RNM Entropy

Solutions and ideas under Creative Commons 4.0, Noncommercial, Share Alike

Oil cooling of electric vehicles parts with nanofluids

Posted by RNMentropy on 26 de February de 2014

I found an article online about adding nanoparticles of  hexagonal boron nitride to transformer oil and it got me thinking:

  • What if we could apply this mineral oil + nanoparticles of boron nitride to help cool electric motors, power electronics (like motor controller, battery charger, dc-dc converter) or even the battery pack (cells or BMS)?

Starting with the following scenario:

  • Internally oil cooled electric motor (oil flowing through to stator and rotor via a pump and heat exchanger)
  • No magnetic limitation (iron saturation, permanent magnet limitation, etc.)
  • 95% efficient motor
  • 100 kW continuous power output

The continuous power output of an electric motor (or any machine or system) follows a very simple principle:

Pout = Pin – Plosses

Since it’s 95% efficient that equation would look like: 100 = 105.26 – 5.26.         What this mean is that the power output is effectively limited by the cooling system, which is only capable of transferring 5.26 kW of heat.

Enter the oil + nanoparticles, and let’s say we could achieve that 70% increase in thermal conductivity. So we could increase the motor output to 170 kW? Well, “the plot thickens”…

Forced oil cooling is, in fact, forced convection heat transfer, which normally corresponds to  , where h = convective heat transfer coefficient of the fluid, A= area of heat transfer, Ts = temperature of the heat source and T∞ = temperature of cooling media (convection includes both advection and diffusion). 

BUT this is a simplification of the equation ,

where k = thermal conductivity of the fluid and eq represents the thermal boundary layer thickness (“y” in the below picture).

Therefore, hBN shows great potential to dramatically increase convective heat transfer. Since 1 test is worth a 1000 calculations and opinions, let’s see if someone tries this in an electric motor or power electronics…

From the top off my head, Mission Motors or Rimac could benefit a lot from this kind of technology, further improving their already über powerful motors and controllers.

Mission Motors
85 kW peak, 15000 rpm, 23 kg PMAC version

Rimac’s Concept One rear motors.
386 kW continuous, 654 kW peak, 115kg for 2 motors!

Any drawbacks in using hBN? Don’t think so because:

  • Adding only 0.1% wt of hBN didn’t affect the oil viscosity, so no need to change pumps, heat exchangers, piping, etc.
  • Hex boron nitride isn’t very expensive, although it’s probably necessary to mix the 70 nm powder ultrasonically in isopropyl alcohol, dry it and mix ultrasonically that with the intended oil.
  • High Temperature Stability, 1000o C in air.
  • Excellent Lubricating Properties due to low Coefficient of Friction at 0.15 to 0.70.
  • Good Chemical Inertness.
  • Electrical Insulator.
  • Environmental-friendly (from the found online data…)
  • No damage to copper windings or enamel, think as adding soft Teflon particles. BUT, if we add cBN (Cubic Boron nitride), it’s like adding diamond dust, very abrasive.

Apparently, we could further improve this by replacing hBN, in terms of heat transfer up to 260%, with BNNTs (Boron Nitride Nano Tubes) BUT, it seems to be cytotoxic, even at low concentrations (think as being as safe as asbestos). “The perfect is the enemy of the good”… 😉

“Enough of theory!” some would say, others “I need hard data!”. Here you go mates, tried and tested (!):

  • Alumina (Al2O3)

Heat transfer through heat exchanger using Al2O3 nanofluid at different concentrations“. Ignore the abstract’s “convective heat transfer coefficient of nanofluid is slightly higher than that of the base liquid”. In the conclusions we can read “For example at the particle volume concentration of 2% the overall heat transfer coefficient is 700.242 W/m2 K and for the water it is 399.15 W/m2 K for a mass flow rate of 0.0125 L/s so the enhancement ratio of the overall heat transfer coefficient is 1.754″.

That’s a 75.4% increase in heat transfer. But I still would go for the hBN because 2% alumina seems to increase friction losses and pressure drop.

  • Copper nanoparticles

Convective heat transfer and flow characteristics of Cu-water nanofluid

“Compared with the base fluid, for example, the convective heat transfer coefficient is increased about 60% for the nanofluid with 2.0 vol% Cu nanoparticles at the same Reynolds number.”

Shared under Creative Commons 4.0, Noncommercial, Share Alike










Posted in Electric motors, Electric vehicles, Motor controllers/ inverters | 3 Comments »