Freewheeling or Generation(Motor)

Yes, depending on the application.

Yes. It's quite useful for braking, if that's what you want.

"Waste". You're so judgemental.

Keeping the H-bridge low impedance (i.e. always having two transistors on, except as necessary to avoid shoot-through) will make the control to your motor more positive and predictable -- your motor control will have a known average voltage, which means that on average it'll act like a linear system and make wrapping it with something like a PID controller easier.

But it'll make the motor burn up more power in it's windings ("waste" is such a strong word...).

Allowing the motor terminals to float as the motor freewheels will, indeed, prevent that energy from going away as heat. But as you change the duty cycle of your bridge the motor's effective mechanical damping will change. If this is of no concern for you, then you can probably go ahead and let them float.

Are you supplying DC or AC power to your H bridge? Is your controller designed for single quadrant or four quadrant operation. Does your motor only rotate in one direction or is it a bidirectional motor? Do you want your controller to act as brake when stopping your motor?

Maybe John is thinking that the manufacturer will soon be slapped with a carbon tax if he includes freewheel braking in the design ;-)

Like the pirate-copy tax the Europeans supposedly slap on things if there is a CD burning capability included.

They did it with DAT here in the U.S. as well:

Absurd, but then again... maybe an "we assume all of you are pirates" tax is preferable to the heavy-duty DRM seen on Blu-Ray discs, Winhdows Vista, etc.?

I just resent it when a government body makes a blanket assumption that the majority of users is engaging in criminal behavior. It is not right. I've never pirated any music, software, movies or whatever and I do not plan to. Heck, call me a luddite but I wouldn't even know how to download music. But I also do not want to pay a tax for something I do not commit.

Certainly they tax blank DVDs in France (and AFAIK most EU countries) which makes them about 100 Euros for a spindle or about 10 times more expensive than in the UK. For that reason, there are several online shops in the UK that sell blank DVDs and their webpages are, for some very strange reason, in French. Most French consumers that I have talked to don't seem to realise that >90% of the price of blank DVDs in their country is tax that AFAIK goes to some pretty arbitrary companies, otherwise I doubt they would tolerate it. I think they also tax USB hard drives now, and they re-adjust the tax per gigabyte every so often, so that it stays just below the threshold where consumers would start to ask where all that money is going. And then even after you've paid the private tax, they still say it's illegal to sell copied movies etc. There is some weak excuse like that in exchange for the tax you are allowed to make copies of CDs for "close family" or something like that but that is very discriminatory against people who either make their own music and movies, or just feel like recording some data for whatever reason that is nobody else's business. They should not get to have it both ways, (either stop the private taxation or declare any and all piracy legal when using highly taxed media) but I guess these companies must have very good lobbyists.

Chris

normally, Hbridges have both High and Low side off during the low cycle. And in your case, you need to have the other low side on to keep the boost cap in check in the low cycle.

You can turn on both low sides to do regenerative braking, in this case it wouldn't matter if the cap got charge or not because you're not requested a drive current how ever, in both cases the low side fets should come on with a low enough Ron to provide a low side path for the boost cap. If this isn't the case then it's possible the motor you're using is generating more REGEN energy than the low side fets can short out. But like I said, you should always have the driving FETS off during the low cycle to not brake the motor and only the opposite low side to keep the boost happy for the next on cycle..

During a slow down braking, you don't need the high side any ways so there really isn't a problem here that I can see and you'll only miss one on cycle on restart of driving state.

You can short the motor, freewheel it, regenerate, or divert the current via a load resistor. The options have different advantages and disadvantages. If it's freewheeling, it's unbraked, so it rotates until the friction uses up the inertia. If it regenerates the energy can be recovered, useful in battery applications, however keeping the voltage under control can be a problem and the circuitry is more complex. Shorting the motor brakes it, but the inertial energy is converted to heat in the windings, so you have to be careful about the motor rating and cooling arrangements. The load resistor dissipates the heat elsewhere, and you can achieve different levels of braking if you need to, but again the circuit is more complex. And don't forget about the possibility of an overhauling load.

JS

You have some good answers. It seems some facts were missed though, especially if you're driving a big motor.

If you short your motor, you can get almost as much current as when you short your power supply with a DC resistance the same size as the load's. The L/R tc means it takes time to build up, but if there are enough joules freewheeling it will still hurt. If you have enough resistance in the loop to stop it hurting, that resistance dissipates the braking energy.

If you don't have enough resistance in the loop, you have to PWM the short-circuit based on the L/R time constant, alternately routing the developed current back into the battery, which stores the energy instead of dissipating it as heat. *That" is how regenerative braking is done.

If you don't want to regenerate and can't or don't want to short the motor (i.e. you want true freewheeling) you may need to PWM until the motor current reaches a low enough value that you can safely open the circuit without the inductive spike causing pain.

I.e. a short may not be feasible but instantly opening the circuit can cause pain also.

Clifford Heath.