Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Motor Driver

Status
Not open for further replies.

cosmonavt

Member
I am new to the field of robotics. I have googled up motor drivers to learn that they are used to control the speed of a motor and acts as an intermediate device between the microcontroller and the motor.

If we are getting 5V out of an MCU, can't we simply use a transistor to amplify this to 12V to power on a motor? What are the drawbacks of this method? When it comes to Pulse Width Modulation, can't we simply turn on and off the voltage from the output of the MCU?
 
Yes and Yes. However controllers will do more like control direction and braking. Thay can be used to take some of the prosessing off the micro and save on I/O ports. Andy
 
Last edited:
Well if I add a separate brake system to the robot and control direction using multiple outputs from the MCU, do I need a controller in that case?
 
You can do it directly with transistors, that's how I usually do it. I will add a couple of gates or transistor logic to the "H-drive" (forward and reverse) to ensure that the high and low drive transistors do not come on the same side at the same time. This will cause "shoot through", a short circuit, which will blow the drivers.

With ideal programming and operation this would never happen, but in practice it happens a lot - bad code, electrical glitch that sends the CPU into the weeds, something.

The "brake" generally consists of just turning the two low-side drivers on at the same time, or give it a kick into reverse for a more pronounced effect. This works great with small motors. For anything pulling more than twenty watts or so, you want to start PWM'ing the brake. For motors more comfortably measured in HP, you probably want to start using a separate dynamic or regenerative brake circuit with its own power resistor.
 
Last edited:
Thanks a lot guys. Yes, the H bridge is what I was looking into, could you please direct me to some comprehensive tutorial for making an H bridge?
 
Better to buy 298 H bridges chip (2A) rather than making it with transistor/MOS....
 
Last edited:
Thanks a lot guys. Yes, the H bridge is what I was looking into, could you please direct me to some comprehensive tutorial for making an H bridge?

Before you go down that route, you need to ask yourself some questions:

1) What are specifications of my motors (voltage and current needs, particularly worst-case current needs, called "stall current")
2) What is my experience with designing and building an h-bridge (I would say nil at this point, but I could be wrong...)

Depending on the answer to "1", the fact that you have no experience may not be an issue. If, for example, your motors were small "hobbyist" motors pulling only (at max) maybe a few amps when stalled, then not having any experience may not be a real issue, and you could learn a lot along the way. You might blow a few parts, and spend a little money, but what you gain in knowledge and experience will probably outweigh that. If, on the other hand, you are more interested in getting something working, and not looking at how to learn to design h-bridges, then using a motor driver IC (like an L293 or L298) might be the better way to go.

Now - if your experience is low (which I suspect it is), and the motors you need to control have a stall current rating of over, say, 10 amps - you are on "shaky" ground; if they are over 25 amps - I would say forget it. In either case, you might do well to give long and hard thought at what you want to accomplish, and what your end goals are. This is simply because at those current needs, learning how to design an h-bridge to handle them (whether that h-bridge is composed of discrete transistors, or more likely, discrete MOSFETs) is something that could lead you down the road toward spending a lot of money. You could easily end up blowing $5.00/ea USD (or more) MOSFETs on a regular basis. Now - if your goal is to learn - really learn - large-scale MOSFET h-bridge design, then this expense might just be considered part of your education. But, if your goal is to build a running robot, you might end up in a "money-sink" situation.

Just some things to keep in mind and look into, before you make your decision. Good luck, whatever option you take! :)
 
Yes, the problem is that I want to build a running robot which would weigh no less than 1 KG and I am using differential drive (2 motors). If I choose this motor:

**broken link removed**

and use two of these, would it classify as overkill? I am new to the field of robotics but when it comes to calculations on amperage, voltage and stuff, I am OK with that.

Also, I found a tutorial on this site:

**broken link removed**

Which uses only two logic chips (they cost around 0.25 dollar) and a bunch of transistors (even cheaper). That way it won't lead me to a money sink situation...
 
Yes, the problem is that I want to build a running robot which would weigh no less than 1 KG and I am using differential drive (2 motors). If I choose this motor:

**broken link removed**

and use two of these, would it classify as overkill?

I don't think so, as long as you keep the weight of the robot (and everything on it) under 2-3 kg; those motors are fairly small, and look to have only bushings for bearings; so side loads on the shafts should be kept fairly light. The torque will probably be ok (keep the wheels to around or under 10 cm or so in diameter); you could mount the wheels on shafts, and use small pillow blocks or something to take the side-load off the motor and gear housing, if you wanted; this will allow you to carry more weight without overburdening the motor as much.

I am new to the field of robotics but when it comes to calculations on amperage, voltage and stuff, I am OK with that.

Also, I found a tutorial on this site:

**broken link removed**

Which uses only two logic chips (they cost around 0.25 dollar) and a bunch of transistors (even cheaper). That way it won't lead me to a money sink situation...

Before you sink any money into an h-bridge solution, you need to know the specs of the motors you'll use; I noticed that the motor you linked above didn't mention any current specificiation. I think the motor will likely be fine for a small robot like you are looking at, but you still need to know those current specs (otherwise you'll end up under or oversizing your h-bridge; over is ok, as long as you don't get carried away and spend more than necessary, under is bad, though, obviously).

So - either find out those specs before or after you purchase the motors, I guess. That h-bridge you linked to might be ok for these motors, but then again, I don't know the specs (I'm merely basing that opinion on what I have seen spec-wise with other similar motors). You might find that you can run those motors you linked using something far simpler to assemble, like an L298 (or if you are really lucky, and the motors are efficient, an L293, or two of them stacked).
 
I noticed that the motor you linked above didn't mention any current specificiation.
I noticed that to, weard thay would leave out the current spec. Maybe the OP can call or wright to get the current specs.
 
I noticed that to, weard thay would leave out the current spec. Maybe the OP can call or wright to get the current specs.

Probably be better to write; even then, being that it's drop-shipped from china, you may or may not understand the response (some suppliers have great english skills - others not so much)...

Regarding those driver boards you listed - they seem a bit pricey; I mean, if you're going to get your motors from China, you might as well get the driver boards too (either can be had for around $10.00 USD off ebay). It would be cheaper still just to buy the ICs and brew it up yourself (though I will admit, the L298 is a pain to work with - it doesn't have a "standard" 0.1 inch pin spacing, and heatsinks are hard to find for it).

Still - they were great links for the OP to know what to look for, if he didn't want to go the homebrew route... :)
 
Actually, I have to build the H bridge from discrete components so cannot use L298.

And yes, I have contacted the supplier to tell the amperage.
 
Actually, I have to build the H bridge from discrete components so cannot use L298.

And yes, I have contacted the supplier to tell the amperage.

Is the reason you can't use a pre-made h-bridge (like the L298) because it's a school project (or something where you need to learn the operation of the components)?

Once you know the amperage (the stall current is the main one you want; build the h-bridge for this plus maybe 10-15% overage, and you should be ok), you can select the components for the h-bridge properly. Remember that you may need (or want) a heatsink for the switching elements...
 
OK he says 0.16A. Wouldn't this produce too little a torque for the robot to budge forward, let alone run?

Also as per my knowledge, the RPM is max when there is no net force, then would the wheels really reach 60 RPM at terminal speed? (when all the resistive forces have been overcome)

And yes, this is a project, not a school one, but I want to make it out of discrete components. Is there any hope of getting a motor of a decent torque and RPM within 10-15 dollars?
 
Last edited:
Is there any hope of getting a motor of a decent torque and RPM within 10-15 dollars?
Well yes, have you looked at any of the online robot shops?
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top