Here's a relay drive circuit.

What's next is the PWM and input circuit, then the power supply.

Do you want to add in an emergency stop option? Did the original board support this? Normally, the E-stop switch is closed, but when it goes open the relays are both switched off and the motor will stop. In fact, the motor should actually brake, as the terminals are shorted. This is a transistor controlled E-stop, so it doesn't require a high current switch, or additional relays. Also, as it controls the relays, it should still work in case of MOSFET or diode failure.

 

There was circuitry and a physical brake device attached to one end of the motor. If my thumb slipped off the accelerator pot the scooter would come to a screeching halt. It was quite troublesome. While it was a safety feature I actually removed the braking device and bypassed the switching. With it bypassed, the scooter still comes to a quick stop, but not nearly as bad as with the brake. The only issue is that when on a hill the scooter moves down the hill very slowly. The slow movement was due to the differential gearing. I really don't mind that. The worst scenario is if the power is off, then the scooter is free wheeling and can pick up an unsafe speed.

I am guessing that the E-stop configuration might be just a low resistance(even short) across the motor leads. This would be an electrical solution, and not the parts solution the scooter originally came with. I have reasons to want it as well as not want it. A switchable option would be nice, but if not possible I would opt to exclude this feature.

 

There was circuitry and a physical brake device attached to one end of the motor. If my thumb slipped off the accelerator pot the scooter would come to a screeching halt. It was quite troublesome. While it was a safety feature I actually removed the braking device and bypassed the switching. With it bypassed, the scooter still comes to a quick stop, but not nearly as bad as with the brake. The only issue is that when on a hill the scooter moves down the hill very slowly. The slow movement was due to the differential gearing. I really don't mind that. The worst scenario is if the power is off, then the scooter is free wheeling and can pick up an unsafe speed.

When there is no power, both relays will be in the NO position. This would short both motor pins to +24V and brake the motor.

I am guessing that the E-stop configuration might be just a low resistance(even short) across the motor leads. This would be an electrical solution, and not the parts solution the scooter originally came with. I have reasons to want it as well as not want it. A switchable option would be nice, but if not possible I would opt to exclude this feature.

Well the E-stop would be a button you press if you want to stop the scooter, not something that would activate when the accelerator pot snaps back to zero. (This puts both relays in the NO position - same as no power.)

 

When there is no power, both relays will be in the NO position. This would short both motor pins to +24V and brake the motor.

Well the E-stop would be a button you press if you want to stop the scooter, not something that would activate when the accelerator pot snaps back to zero. (This puts both relays in the NO position - same as no power.)

Then the E-Stop button can be a key'd switch for powering on the scooter as the scooter once had... Yes? This would be the most ideal. In a sense, the Power button( the key's switch) and the E-stop button can be one and the same. Seems like there would be no reason to have both.

 

Then the E-Stop button can be a key'd switch for powering on the scooter as the scooter once had... Yes? This would be the most ideal. In a sense, the Power button( the key's switch) and the E-stop button can be one and the same. Seems like there would be no reason to have both.

I don't know how the power switch works, but would it be easier for you just to use the key and not have the E-stop? I will leave the spaces for the components but you don't have to include them.

 

Are you driving one or two drive motors? If you are using a dual drive then there might be a problem with only one motor pulling all of the load. How much current(amps) is flowing to the drive motor(s)? Are you trying to build another motor controller to replace the bad one? Also, the e-stop is there in case the motor controller goes crazy and you need to stop quick.

 

Are you driving one or two drive motors? If you are using a dual drive then there might be a problem with only one motor pulling all of the load. How much current(amps) is flowing to the drive motor(s)? Are you trying to build another motor controller to replace the bad one? Also, the e-stop is there in case the motor controller goes crazy and you need to stop quick.

Just one motor is being used.
Yes, we are building a new motor controller to replace the bad one.
Estimated current is between 15A - 20A. An e-stop or key switch are both viable ways to cut power to the controller and motor and thus fulfill the same task.

@TOM66, Yes that will me fine as I think the key switch for power is good for
both situations. Leave the spaces for the components. Thanks!

 

Ok, I was just wondering, I work on industrial electric forklifts and the theory is very similar. An H-Bridge works well in this type of situation, but you probably know that already. Have a great week!

 

Ok, I was just wondering, I work on industrial electric forklifts and the theory is very similar. An H-Bridge works well in this type of situation, but you probably know that already. Have a great week!

What's better about an h-bridge design compared to the current design we have been discussing?

 

What's better about an h-bridge design compared to the current design we have been discussing?

The major difference with a H-bridge is that it is quicker to reverse and you can drive it in full speed reverse, but those shouldn't be issues for you.

 

Sounds good.

 

Just thinking. Now I realise it might not be best to short the motor terminals while it is running. Are you okay with it just switching off PWM for emergency stop, the motor wouldn't actually brake (It would just free-wheel.) Otherwise an array of power resistors will be necessary to convert the kinetic energy to heat.

 

Just thinking. Now I realise it might not be best to short the motor terminals while it is running. Are you okay with it just switching off PWM for emergency stop, the motor wouldn't actually brake (It would just free-wheel.) Otherwise an array of power resistors will be necessary to convert the kinetic energy to heat.

That's sort of what I was getting at earlier. Shorting the Motor terminals will pretty much lock the motor up and put the scooter into a skid, so to speak.

As mentioned previously, with the existing Motor controller and the brake removed, turning the power off from a moving perspective would first jolt the scooter to a stop and then free-wheel. In other words right now if the scooter was ON but at a stand-still and I turned the power off, the scooter would begin to move if on a hill. Thus Shorting the motor terminals for the power off switch would be very beneficial. With this same setup if I let the accelerator pot spring back to center position(2.5k on the out) the scooter would slow down substantially. This pot position is neither forward or reverse and I assume would be no PWM. This is actually what the E-switch could do.
The trick is that when power is on, there must be some resistance across the motor terminals, not infinite(free-wheeling) and not shorted(frozen). In other words the E-switch must stop the PWM but keep the circuit resistance on the motor. The power switch energized a relay that connects the circuit to the motor.

I hope that makes sense. Thus I think that what you are suggestion is what we want.

It would be interesting to know what sort or resistance would provide a manageable amount of braking, and at what sort of power rating. My friend has an assortment of power resistors I could play with that. The scooter can run at about 4.5mph at 100% duty cycle.

 

That's sort of what I was getting at earlier. Shorting the Motor terminals will pretty much lock the motor up and put the scooter into a skid, so to speak.

As mentioned previously, with the existing Motor controller and the brake removed, turning the power off from a moving perspective would first jolt the scooter to a stop and then free-wheel. In other words right now if the scooter was ON but at a stand-still and I turned the power off, the scooter would begin to move if on a hill. Thus Shorting the motor terminals for the power off switch would be very beneficial. With this same setup if I let the accelerator pot spring back to center position(2.5k on the out) the scooter would slow down substantially. This pot position is neither forward or reverse and I assume would be no PWM. This is actually what the E-switch could do.
The trick is that when power is on, there must be some resistance across the motor terminals, not infinite(free-wheeling) and not shorted(frozen). In other words the E-switch must stop the PWM but keep the circuit resistance on the motor. The power switch energized a relay that connects the circuit to the motor.

I hope that makes sense. Thus I think that what you are suggestion is what we want.

It would be interesting to know what sort or resistance would provide a manageable amount of braking, and at what sort of power rating. My friend has an assortment of power resistors I could play with that. The scooter can run at about 4.5mph at 100% duty cycle.

Didn't you keep the old brake array? You could probably mod it to do what you want. Remove some of the resistors.

If not, an array of ten 10W resistors preferably with a fan (a small 12V dc one, only running when the array is active) could probably provide braking which would act in about 10 seconds.

Does the motor have a hp (horsepower) or W/kW rating?

Also you can consider regenerative braking - the motor's kinetic energy being converted into electricity to recharge the battery - but it could be tricky to do right. What battery type is used, I'm assuming lead acid?

 

Didn't you keep the old brake array? You could probably mod it to do what you want. Remove some of the resistors.

If not, an array of ten 10W resistors preferably with a fan (a small 12V dc one, only running when the array is active) could probably provide braking which would act in about 10 seconds.

Ten 10W, but at what resistance, any idea?
I was thinking If we could create a armature braking system where an increase in the movement continued to to lower the resistance. Would be neat, but not sure how to accomplish it.

The "Brake" I removed was not a resistive mechanism but rather sort of like a disc brake. It was a plastic disc that slipped over the open end of the motor shaft which was coupled with a 5 sided piece of metal. When the accelerator pot returned to neutral a switch was made that pressed the rotating disc against a housing. I was having a bit of trouble with it. I believe I do still have it but it was a pain in the A as it was a very jolting stop if my thumb slipped off the accelerator pot.

Does the motor have a hp (horsepower) or W/kW rating?

The only specs on the motor are 3,500 RPM 15A

Also you can consider regenerative braking - the motor's kinetic energy being converted into electricity to recharge the battery - but it could be tricky to do right. What battery type is used, I'm assuming lead acid?

Although a nice thought, I often wondered what real energy gain I would obtain from such a scheme.

The differential gearing and power-on combination provides ample braking for me. The only feature I would find helpful would be to lock the scooter wheels via the shorting method for when the scooter is on a hill.

 

Ten 10W, but at what resistance, any idea?
I was thinking If we could create a armature braking system where an increase in the movement continued to to lower the resistance. Would be neat, but not sure how to accomplish it.

Well with 24V, to dissipate 100W (peak) the resistor must be 5.76 ohms. Call it 6.8 ohms (closest value) which would dissipate 85W. Make each resistor 68 ohms as ten in parallel is 6.8 ohms. Or use 20 x 120 ohm resistors which would lower the load on the resistors a bit.

The resistors must be kept cool for reliable operation so a fan will be essential unless you can devise some other method of cooling them?

The only specs on the motor are 3,500 RPM 15A

Experimentation will be necessary then.

Although a nice thought, I often wondered what real energy gain I would obtain from such a scheme.

The differential gearing and power-on combination provides ample braking for me. The only feature I would find helpful would be to lock the scooter wheels via the shorting method for when the scooter is on a hill.

How long do you usually get out of one charge?

I could design it such that the full E-brake (shorted terminals) can only be activated after a 5 second delay and the resistive load is initially used to slow it down. That should be good enough. That way, the motor won't stall immediately but will come to a quick stop.

 

Well with 24V, to dissipate 100W (peak) the resistor must be 5.76 ohms. Call it 6.8 ohms (closest value) which would dissipate 85W. Make each resistor 68 ohms as ten in parallel is 6.8 ohms. Or use 20 x 120 ohm resistors which would lower the load on the resistors a bit.

The resistors must be kept cool for reliable operation so a fan will be essential unless you can devise some other method of cooling them?

At first glance it seems a bit excessive as the entire braking process could take less than 5 sec. Ohmite makes some nice resistors that could be thermally mounted to an additional heat sink. I will check into this as well.

Experimentation will be necessary then.

I can see if I can't find additional spec by looking up the SN or Part No. on the net.

Nope SN or Part No. returned nothing at all.

How long do you usually get out of one charge?

Hard to say. I usually charge after every use. The scooter initially came with gel cell batteries, (2) 12V 32AH
I later switched to Lead Acid with 35AH
I am currently using Glass Matt with 40AH

I could design it such that the full E-brake (shorted terminals) can only be activated after a 5 second delay and the resistive load is initially used to slow it down. That should be good enough. That way, the motor won't stall immediately but will come to a quick stop.

'm thinking maybe 3 seconds. The scooter will generally be nearly stopped within 5 - 7ft of when I let go of the accelerator. It doesn't take long with the current resistive load on power on situations.



Brake

Throttle

 

I'm quite busy at the moment (school work) but I will finish the relay driver shortly.

So here's the plan:

When accelerating forward, motor is put into forward position using relays. MOSFET is driven with a PWM signal through a gate driver IC.

When accelerating backwards, motor is put into backwards position using relays. MOSFET is driven with a PWM signal through a gate driver IC. Max speed in reverse 50-60% full speed.

When pot released, MOSFET switching stops, and relays connect to the resistive brake. (Maybe make it so they don't connect to the resistive brake? But it would require 4 relays total, or another power MOSFET.)

When E-brake pressed, both relays connect to resistive brake. After 3 seconds (RC delay - approximate) the relay connects to the shorted position so the motor stalls. MOSFET switching is also inhibited.

So the final solution requires 3-4 relays and one or two MOSFETs to add all the requested features. Is this good for you?

I should be able to use two LM324s and some 2N2222 transistors for the most part, and a gate driver IC from Microchip.

I think it will be best to keep all relays off the main PCB - and just solder wires to the terminals and connect those to the PCB. This would also make the PCB smaller and cheaper.

 

If you use a FET halfbridge (2 FETS) instead of a FET and diode to do the PWM, it will regeneratively brake back to the battery.

The regen braking will be typical ie if PWM ratio is > motor speed the motor will drive forward if PWM ratio is < motor speed the motor will brake and current will be returned to the battery.

No need for "braking resistors" and much improved energy efficiency.

 

If you use a FET halfbridge (2 FETS) instead of a FET and diode to do the PWM, it will regeneratively brake back to the battery.

The regen braking will be typical ie if PWM ratio is > motor speed the motor will drive forward if PWM ratio is < motor speed the motor will brake and current will be returned to the battery.

No need for "braking resistors" and much improved energy efficiency.

We were talking about using regenerative braking but we were not sure how useful it would be.