Inductive kickback Voltage

[venkatgandham]

I am trying to calculate the inductive kickback voltage for the motor. I am having a problem matching theoretical calculations with the measured inductive kickback voltage. Can anybody please help me with identifying the issues with my calculations?

Continuous current = 25 A,
Max voltage = 16.
Peak current= 40 A.

VIk = Li di/dt. 90 uh * 25 A / 70 us = 32 volts.

Measured values on the oscilloscope are between 23 to 28 volts depending on the load. At maximum load I can get the inductive kick back up to 28 volts.

I want to calculate the dt without using the oscilloscope. (i.e the discharge time of the inductor)

Should the dt be calculating by tou = L/R ~ dt or with some other parameters needing to be considered for calculating the discharge time?

 

[Tunelabguy]

Since you are dealing with a motor and not a simple inductor, the motion of the motor is going to figure into the inductive kickback. You would have to know a lot more details about the motor to model it precisely - things like how fast the motor was turning at the time you switched off the power, etc. Why do you need to get theory to match the measurement? It is not going to be a simple theory - not like switching off a simple inductor.

 

[venkatgandham]

The reason for my theory to match the measurement is that I am choosing a gate driver which should be able to handle the inductive kickback voltage. The present gate driver on the board is getting damaged by the inductive kick back voltage.

At 2000 RPM at 90% of load allowed, will generate a 26 volts of inductive kickback.
At 3000 RPM at 70% of load allowed, will generate a 28 volts of inductive kickback, which is damaging the phase pins of the gate driver. I can protect it by having a TVS diode but I want to calculate the inductive kickback voltage theoretically to match my measurement at different speed and load.

Do I need to consider the speed and load factor into equation to get the exact values. Also, I am thinking of calculating the worst case scenario, this can also work. but I am stuck with dt value. Could you please help me in calculating the dt value for the inductance or it needs to have more parameters like speed and load into equation to calculate it.

 

[BradtheRad]

Does the commutator spark? Then it indicates higher voltage than 26 or 28 V.

Anyway ignoring that, you are measuring single windings that are switched on and off, one after another, each for a time that is inversely proportional to the rpm.

Each winding builds up a Weber value ( A x L ) while it is On. The LR time constant determines how rapidly current builds.

When the winding is switched out, it generates a voltage. How high it is depends on the resistance the winding sees as it discharges that current.

 

[Tunelabguy]

The reason for my theory to match the measurement is that I am choosing a gate driver which should be able to handle the inductive kickback voltage. The present gate driver on the board is getting damaged by the inductive kick back voltage.

Do you have the option to modify the circuit, or is your only option the selection of a gate driver component? I would say, if it is possible, design the driver with a kickback voltage limiter, setting the limiting voltage just under the safe maximum voltage the switch can handle before breaking down. That way, if the kickback voltage does try to rise too high, instead of a burned out switch, the only negative consequence will be a lack of peak performance of the motor (torque and speed). If the performance is not sufficient, you will need to select a higher kickback voltage limit and a switch that can handle that higher limit.

Depending on the circuit, a kickback voltage limiter might be as simple as a Zener diode in series with a regular diode connected between the switch and the positive rail.

 

[schmitt trigger]

Are you measuring the kickback with the motor fully loaded or unloaded? And its speed?
On a motor, its electrical characteristics are fully dependent on its mechanical output.

 

[venkatgandham]

Yes, I can use the TVS diode to protect the gate driver from the inductive kick back voltage instead of going to a new gate driver. I do see the relationship between speed, current, load and inductive kick back voltage.

When the motor is loaded from 80 to 100 % of allowed load at max speed and the current draw is max, that generates a high inductive kickback voltage. That puts the gate driver in danger. I was able to save the gate driver by placing a TVS diode in between the phase pins and the ground.

Unfortunately, my software team was able to fail the same board during software development.

It failed because the software is still in a development stage and they are tuning the limits of current through the motor. When the motor is asked to spin at a high load and high RPM, it will draw more current but it will compensate the current draw by reducing the speed.

My question is, in which scenario can I fail the board due to software?

For example by subjecting the board to high current draw, which will result into too high of an inductive kickback voltage.


I have a few more questions related to the inductive kick back voltage.

What if I don't provide the discharge path for the inductive kickback voltage? Will it increase indefinitely or will it dissipate into the gate driver as it has no path to discharge?


I see the relationship between max load and inductive kick back. That means max the load equals the max current draw.

In that case:

Vik = Li di/dt and E = 1/2 (I)2 L = is the energy stored in the inductor. How would I calculate the exact inductive kickback voltage based on this equation?

What factors should I consider for it?

 

[BradtheRad]

What if I don't provide the discharge path for the inductive kickback voltage? Will it increase indefinitely or will it dissipate into the gate driver as it has no path to discharge?

As a general rule with inductors, EMF will soar instantly to whatever level is needed, in order to keep current flowing at switch-off.

Vik = Li di/dt and E = 1/2 (I)2 L = is the energy stored in the inductor. How would I calculate the exact inductive kickback voltage based on this equation?

What factors should I consider for it?

Suppose a winding sees 1000 ohms, and current is 2A. Then you get 2,000 V. (Again, a general rule for inductors. My knowledge of motor dynamics is incomplete.)

Notice you cannot just put a diode across each winding (as we frequently are advised to do with a coil). A diode defeats the motor's ability to generate torque.

I could be wrong.