One of my first engineering-related internships was regarding gearing mechanisms in BLDC electric motor powertrains.
The Problem
Nylon gears are often incorporated into low max-load powertrains in order to reduce noise. This often causes reliability issues since suppliers do not know what limits the motor will be pushed to; such as the application of 70km/h electric bike. My job was to propose several cost-effective material solutions to this issue along with full quotes from suppliers, and testing data.
The main plan was first to quantitatively identify the limits of the gear design and then apply the different mechanical properties and characteristics (such as ductility, young modulus, strength, toughness, modulus of rupture, etc.) of candidate materials to analyze their viability for this application in order to then gather information to quote.
One of the ridiculously powerful dual-suspension pedal-assist mountain bike motors I had the honour of working on.
For the application of helical gears, I found the use of “Hertzian Stress” Methods was most applicable for FEM Analysis to simulate a close estimate of each materials limits in terms of cycle lifetime and maximum stress (sustanied and non-sustained) while also taking into account the effects of temperature(thermal softening), specific material characteristics such as crack tip intensity, and several production methods and/or treatments on its behaviour. By localizing contact points to a squashed rectangle (like a tire’s contact area with the road), the Hertzian method works quite well to determine the maximum and mean stress the gear will face as shown below. By 3D modelling the system of gears I can mesh them in an assembly and find the curvatures needed to set up equations for the maximum Hertzian stress the gear can deal with.
From there,
I can use this localized stress value in an FEM simulation in order to find the point of failure for each material solution. With this data and info from ASTM D671 (American Society for Testing and Materials) I can work backwards by using AGMA’s (American Gear Manufacturers Association) Standard 2105-D04 rating formulas to find the number of load cycles expected by fatigue fracture corresponding with stress cycle factors of the gear teeth to find the gear’s eventual fatigue failure due to the cyclic nature of the load applied at different rates of use. Then I can use standardized values to determine a reccomendation and conservative probablilities of failure for best judgement.
A sample comparison of Nylon vs Steel at near-maximum loading of the Nylon gear.
In the end, some of the most promising solutions turned out to be:
Steel Core Fiberglass-reinforced Nylon 66:
Maintained noise reduction
Self-lubricating
38% more torque capacity
Moderate cost
Easily machineable
low dimensional stability and weaker than other steel gears
66 Crystal structure designed to bear load
Low friction and moderate fatigue resistance
Duratron® PBI:
One of the strongest thermoplastics (2 times nylon)
maintained properties above 400°C (extreme dimensional stability)
maintained noise reduction
Easily Machineable
Expensive
still weaker than other steel gears
Low Friction, high fatigue resistance
S45C Heat treated Steel
cheap
as strong as all other components
easy / standardized manufacturing
Noisy
Overall, my experience in learning proper gear analysis help me in more affective analysis in design decisions of my own projects and I look forward to digging dfurther into this stuff,
Thanks for reading. Brian.
The steel-cored fiberglass reinforced nylon, Duratron PBI, and heat-treated steel solutions respectively