Thermals Design

Thermals Design

 

 Problem statement


  • Optimize cold plate design

  • Re-finalize radiator and pump part selection

  • Finalize a manifold design

 

 Questions to Answer


  • Optimized pathing/channel for tubing?

  • Is the pump powerful enough to support the pathing chosen above? How do we determine that?

  • Can our radiator dissipate heat from fluid given the chosen fans?

 Solution hypothesis


 

Design Constraints/Given Architecture

Area

Constraint

Reasoning

Area

Constraint

Reasoning

Nose Intakes

Double NACA cutout placements on aeroshell front

CFD by Aerodynamics optimized holes in area of highest pressure (most airflow in)

Nextbreak Shell

Rear Exhausts

Double NACA cutout placements on aeroshell rear

Nextbreak Shell

Ducting

4in. Aluminum ducts

Size Constraints/interferences

Battery Ventilation

Must have exhaust Ducting

Regulations

Cold Plate

  • Battery base plate quarter in. thickness

  • Round Copper Tubing 0.25in OD

>0.25in base plate is egregious, 0.25in Copper tubing OD constrained by cold plate thickness.

Round tubing since no square tubing manufacturer available

Radiator/Pump

  • Location and tube routing

  • Already placed so that length is reduced

  • Only place where pump and radiator fits properly

Calculations and Scripts

Name

Medium

Results

Notes

Name

Medium

Results

Notes

Kaden Battery Steady State Analysis

Matlab

 

image-20251120-030526.png
Q vs. Mass Flow Rate
image-20251120-030612.png
Conduction through plate

At conditions:

  • T_cell = 45; % cell surface temperature [C] (ideal)

    T_plate = 35; % temperature of cold plate [C] (warm to touch, about 95F)

    T_Bair = 35; % ambient temperature inside battery enclosure [C]

Results:

  • MFR needed to for adequate heat transfer: 0.0016 kg/s

  • Delta T between plate and cells needed for conduction: 13.1

  • Q (natural convection): 108.1 W

  • Steady State

    • Does not account for Multiple channels and fluid temp increase further down the channels

    • Static material and fluid temperatures, assumes all heat is transferred through material

    • Assumes constant current draw and heat gen of 600W

  • Needs time step function to simulate true scenario

    • very hard. Can separate tubing into smaller dx? but don’t know how to calculate heat transferred per dx

  • Solution is to do thermal sims

    • idk how

Ava Heat Exchanger Calcs

Python

Heat Exchanger Calcs

 

 

 Cold Plate Optimization

  • Plate thickness and tubing OD is locked in place. So the only cold plate optimization that can happen is tube pathing

  • Note that double channel worked better for rectangular tubing due to larger volume and surface contact

  • Rectangular to Circular

    • 23mm → 20.3mm inner tube cross sectional area (12% decrease in water volume)

    • 16.5mm → 10mm cross section contact length with base plate (40% decrease in contact area)

    • Need to make up in other areas to make up for losses

 

Double Channel

Six Channel

Quad Channel

 

Double Channel

Six Channel

Quad Channel

Overview

Manifold Y-Split the flow into two channels. One per four segments

Manifold three way split into three channels.

Manifold four way split into four channels. One per two segments.

Screenshot

image-20251120-005439.png
Manifold Velocity Distribution

 

 

 

Notes

 

  • I made one in the summer but I lost the file

Pros

  • Lower pressure loss within manifold

  • Middle ground between lower pressure loss and higher heat transfer rate

  • Theoretical higher heat transfer rate across shorter channels

  • Easier pipe manufacturability, shorter channel → easier tolerancing

Cons

  • Lower heat transfer rate (fluid is hotter farther down longer piping)

  • Harder to split between eight segments to ensure uniform cooling

  • Inconsistent pressure loss within manifold

  • Inconsistent flowrate into each channel

Criteria

  • Need to redesign y split to configure with round tubing

  • Need to design a four channel manifold

  • Simming to optimize flow

  • Need to design a four channel manifold

  • Lots of simming to optimize flow

  • Takes a lot more time to design

 

 

 

 

Radiator Optimization

  • Due to space restriction, we are limited to a 2x120mm fan radiator setup with one intake and exhaust duct

  • Radiator must have copper interior to prevent galvanic corrosion between aluminum and copper

  • Better if we order from koolance since they’re an approved vendor

 

34mm Thick, 17-FPI Copper

54mm Thick, 30-FPI Copper

34mm Thick 30-FPI Copper

 

34mm Thick, 17-FPI Copper

54mm Thick, 30-FPI Copper

34mm Thick 30-FPI Copper

Overview

 

 

 

Screenshot

image-20251120-021742.png
image-20251120-022003.png

 

image-20251120-021748.png
image-20251120-021913.png
image-20251120-021657.png
image-20251120-021843.png

Link

https://koolance.com/hx-240yc-radiator-2-fan-120mm-17-fpi-copper

https://koolance.com/hx-240xc-radiator-2-fan-120mm-30-fpi-copper

https://koolance.com/radiator-2-fan-120mm-30-fpi-copper

Pros

 

 

 

Cons

 

 

 

Criteria

  • We have ordered this

  • We have this at pickle

  • Would need to order

 

 

 

 

Pump Optimization

  • Must have extensive documentation especially over PWM control

  • Can be 12V now since pump controller has a 12v buck Rev B (a lot more options)

  • Current pump was selected since it was the only 24V pump with proper documentation.

 

Kaden’s Notes:

  • Plate size and tubing size is locked into place → only things we can change is pathing, which in turn dictates manifold design. We shouldn’t do less than one “loop” per two segments due to contact area already shrinking due to round tubing.

  • Need to decide pathing and manifold design

  • Other points to figure out: pump strength and radiator op

 Follow up

Decision

Status

Next steps

Decision

Status

Next steps

decided / in review / other

 

 

 

 

 Source files

Type /link to add links to design files.