or with a battery temperature feedback loop. Initially my thoughts are 'Enable / Disable charging, charge rate setpoint, readout of battery temperature, charger temperaure,  and any kind of fault status. Finally J1772 does provide a charger handshake to indicate what the EVSE is capable of. There are some evse's that may limit to 3kw, in which case the charger should try to throttle down to whatever that threshold is.
Size
I really like the 'battery pack is a slide out module with a handle' approach to electric bikes. My "city bike" implemented this, and when implemented well really is great for modifications. My thought was to use the flat surface that sits along side the battery for the charger, so something long and flat. Unfortunatley J1772 connector is a bit large, so possibly utilizing the wheel well to hold the charger plug would be a good plan.
Interfacing

Digikey Shoe-shopping
As an exercise its sometimes easier to just see what presently exists and possibly design around it. The following is a few options i found when digging up off the shelf power supplies.

Digging into Lithium Titanate Chemistry
For the sake of having some intuition about these cells, lets 


N


S

(There's other photos in the photo gallery)
Concluding Remarks:

If you have questions or comments, ask below or send over an email.
Comments:
HTML Comment Box is loading comments...

Stay safe when working with electrons in aqueous / semi-aqueous environments. Also wear sunscreen, I'm not responsible for your newly aquired winter-farmers tan : ]

Dane.Kouttron
Rensselaer Polytechnic Institute 
Electrical & Electrical Power
631.978.1650