1
\$\begingroup\$

I'm designing an H-bridge and one of the features I'd like to have is to allow the motor to keep it's momentum even when the H-bridge is off.

Obviously flyback diodes are used in H-bridge designs to allow the coils to discharge. I feel like I don't have a very intuitive understanding of this concept, but my hypothesis is that this will cause the motor to stop spinning (or at least slow down a bit until the voltage on one of the motor's terminals is no greater than the power supply +0.7V). Is this correct? What does this mean from a mechanical perspective?

\$\endgroup\$
  • \$\begingroup\$ I think you've got a pretty good handle on the realities of it. \$\endgroup\$ – Daniel Feb 1 '17 at 1:48
  • 1
    \$\begingroup\$ My question would be: how are you getting the motor to turn faster than the supply voltage can spin it? \$\endgroup\$ – Daniel Feb 1 '17 at 1:49
  • \$\begingroup\$ Bridge RPM with no load is rated by V/RPM of motor. Coasting only means back EMF is same V/RPM but any current is reversed if loaded. Otherwise current is only when torque is applied. (neglecting friction and losses) \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Feb 1 '17 at 2:08
2
\$\begingroup\$

The snubber diodes only provide action in dissipating the energy "stored" directly in the windings of the motor. Since the direction of current flow in the motor winding is in the same direction as when being driven, the dissipated energy will in fact continue rotation and not retard it (brake the motor).

If you consider the effect of a snubber diode on a relay for example, adding the diode makes the relay change state slower when power is removed.

You can control the time taken to dissipate the energy by allowing the flyback voltage to increase to some larger controlled level.

Here with a basic snubber: enter image description here

Here with a Zener to increase the snubber dissipative power to reduce the run-on effect: enter image description here

Notice here that the time taken to discharge the energy is much reduced.

The time taken to dissipate this energy becomes important where you want to reverse the voltage across the motor. The time taken becomes the limit of your switch time:

enter image description here

The last element is the ability in an H-bridge to "brake" the motor by turning on either both high side or both low side switches at the same time to dissipate any inertia or momentum stored in your system.

\$\endgroup\$
1
\$\begingroup\$

Consider the image below. These are the five legal states of a H-bridge. There are two illegal states (will cause damage) and four that don't do anything.

All switches OFF - any current would freewheel by the supply & decay

1 & 2 ON - positive current can build up in the load

3 & 4 ON - negative current can build up in the load

1 & 3 ON - zero volt loop that minimises load current decay OR shorts winding.

2 & 4 ON - zero volt loop that minimises load current decay OR shorts windings.

For each state, the current path is shown. THe diodes are present to ensure there is always a path that the load inductors current can take.

There is however a massive difference between the freewheel paths of State1 & State{4,5}

In state1, the "natural" freewheel path, the safe & default path, the main DClink is in-circuit. Thus to facilitate current flow, the load's voltage must be greater than this voltage. This will cause the current to decay relatively quickly and thus there will be some deceleration torque experienced at the shaft

In state{4,5} a "zero volt loop" can be established across the load and thus the free-wheel conserves the the load inductors current. Once the rotor has stopped however these schemes will facilitate a pseudo-locking mechanism

enter image description here

\$\endgroup\$
0
\$\begingroup\$

You have to distinguish between two different effects:

  1. The inductive energy stored in the motor coils, which generate "flyback" voltage pulses when you switch off current
  2. The back EMF of the motor, which (when the bridge is switched off) is simply the EMF of the motor acting as a generator.

The flyback diodes simply return the inductive spikes to the supply instead of generating excessive voltage which could destroy the transistors. The energy involved is normally relatively small and has little effect on a freewheeling motor. (And if it does, the effect can be minimised using the snubber tricks in Jack's answer).

Back EMF - or generator EMF - is dependent on speed, and will generally be less than the incoming supply voltage. Thus, if you simply switch off the bridge, letting the motor freewheel, the generated EMF is not high enough to turn on any of the diodes, and the motor will normally freewheel - with two important exceptions.

  1. If you also remove the supply voltage from the bridge, then the generator will power the circuit through the diodes. (One form of regenerative braking is to use a boost convertor to charge the battery while this is happening). But as you don't want braking, you can avoid this by keeping the bridge powered even while all its transistors are off.
  2. If the motor is being driven by its shaft - e.g. a car running downhill - faster than its top speed, the generated EMF can exceed the supply voltage, turning the diodes on. This will increase the supply voltage and either charge the battery or possibly destroy other electronics on the supply.
\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.