NREL/Oak Ridge thermal measurements of packed copper wire enables better electric motor designs
11 July 2017
A recently completed study by researchers from the National Renewable Energy Laboratory (NREL) and Oak Ridge National Laboratory (ORNL) of the anisotropic—i.e., directionally dependent—thermal conductivity of packed copper wire for electric-drive vehicle (EDV) motor applications is providing a baseline for the assessment of new materials and winding structures. The findings of the study are published in the ASME Journal of Thermal Science and Engineering Applications.
Improved thermal management of windings in electric motors makes it possible to maximize operational efficiency and longevity and reduce component footprint, allowing manufacturers to meet consumer demands for high-performance, reliable, and long-lasting EDVs.
Anisotropy is the quality of exhibiting properties with different values when measured along axes in different directions. As a relevant example, in 2015, a team led by a group of researchers at Berkeley Lab experimentally confirmed strong in-plane anisotropy in thermal conductivity, up to a factor of two, along the zigzag and armchair directions of single-crystal black phosphorous nanoribbons. (Earlier post.)
When the NREL/ORNL researchers evaluated packed copper wire windings used in vehicle applications, they found anisotropic properties, with distinctive differences between their parallel and perpendicular measurements. The thermal conductivity proved to be over two orders of magnitude higher in the direction parallel to the wires than in the perpendicular direction, for a wire packing efficiency of approximately 50%.
Researchers examined 670- and 925-μm-diameter varnish-impregnated copper wire specimens with an insulation coating thickness of 37 μm. The interstices were filled with a conventional varnish material and also contained some remnant porosity. The apparent thermal conductivity perpendicular to the wire axis was about 0.5–1 W/mK, whereas it was more than 200 W/mK in the parallel direction.
The thermal conductivity of the wire was measured using laser flash, transient plane source, and transmittance characterization methods both parallel and perpendicular to the axis.
A measurement of apparent thermal conductivity (κ_app) was used to factor in not just the bulk thermal conductivity (κ) but also the interfacial thermal resistances, which can lower the apparent thermal conductivity.
The collective results from all three test methods indicated that the κ_app of the packed copper wire was significantly higher in the direction parallel to the wires than in the perpendicular direction.
The low κ_app values in the perpendicular direction indicated the copper wires were isolated and did not significantly affect heat conduction in this direction. On the other hand, heat conduction parallel to the copper wires showed the wire had a significant impact.
This supports the expectation that increasing the thermal conductivity of both the wire-insulating coating material and the material in the interstices can significantly increase the apparent thermal conductivity perpendicular to the wire orientation in packs of aligned copper wire.
These test results provide a valuable baseline for comparing new materials, and for highlighting methods for examining the thermal impact of new materials for winding structures relevant to motor applications. This ongoing study will help manufacturers and researchers design high-performance, long-lasting motors for EDVs.
Resources
Wereszczak AA, Emily Cousineau JJ, Bennion K, et al. (2017) “Anisotropic Thermal Response of Packed Copper Wire.” ASME. J. Thermal Sci. Eng. Appl. 9(4):041006-041006-9. doi: 10.1115/1.4035972
There have been recent stories about the use of square wire in motors, specifically to increase the packing fraction, reduce electrical resistance, shrink the physical size of the windings and reduce the magnetic flux path length.
One of the consequences of square wire is that the area of contact between wires will be much greater. It would be interesting to know if this has a corresponding effect on the cross-conductor thermal conductivity, or if the interface effects between conductor and insulation dominate.
Posted by: Engineer-Poet | 11 July 2017 at 01:10 PM
E.P. my edification suggests that the dominant current carrier is the skin or surface effect.
There being little real difference between a copper tube or solid.
That suggests that the Cu weight or density is irrelevant. (Cu being non magnetic)
Sq wires are supposed to have demonstrable advantages
by virtue of the sharp 90o sharp penetrating contact to for E.G. electronics wire wound binding posts.
They are also touted as non slip binding applications beneficial for stringed musical instruments when the core is sq.
That high current hollow or surface carrier tube windings have been shown to be as efficient (majority charge) carriers compared to solid wire may help inform your investigation.
Posted by: Arnold | 12 July 2017 at 01:30 AM
If we visualise the conventional circular rotation of EM as textbook description, then I see a correlation with conservation of angular momentum which suggests 90o angles are NOT conserving angular momentum but rather 90o of deflection resembling the brick wall the - T intersection.
Obviously gravimetric density increases with packaged density but that is not necessarily ,consistent with the conventional understanding of electromagnetic field's curved character.
From my 'old school' perspective, without prejudice to real world progress, is that counterintuitive results are described.
Posted by: Arnold | 12 July 2017 at 02:55 AM
I've been through the "old school" EM curriculum, and what you just wrote is incomprehensible.
Skin effect is an issue with very large conductors and high frequencies. RF applications often used braided "Litz wire" to distribute current through the bulk of the conductor. This is generally not an issue in the small wire used for electric motors up to a few tens of horsepower and frequencies of 50-60 Hz; the "skin depth" is much greater than the conductor size.
Posted by: Engineer-Poet | 12 July 2017 at 07:12 AM
It's probably a scam because i read the article 2 time and the 4 blog posts and i still understand nothing, so this is a scam.
Posted by: gorr | 12 July 2017 at 08:46 AM
No, gorr. It's you, not them. You aren't tall enough for this ride.
Posted by: Engineer-Poet | 12 July 2017 at 09:32 AM
LMAO!!!! Thanks for the laugh E-P.
gorr,
This is Oak Ridge...a National Lab. These are the guys who did the uranium enrichment for the Manhattan project and the design of many different nuclear reactors and all kinds of advancements in materials science, propulsion for space exploration....you know, real science.
Again, adjust your meds, get a grip and stop with the delusions and conspiracy theories. It's getting really tiresome.
Posted by: DaveD | 12 July 2017 at 01:08 PM
Just image if we can find a material that's much lighter and even more conductive than copper- we usually are just waiting for better batteries but there are other innovations that can advance EVs- I like the advances in power electronics and I'm watching developments in carbon nanotube yarns, covetic alloys, superconducting composite tapes, graphene coatings- I'm interested in the benefits of hub motors but to make them practical we probably need a substantial increase in motor power density so lighter more conductive wire would be a godsend.
Posted by: William Stockwell | 12 July 2017 at 02:06 PM
No kidding. Lighter, more conductive wire (looking at you, Richard Smalley nanotube guru) would change so many things it's hard to come up with even a short list.
Especially if it's made from carbon. All the talk about materials shortages—POOF!
Posted by: Engineer-Poet | 12 July 2017 at 03:59 PM
Improved lighter e-motors and control electronics will benefit HEVs, PHEVs, BEVs and FCEVs.
It would be an effective way to increase distance on electricity, reduce fuel consumption and GHG etc?
Posted by: HarveyD | 12 July 2017 at 04:44 PM
Engineer-poet not sure if you were joking but Richard Smalley died in 2005 - so probably no big innovation coming from that direction.
Posted by: William Stockwell | 12 July 2017 at 08:17 PM
Not a joke so much as a tip of the hat, and a dog-whistle to see who remembers his name.
He contributes no more but the battle he began continues to gain ground.
Posted by: Engineer-Poet | 12 July 2017 at 09:50 PM
Constructive criticism always welcome.
I was of course thinking of benefits for large rotating machine winding but these smaller compact examples are benefiting from improved thermal properties.
The parallel thermal conductance fits with higher Cu density and the other reference - to perpendicular thermal isolation - hints at investigation of higher thermal conducting insulation.
Brings to mind
http://www.greencarcongress.com/2017/01/20170110-honda.html
Both motors utilize segment conductor winding, where the coil is inserted from the axial direction of the stator core, as well as square copper wire versus the previous round wire, providing a denser and more compact stator. These changes enable the electric motors to be significantly smaller and more than 23% lighter than the previous motors.
Posted by: Arnold | 12 July 2017 at 11:58 PM
I don't recall seeing that Honda piece before. Thanks for the heads-up.
IIUC, large machines such as multi-megawatt alternators have long used, not just square wire, but hollow wire with the central passage used for coolant. Hydrogen appears to be the most commonly-used coolant, which is why you occasionally read about hydrogen explosions at power plants.
Posted by: Engineer-Poet | 13 July 2017 at 05:58 AM