Transient Spike Reduction by John Tetz
 Current Limit with Transient Spike Reduction
Aug 2008 - An article by John Tetz
Previously developed was the Current Limiter Circuit, which limited the current delivered by the RC electronic speed control (ESC) from damaging the lightweight gearing of the low power E-Assist system as well as preventing the system from wasting power. This is very important considering the tiny batteries used in this system. In this article, the system is further refined to limit the initial system turn-on current, which as shown below, will also save power and improve drivetrain reliability.

I had been seeing high start up current/watt spikes on previous data plots.

Here, the motor is being turned on/off with the rear wheel up off the ground. Orange is watts. This plot is at a one measurement per sec record rate, typical of what I had been using for previous data.

The following plot is at a 10 measurements per second rate showing that the startup spikes are actually very high . The first 3 spikes are with the wheel accelerated from a dead start. The next three are with the wheel spinning. Because I am holding the rear end off the ground I can feel the sharp impulse of getting the wheel up to speed. I feel much less of an impulse with the wheel spinning. Those 200 watt and up to 20 amp spikes stresses the systems gear box and battery but does very little to power the vehicle.

I tried changing the values of the capacitors across the op amp feedback resistor. Very little improvement occurred. Then on advice from Richard Flechter I connected a 100 ufd capacitor from the center of the Servo Tester pot to ground. This softens the start impulse I felt when holding the rear wheel of the ground. Note the spikes are no higher than 100 watts and 8 amps and are very uniform regardless if the wheel is accelerated from a stop or spinning.

Then I tried it out on the road, shifting normally. No spikes. Note how flat the power is. The level of the power allowed by the current limiter circuit can be adjusted by setting the voltage on the + terminal of the op amp. A digital voltmeter is needed to measure this voltage.

The following is a previous plot made without the current limiter circuit where I shifted into a much higher gear. Bright Red is the motor RPM. Note how high the power jumped at 22 seconds at 3000 motor RPM and how the power drops as RPM increases. This is typical of most of the road data. This is at a low data logger resolution rate with a higher sample rate the peaks would be much higher.

The following is the plot showing the low speed test I had performed with no current limiter circuit. I snapped off the gearbox shaft at an RPM of 2000 at the 6.5 minute point.

The following plot shows the power vs low to high RPM with current limiting. No big peaks and quite flat power. This broadens the power band without stressing mechanical components or the battery.

 

Here is the Current Limit Circuit with spike suppression.

I am very pleased how the overall system has been improved by adding a tiny op amp. Now, with a current limiter and spike supression, its a closed loop system.

John Tetz

More about John's project and others on the ULEA (Ultra Light Electric Assist) Yahoo Group

Back to the electric bike projects