Case Study 1 – Joining Components of a Cold Plate using Structural Epoxy
Case Study 1 – Joining Components of a Cold Plate using Structural Epoxy
Project Background
During the period the R&D team set out to understand and develop a way of joining the two or three components of a cold plate by using a structural epoxy. If successful this technology could be used to produce small and large form factor cold plates. The large form factor cold plates used for electric vehicle battery cooling are often too large for conventional joining techniques such as FSW (Friction Stir Welding) and vacuum brazing.
As well as ensuring the materials join, the R&D team would have to ensure the product met the requirements of pressure testing. Consideration would also be required as to the input and output connections. The projects that would be considered for this type of cooling would require a simple serpentine liquid flow path, specified parameters would be met in respect of power dissipation and liquid temperature flow rate of the coolant. The liquid cold plates would need to meet specific pressure tests in the region of 9bar. Consideration of price would also have to be factored into the development to ensure marketability of the overall product.
The R&D team looked at the potential to manufacture the plates out of two and in some cases three parts. Where the form factor was small the liquid flow path could be CNC machined into the lower part and then a simple flat lid could be fitted. Where the form factor was too large to easily fit onto a CNC machine, the liquid flow path could be water jet cut from a sheet and a simple flat bottom and top could be fitted. Input and output locations could be fitted with screws and O-rings.
In each case either the two or three parts would then need to be joined together to ensure a robust product.
Technological advancement
The joining of two or three components to form a liquid cold plate with no welding, allowing for thermal performance and zero joint leakage.
Technological uncertainties faced and overcome
Initial industry reaction was sceptical as to the practicality of bonding plates together particularly when having to produce leak free joints capable of passing pressure testing.
There were also barriers to consider in respect of selection of the structural epoxy, the amount and spread that would be needed (joint design rules), the method of dispensing the epoxy, how to physically locate the parts together, how to hold them and how to cure the epoxy.
In order to overcome the barriers a simplified cold plate was designed which could be changed to help maximise joint design rules. Several epoxy product manufacturers were approached in order to select the most suitable epoxy, to investigate how the product could be dispensed and how it could be cured.
To facilitate the physical location of the parts together, to clamp them during epoxy cure time and to add some structural strength it was decided to add some screws. To match the thermal coefficient of expansion, to reduce weight and to enable easy fly-cutting of the surface if needed it was decided to use countersunk aluminium screws.
Structural Epoxy adhesives
Some of the products were two part epoxies that would require accurate measuring and mixing and had a limited life once mixed, others required heating in an oven in order to cure.
In view of the fact that the single part, temperature cured epoxy adhesives were easier to handle, easier to dispense and had excellent shear strength performance if cured at >150°C, they were chosen ahead of the two part variety. Loctite EA9514 was the selected product.
Experimentation enabled the joint width and screw spacing to be optimised.
In production it was proven that a robotic “X-Y” table could be used to dispense the epoxy adhesive. For prototypes and small volume production, the epoxy could be applied by hand.
Test pieces were manufactured. The base was machined to size 75mm x 95mm from AL6082, a liquid cavity of 59mm x 79 mm was machined, leaving a bonding area of 8mm width. Four, M4 screws for fixturing and mechanical support were added one in each corner.. The lid was a flat 3mm thick plate.
The parts were intended to withstand proof pressures in the region of 9bar.
A Micropac hand pressure pump was used at a test temperature of 19.6°C. No additional support other than the 4 x 4mm aluminium screws was provided during the test. Tension cracking sound was noted at 27bar final pressure was >30bar. In all cases the test pieces exceeded expectations.
Further a Helium leak test with stringent limits was applied, all tested parts achieved leak rates smaller than the 10-6 mbar l/s limits when tested in the outside-in method.
Future developments
The major step forward in this R&D effort relates to the relationship between joint geometry, structural epoxy selection, fixturing and structural support. The benefits of being able to design and manufacture cold plates that are too large for standard manufacturing techniques.
Next development is to conduct pressure cycling of some 200,000 cycles to prove the integrity of the structural epoxy adhesive joint.