The CATL Qilin CTP 3.0 is their second generation cell to pack design. Qilin is named after a legendary creature from China.
The latest CATL post suggests that this integrated system can increase the energy density to 255Wh/kg for ternary battery systems (NMC, NMCX etc), and 160Wh/kg for LFP battery systems. Essentially removing the overheads of a module.
The pack design on the outside is quite traditional with a tapered section at the front to accommodate crash and suspension structure in a normal vehicle design.
The pack has bolting locations down each side and at the back edge of the pack (bottom of this image).
Note that there are no fixings in the centre of this pack. This is surprising for a pack of this size.
“In the CTP 3.0 battery, the internal crossbeam, liquid-cooling plate and thermal pad have been integrated into a multifunctional elastic interlayer.”
The sandwich layer with the cells and cooling plates running across the pack along with a lower and upper closing panel is not really described in enough detail to understand how this is assembled. How are these parts fixed to each other?
The new CTP3.0 design appears to rely on the structure in the cells and the cooling system to react the side impact loads seen in a crash.
This image doesn’t show the detailed mechanical structure around the cell connections and the coolant hoses. How do the coolant pipes connect to each plate and how is that held mechanically?
Earlier in the video this is described as a system that allows them to remove the internal crossbeam from the pack. That means the cooling plates and cells must tie into the pack side beams in order to react the side impact loads.
Large packs tend to have a “panting” mode and hence need structural beams to add stiffness. Maybe the cells are bonded to the lid/upper panel?
The cells point downwards and this is sensible from a venting point of view as it directs particles and gases away from the passenger floor. However, there is not enough detail to understand how the cells are supported mechanically or how the electrical connections are made.
The pack baseplate and upper lid/panel could work as shear planes, but with no structure in the centre of the pack for these to tie into they would probably crumple.
“The internal crossbeam, liquid-cooling plate and thermal pad have been integrated into a multifunctional elastic interlayer”. The area of the cooling plates has been increased by a factor of 4. However, the cooling plate is now applied to the large face of the prismatic cell and hence the heat in the core of the cell now has to conduct through the active layers. The thermal conduction through the active layers is around 30x lower than the conduction in-plane.
The elastic interlayer cooling plate can be seen being compressed by the cells as they expand. The cells will continue to expand over the lifetime of the pack. It will be interesting to see how the cooling plate end connectors are designed with the stress that this compression will generate in the joint.
Also, this means the cooling channel width will decrease as the cells age and expand. Hence restricting the flow of cooling fluid. As cells age the internal resistance increases and so the amount of heat increases for a given drive cycle. Less cooling as the cell ages will result in much higher cell temperatures. Higher temperatures will result in more rapid cell ageing. Thus forming a positive feedback.
As the cooling plate is compressed the fluid has to go somewhere and hence the header tank will need to be large enough to accept the fluid that is displaced.
Each cooling plate has a connection at each end. Therefore, lots of possible leak points in this design.
Some big claims.
A lot of the details are missing. Issues and quite a few questions that need proper answers before we can really see how this could ever work:
- Can the cells and cooling plates replace the cross-pack structure?
- How do the cells and cooling plates connect into the side beams of the pack?
- The cooling is throttled as the cells expand with age. Lower coolant flow as the cell resistance increases doesn’t appear to be sensible.
- How do they ensure a robust cooling connection to the extruded cooling plate?
- There is no structural detail explaining how the cells are located vertically.
- Lots of cooling connections each side of the pack, lots of possible leak points. In side impact all of these connections are very vulnerable.
These are my first observations on the CATL Qilin CTP 3.0 and we will expand our analysis and understanding as more data becomes available. Please comment below.
Early numbers for the Zeekr 001 and the CATL CTP pack are showing that it is achieving 200Wh/kg.
CNEVpost  give some initial values:
- 1032km at 14.9kWh/100km = 153.8kWh
- However, I assume the 14.9kWh/100km is at the plug and hence the vehicle is achieving roughly 14.9 x 0.9 = 13.4kWh/100km
- 1032km x 13.4kWh/100km = 138kWh
We need more data and cell specifications to confirm this, but this is looking like the highest energy density automotive pack in production.
- CATL CTP battery pack thermal characteristics, Dr Kun
- Zeekr 001 with CATL’s Qilin Battery will have 1,032 km range, CNEVpost
The BYD blade cell to pack design is perhaps more interesting as it has been designed by a company that understands vehicle design. Also, this pack design is used in production in the BYD Han.