Here are the basics of a brushless DC motor. Beyond the background info, they go into using microcontrollers to sense and switch the power. Do you have access to a Microcontroller programmer? If so, which brand?

If not, are you sure you understand how PWM will work with A brushless motor?

You can simply control the speed by controlling the switching speed (no pulse-width modulation, always a constant duty cycle, just faster switching frequency).

 

Here are the basics of a brushless DC motor. Beyond the background info, they go into using microcontrollers to sense and switch the power. Do you have access to a Microcontroller programmer? If so, which brand?

If not, are you sure you understand how PWM will work with A brushless motor?

You can simply control the speed by controlling the switching speed (no pulse-width modulation, always a constant duty cycle, just faster switching frequency).

I believe you mean commutation frequency. Switching frequency (in my mind) refers to PWM switching frequency. EX. changing from a 50% constant duty cycle @ 2khz to a 50% constant duty cycle @ 16khz would not increase speed.

 

You need to be cautious about minority carrier current in the output MOS transistors and free-wheeling diodes. Allow sufficient time for the transistor and diode current to drop to practically zero before the complementary transistor is turned on! You will have a possible short circuit between DC + and DC- if any of the six transistors gets stuck "on" it's complement must not turn "on" to prevent the loss of "magic smoke". Your motor nameplate data does not mention Hall effect sensors. You could attempt to purchase and install them, but improper phasing of these signals would release the "magic smoke". Texas Instruments sells a product called "easy spin" which does not require motor mounted sensors. My advise is to go that way being aware that a slow free-wheeling diode will contribute significant inrush current each time a motor phase changes polarity causing significant losses. ST is another semiconductor vendor with a full line of controllers and switching devices designed to control 3 phase brushless DC motors without sensors. Caution is necessary to protect your semiconductor and transformer investment. In your opening you mentioned 40 Amperes of motor phase current, you should have quoted 20 Amps. Look it up a 3phase motor has phase current of 1/Square root of 3 less current than the total power delivered to the motor. 1 horsepower is defined as 746 watts. Watts = Volts X Amp effective for the motor/efficiency. With 80% efficiency, the full power current per phase will be ((746/48)/.8)*1/sq. rt (3) or 11.2 Amperes.

 

Good 5 volts for the hall sensors. So the IC in post 7 will work. I would just build a simple PWM using a 555 to control the speed. I think there are lots of them on the web.

 

I still say that using the MC33035 in my post #8 is the easiest way to go with this. Using this chip gives all of the logic needed to drive a BLDC. The out board power mosfets can be chosen for any voltage and amperage that is needed by the motor. It is a proven chip that has been used for a long time, no microcontroller needed, but one can be used along with it if wanted.

Edit) it even gives a 5V output for the Hall sensors.

 

The chips are almost the same. With the 33035 you need to build a driver (X3) to interface the low voltage of the IC to the higher voltage PFETs or build a charge pump to use NFETs. With the other one you need to build a pwm and it uses more efficient NFETs.

 

I am working on a school project and i'm working with a design team to make a simple motor controller for a 48v 40a bldc motor. We are just a high school class and have no experience with this topic. Anything you could do to point us in the right direction would be great.

Wow. A 48v 40A 3-phase sensored brushless motor controller is so far away from a "high school project" that you might as well ask us how to build a Space Shuttle for your science fair.

I've seen good designers with many years experience smoke up hundreds of dollars worth of FETs and PC boards making 40A motor controllers from scratch, and you say you "have no experience with this topic".

It's time to re-think what you are doing. Either go to a brushed DC motor, that you can control with PWM and one large simple FET, or just buy a 3-phase brushless motor controller to suit your motor from one of the hobby robotics suppliers.

 

Wow. A 48v 40A 3-phase sensored brushless motor controller is so far away from a "high school project" that you might as well ask us how to build a Space Shuttle for your science fair.

I've seen good designers with many years experience smoke up hundreds of dollars worth of FETs and PC boards making 40A motor controllers from scratch, and you say you "have no experience with this topic".

It's time to re-think what you are doing. Either go to a brushed DC motor, that you can control with PWM and one large simple FET, or just buy a 3-phase brushless motor controller to suit your motor from one of the hobby robotics suppliers.

I started a reply along these lines but I didn't want to be the lone naysayer. A couple of years ago I delved into making a high power brushless controller, and WOW. This is not a project for the novice. After a couple of months of research, I decided to use a brushed motor with only a single low-side switch controller. Even that would be a challenging project for a high school team. Add five more switches and the complexity goes up by an exponent for each one. I am wondering if OP goes to a private school for exceptionally gifted savants or his instructor is an idiot, or both.

 

Wow. A 48v 40A 3-phase sensored brushless motor controller is so far away from a "high school project" that you might as well ask us how to build a Space Shuttle for your science fair.

I've seen good designers with many years experience smoke up hundreds of dollars worth of FETs and PC boards making 40A motor controllers from scratch, and you say you "have no experience with this topic".

It's time to re-think what you are doing. Either go to a brushed DC motor, that you can control with PWM and one large simple FET, or just buy a 3-phase brushless motor controller to suit your motor from one of the hobby robotics suppliers.

I started a reply along these lines but I didn't want to be the lone naysayer. A couple of years ago I delved into making a high power brushless controller, and WOW. This is not a project for the novice. After a couple of months of research, I decided to use a brushed motor with only a single low-side switch controller. Even that would be a challenging project for a high school team. Add five more switches and the complexity goes up by an exponent for each one. I am wondering if OP goes to a private school for exceptionally gifted savants or his instructor is an idiot, or both.

We are actually just a group of students that are competing in the Shell Eco-marathon. Last year we purchased this motor and its accompanying motor controller; however, this year they changed some rules that now requires us to build our own motor controller. Honestly it is probably the worst rule to change because I foresee a lot of car's motor controllers and motors possibly catching fire or blowing up, but we can't change the rule.

After doing a lot research building a BLDC motor didn't seem impossible, but definitely complicated. We were hoping to be able to get away without buying a brushed motor controller, but that seems like what we should do. We would still have to build a motor controller for the Brushed DC motor, but it would be a TON easier.

Do you happen to have any advice on that???????

 

We are actually just a group of students that are competing in the Shell Eco-marathon. Last year we purchased this motor and its accompanying motor controller; however, this year they changed some rules that now requires us to build our own motor controller. Honestly it is probably the worst rule to change because I foresee a lot of car's motor controllers and motors possibly catching fire or blowing up, but we can't change the rule.

After doing a lot research building a BLDC motor didn't seem impossible, but definitely complicated. We were hoping to be able to get away without buying a brushed motor controller, but that seems like what we should do. We would still have to build a motor controller for the Brushed DC motor, but it would be a TON easier.

Do you happen to have any advice on that???????

Yes, I would definitely go with the brushed motor controller. This will allow you time to focus on building the car. Otherwise this controller is going to consume the whole of your build and you probably still won't get it right. You could; I'm not saying it's impossible, but it's not practical.

I applaud you for entering the eco challenge. I was at eco challenge marathon last year in Houston as a spectator and it looked awesome; made me wish I was in school so I could enter the competition.

The brushed motor controller can be quite simple if you only need to go one direction, with all the heavy lifting performed by a single MOSFET or IGBT. The mosfet or IGBT should be the module type (there are several packages) with super low RDSon and a large thermal transfer pad at the bottom, and bolted to a heat sink. It should be over-rated well above the intended volts and amps to be safe. Not everybody will agree; if you look inside a kelly or alltrax, controller, instead of a robust module, you'll see a bunch of parallelled to-220 MOSFETs, which is chicken **** in my book, and you'll also find a bunch of pictures online of these controllers with their cute little FETs blown to smithereens.

When I was planning my 300A 72V brushed motor controller for an electric go-cart project, I bought the following:
IXYS VMO1200-01F 1200A 100V MOSFET module.
IXYS 200V 582A diode module
Motoenergy ME0708 8HP (19HP peak) brushed DC motor. more info

I never got around to building the controller or the cart. The stuff still sits in my garage. I am willing to sell it for less than I paid, if you want to go that direction. But this stuff I have is probably overkill for you eco challenge build. I was planning a high performance toy, not a model of efficiency.

 

I'll respond in your new thread, here;
http://forum.allaboutcircuits.com/showthread.php?t=92117