Increased DC motor braking

I have a DC motor (nominally 12v and 4A stall) control circuit that has to work within the confines of the electrical circuit shown in the diagram below. I only have access to what is inside the box so that the GND wire must go from power to motor directly, but I can get to it. The 12v supply is interrupted until I connect it.

I have successfully created a half bridge circuit which uses the mosfet between N and L to provide power to the motor using PWM for speed regulation and a mosfet between N and E for braking, again using PWM for regulation – making sure that neither switch is closed at the same time.

Unfortunately the braking force isn’t quite enough, even when “shorted”. I would have liked to have wired it as a full H bridge to put a little reverse (I know when the motor is turning and when its actually stopped) into the motor for a moment to assist.

Unfortunately it seems that the inability to break the GND line stops me from doing this... unless I have gone ‘circuit blind’ and there is a configuration!

If I could generate a negative voltage (doesn’t even have to be -12v) and could pulse that in at N that would probably do it but it’ll need up to 4A.

Any suggestions please as to improving the braking.

Could you put a DPDT relay between motor and control box?
Have you tried shorting between N and E with a piece of wire while the motor is coasting? Non logic level mosfets may have high on state resistance which could reduce braking current.
Have you tried a free wheeling diode between N and E with cathode toward N?

What you are wanting to do is called regenerative braking and may kill your mosfet.

A full H-bridge can apply a reverse voltage. The motor would no longer be connected to GND, but to both half-bridges. The mosfets should be able to have some reverse current for 'normal' shortcut braking, in most cases that is no problem with mosfets.

jcallen:
Could you put a DPDT relay between motor and control box?
Have you tried shorting between N and E with a piece of wire while the motor is coasting? Non logic level mosfets may have high on state resistance which could reduce braking current.
Have you tried a free wheeling diode between N and E with cathode toward N?

What you are wanting to do is called regenerative braking and may kill your mosfet.

No DPDT (or anything) unless it fits somewhere between N, E or L. Those three points are the only things I have access to - the rest of the circuit is a closed system.

Wondered about the effective shorting of the braking circuit. Difficult to manually short it out as braking is all over and done with in under 0.5 second.

I have selected the lowest RDS on mosfets I can... I have 23 different ones on test at the moment plus about 6 or more mosfet and motor controllers. Sure is being a difficult problem to solve.

Yes I have been playing with diodes and haven't given up with this approach yet BUT I think the mosfet is closing properly... however trying to find a way to establish how closed it really is.

Add a resistor of about 0.01 to 0.1 ohm and use a storage scope. That should reveal how much shortcut brake current there is.

Peter_n:
A full H-bridge can apply a reverse voltage. The motor would no longer be connected to GND, but to both half-bridges. The mosfets should be able to have some reverse current for 'normal' shortcut braking, in most cases that is no problem with mosfets.

Love to use an H bridge - can you sketch out what it is you have in mind that would work between N, E and L.

A wild thought just came into my mind (usually foolish as I am not an electronics person and make some pretty howling errors... however).

During the braking phase I short the motor between N and E for 50% duty cycle at 1Hz (numbers chosen arbitrarily).

While this is going on I use the 12v supply voltage to charge up a capacitor.

Then in the next 50% of the duty cycle the NE is connected to the capacitor to provide a reverse polarity at NE.

And so on until the braking phase is over.

Hi,

Possible to get a H-bridge to do the job of the DPDT switch on a capacitor, so it charges on motor run and discharges on brake.

Tom… :slight_smile:

If (or should I say IF) some sort of circuit was possible is the maths something like this:

  • At 500Hz the period is 2mS
  • Half duty cycle takes 1mS
  • 12v supply to 1000uF capacitor through a 1ohm resistor has TC time constant of 1mS
  • 12v stored in a 1000uF capacitor is 0.072J
  • 0.072J released in 1mS is 72W
  • 72W at 12v is 6A

However the drain may not be as much as 6A and so the capacitor will increase in charge rather than decrease and approach 12v...

Is this logic right or??? Does it indicate that there is some merit in pursuing this thinking???

TomGeorge:
Hi,

Possible to get a H-bridge to do the job of the DPDT switch on a capacitor, so it charges on motor run and discharges on brake.

Tom.... :slight_smile:

Like your schematic. Now studying it :slight_smile:

I don't understand. What about just a normal H-bridge ?
That is "design 6" on this page : H-Bridge
A special mosfet driver can be used to drive the gates.
Or a module like these : Pololu - VNH3SP30 Motor Driver Carrier MD01B

Normal braking can be done by making both outputs high or both low. Reverse power for faster braking is no problem for an H-bridge.

Peter_n:
I don't understand. What about just a normal H-bridge ?
That is "design 6" on this page : H-Bridge
A special mosfet driver can be used to drive the gates.
Or a module like these : Pololu - VNH3SP30 Motor Driver Carrier MD01B

Normal braking can be done by making both outputs high or both low. Reverse power for faster braking is no problem for an H-bridge.

I don't think the design you point out will work because the H bridge requires the motor connections to alternate between + and GND depending upon direction. As you can see from my constraint that one of the motor connections is always connected to GND.

I have had to do the motor control myself because the drivers I've tried unfortunately suffer from a hesitancy at during acceleration/braking cycles of about 3-4 times a second which cause the motor to stutter intermittently or they require four wires (like the H bridge design) which I don't have.

One product simply didn't provide any significant braking effect. Its taken me ages to get some individual mosfets to not suffer from hesitancy, failure to switch on fully or switch off or no braking... If I could do the maths that is needed to design it all properly I'm sure it would have been faster and cheaper but I've gotten a stable half bridge now but would really like to get more braking.

Please, if you can see a H bridge design that will fit into the 3 pin environment I have to work with, pop a schematic up I would much rather work with a conventional H bridge as it gives me much more and easier control over the braking side.

Then you would need +12V and -12V.
Is there a way to get to the grounded wire ? Drill or break something open ?

If you really can't break the ground wire, can you get a second 12v supply and connect + to GND and - to your controller? A second battery of about the same size as the original battery? Then you can drive the motor in reverse and get more 'braking' effect.

Your capacitor solution is creating a 'voltage doubler' configured to create a negative power supply. It could work but its efficiency will be low.

It seems a strange restriction when a normal H-bridge would do the job.

acboother:
I have a DC motor (nominally 12v and 4A stall) control circuit that has to work within the confines of the electrical circuit shown in the diagram below. I only have access to what is inside the box so that the GND wire must go from power to motor directly, but I can get to it. The 12v supply is interrupted until I connect it.

I have successfully created a half bridge circuit which uses the mosfet between N and L to provide power to the motor using PWM for speed regulation and a mosfet between N and E for braking, again using PWM for regulation – making sure that neither switch is closed at the same time.

Unfortunately the braking force isn’t quite enough, even when “shorted”. I would have liked to have wired it as a full H bridge to put a little reverse (I know when the motor is turning and when its actually stopped) into the motor for a moment to assist.

Unfortunately it seems that the inability to break the GND line stops me from doing this... unless I have gone ‘circuit blind’ and there is a configuration!

If I could generate a negative voltage (doesn’t even have to be -12v) and could pulse that in at N that would probably do it but it’ll need up to 4A.

Any suggestions please as to improving the braking.

I am sure you are aware the braking by shorting the motor leads is limited by the resistance of the motor windings and brushes. Nothing you can do about that.

Can you introduce a separate 12 volt supply, completely isolated from ground and use that to stop the motor?

Paul

Peter_n:
Then you would need +12V and -12V.
Is there a way to get to the grounded wire ? Drill or break something open ?

No can do.

I'd found something using a 555 that made +- voltages for small devices (op amps) but I need more amps.

MorganS:
If you really can't break the ground wire, can you get a second 12v supply and connect + to GND and - to your controller? A second battery of about the same size as the original battery? Then you can drive the motor in reverse and get more 'braking' effect.

Your capacitor solution is creating a 'voltage doubler' configured to create a negative power supply. It could work but its efficiency will be low.

It seems a strange restriction when a normal H-bridge would do the job.

Its slot car controller. If I put another battery into the system I'd be accused of cheating and I'd have to look for my testicles up my own arse... say no more

Paul_KD7HB:
I am sure you are aware the braking by shorting the motor leads is limited by the resistance of the motor windings and brushes. Nothing you can do about that.

Can you introduce a separate 12 volt supply, completely isolated from ground and use that to stop the motor?

Paul

See my previous reply with respect to introducing another power supply.

All I 'see' are three holes as depicted by N, E and L and I have a three pin plug that goes into them. My world is on the outside of that three pin plug and that is where my mosfets, Arduino, LEDS, pots, triggers etc live.

TomGeorge:
Hi,

Possible to get a H-bridge to do the job of the DPDT switch on a capacitor, so it charges on motor run and discharges on brake.

Tom.... :slight_smile:

I like the concept but its getting a bit mosfet heavy with the H bridge for the capacitor :frowning:

Is there such a thing as a digital DPDT?

The other issue with braking through shorting is that it decreases in efficiency as the speed drops,
the torque falls off with speed, so it can be very effective (dangerously so) at full speed, yet
inadequate at slow speed.

Using current control and an H-bridge allows constant torque braking to be implemented, but
you are limited by the max (reverse) current the supply and bridge can tolerate.