Motor Controller

[Omar.M]

I've got great news.
I was reading some threads and I stumbled upon a fellow member of Electro Tech Online who has an online store (quite small).

I found he was based in the same province as me, so I contacted him regarding the motor ICs. He's agreed to get some chips for me!

Seriously, I am very happy right now. Just in case you guys are wondering:
https://www.dipmicro.com/ is the site.

I guess small businesses really do know how to run a good business, compared to bigger stores in Ottawa.
Omar

 

[Hank Fletcher]

But at the end of the datasheet they show external clamp diodes on a stepper motor.

Yeah, I saw that, too, so I was hesistant to criticize your original comment. I don't know what to make of that either, all I can say is that from experience and with DC motors, you don't need any extra diodes.

 

[blueroomelectronics]

The SN754410 Has clamps according to this, yay. Nice part and still made. I'll use it instead of the L293D on Mongoose.
https://www.kronosrobotics.com/an101/AAN101.shtml

 

[theinfamousbob]

The SN754410 Has clamps according to this, yay. Nice part and still made. I'll use it instead of the L293D on Mongoose.
https://www.kronosrobotics.com/an101/AAN101.shtml

How much voltage are you planning to run through it? I used the Tamiya twin gearbox with a 6v supply (4x1.5v AAs) and stacked three of the SN754410s...they still got quite warm.

 

[blueroomelectronics]

5V (4 x AA NiMH 1.25V each) Warm is fine. Hi Current chips tend to run warm, lots of copper on the PCB GND pins to help it along.

 

[Hank Fletcher]

stacked three of the SN754410s...they still got quite warm.

Each SN754410 comes with a generous amount of four ground pins that are also heat sinks for the chip. If you stack three, you're still just using the four pins on the bottom chip for heatsink, so you're dividing the heatsink in total by three. Add to that the fact that you're sandwiching a chip in the middle, and the result will be that all chips will run much warmer than if, for instance, you ran them side by side. Don't stack more than two if you can avoid it, and put a heatsink on the top one.

Why did you use so many for the Tamiya's? What's the peak stall current for those motors? Surely it's nowhere near 6A, which is what you'd have with 3 h-bridge chips.

 

[blueroomelectronics]

Peak stall is about 2A, that's why I'm using both a 1A thermal fuse (PTC) and stall (rotation) sensors.

 

[Omar.M]

Sorry this may be a stupid question, but what do you do with the stall sensors?

If the motor is stalling then you stop outputs or something? How expensive are these?

Thanks,
Omar

 

[blueroomelectronics]

Sorry this may be a stupid question, but what do you do with the stall sensors?

If the motor is stalling then you stop outputs or something? How expensive are these?

Thanks,
Omar

The rotation sensors do double duty as stall sensors. If you've supplied power to the motor and the rotation sensors aren't changing then you have a problem. So turn off the motor power.

 

[electric.engineer]

hi. i have started to search about dc motor control recently.
i have a project about dc motor control.
i'd like to use a mosfet, its driver, pic 16F628A, and a dc motor.
and i'd like to generate a pwm signal by pic.
**broken link removed**
i'm affected from this site and i'd like to make the second circuit(Low Side Driver Block Diagram)
could you help me about which components i should to use and how i can obtain assembly codes of pic to generate a pwm signal and also how i can control the speed of the motor(by which component i.e will i use a potentiometer/pot?)
thanks

 

[bloody-orc]

OT:
btw how is your mongoose coming along blueroomelectronicsdudewhoserealnameIstilldon'tknow ? haven't seen ant refreshes on the topic of yours. if you are stuck with something maybe i could lend a hand (or two)...

 

[blueroomelectronics]

My name is Bill, I often put a project aside in it final stages to think it over before going to manufacture. Here's the Chassis, it'll get built first.
**broken link removed**

 

[bloody-orc]

Very nice work Bill. glad to hear you are still "playing" with it

 

[theinfamousbob]

Each SN754410 comes with a generous amount of four ground pins that are also heat sinks for the chip. If you stack three, you're still just using the four pins on the bottom chip for heatsink, so you're dividing the heatsink in total by three. Add to that the fact that you're sandwiching a chip in the middle, and the result will be that all chips will run much warmer than if, for instance, you ran them side by side. Don't stack more than two if you can avoid it, and put a heatsink on the top one.

Why did you use so many for the Tamiya's? What's the peak stall current for those motors? Surely it's nowhere near 6A, which is what you'd have with 3 h-bridge chips.

They were lying around, already stacked and soldered in a box 'o junk. I had limited board real estate so I figured there wouldn't be a problem. I threw a generous amount of solder between all four ground pins to act as the board heatsink. It was the BotBoardPlus board (google it), and the proto space is limited, with plated through holes.

 

[Omar.M]

I am no expert yet, but I think at one point you're going to fry the gearbox motors.

This discussion was based upon the fact that the motors run on very little current / voltage, so too much might burn them. You might be giving them TOO TOO much. Surprised they didn't explode and go KA-BLOOOOOO-EY!

Omar

 

[blueroomelectronics]

Are you sure about needing diode clamps for the SN754410? I was under the impression that they were built-in on this IC, and that that was one of the major improvements over the L293D. I've never had any problems running DC motors without any extra diodes on the SN754410 .

As in a furthur post, the SN754410 has the clamps.

 

[Omar.M]

Well, hopefully I shall have all the parts I need in about a week. Until then I need to figure out some things regarding the PWM pulses. It seems that the online shop I'm getting the TI motor controller from doesn't have too many pots and such. So, I was hoping to get a drive to Active and possible pick out a few of those as well as the chip and capacitors.

Now, I was just wondering, which potentiometer / trimmer should I go for? I mean the resistance value; referring to here:
**broken link removed**

The author says 100k pot, but is it possible to use 10K, 50k, or 25k? I want a high duty cycle, because I am merely doing this PWM so the motors don't burn out-- I wish not for my robot to move very stiffly if you know what I mean.

Those are the only pots I have around here, if I need a higher rating, can I just somehow increase it by adding a resistor?

Secondly, I was just randomly wondering (as I am breadboard person, not proto PCB person) that if you can just stick stuff into the proto PCB holes and have it connect, instead of having to solder it down. I've never used them and have always wondered!

By the way, I've got the basic look of the robot done. It's no where near as good as the ones Bill comes up with, but I'll take a picture soon and show it to you!

 

[Hank Fletcher]

I want a high duty cycle...

No, you don't. You want a duty cycle that will limit the voltage coming out of the h-bridge chip to your motors to a maximum of 1.5V (presuming that's still the top voltage you want to run the motors at). What you're thinking of is the PWM frequency, which does have to be relatively high to keep the motor rotation smooth (capacitors excepted to illustrate point). For instance, if the PWM frequency was 1Hz and the duty cycle was 50%, the power to your motors would be turned on for a half second, then off for a half second, then on again, and over and over resulting in jerky robot movement. The link you gave suggests a PWM frequency of about 150Hz, which would be okay for your application, but know that PWM frequencies for motors are usually in the 15,000 to 30,000Hz range. That way the PWM frequency is out of range for human hearing, so the impression is that your motor is running quieter.

I was just wondering, which potentiometer / trimmer should I go for?

At the heart of the 555 as an astable multivibrator (or oscillator) is the RC circuit. No, I don't mean radio-controlled. What I'm meaning is Resistance-Capacitance, which is a basic building block of electronic circuitry that is worth every tinkerer's time to check out, so do a search and read up. The nature of an RC circuit is that it oscillates the charging and discharging of the capacitor in a manner that can be explained in a simple equation:

T = R x C

where T is time in seconds, R is resistance in ohms, and C is capacitance in farads. Multiply the resistance times the capacitance, and you have the amount of time it will take the capacitor to charge and discharge once. This is what controls the rate at which the 555 oscillates, and consequently what your PWM frequency will be. Because of the nature of the equation, you can use a greater resistor and lesser capacitor, or a lesser resistor and greater capacitor, and still end up with the same frequency. Everything after that is just fine tuning. In short, the pots you have now should work fine as long as you can match them up with a capacitor lying around the house that will result in a PWM frequency you can live with. I still say go for the pack of 12 assorted trim pots from the Source, since they're nice and small in size.

I have some issues with the way your link set up the 555 timer, not that it's wrong, it's just not the way I'd do it. It just looks a little more complicated than it has to be. Search for some schematics demonstrating how to set up the 555 as an astable and compare. I've got a 555 here I'm tinkering with to PWM my nose clippers, so get one (a 555, that is) and start playing with it and we'll swap notes. Remember, it's the duty cycle you want to have control over, not the frequency, although you can use pots to control both if you wish. Use your voltmeter on the 555 output pin to verify the change as you adjust the duty cycle. Check the output pin of the 555 with a 330ohm resistor and an LED, and it should dim and get brighter as you adjust the duty cycle length (presuming the frequency is greater than 30Hz). If for kicks you drop the PWM down to 1Hz, then the LED will flash once per second, and the longer you make the duty cycle the longer the time the LED will stay on for each flash.

I was just randomly wondering (as I am breadboard person, not proto PCB person) that if you can just stick stuff into the proto PCB holes and have it connect, instead of having to solder it down. I've never used them and have always wondered!

Although I've seen proto-PCBs that have a configuration similar to that of a breadboard, even then you'd need just as many jumpers as you would need for the same circuit on a breadboard, and yes, you'd have to solder whatever you're connecting to it to make it "stick." So it's time to buy a soldering iron. I've never used proto-PCBs either, I'd just rather use a breadboard and then design a custom PCB if I needed it. I imagine you're going to high school soon, maybe next year? Technology classes vary widely from school to school in Ontario, so make sure whatever school you choose has a good electronics program and lab. Call the school to ask what they have, and try to arrange a tour/interview with the electronics tech teacher. A good high school lab should have its own PCB design and production equipment. Using your school's lab will save you the overhead of buying chemicals, boards, hardware and software for your own lab. Also, if you can convince your teacher to let you integrate your robot project into the course curriculum, you can work on your robot during school time and get credit for it!

 

[Omar.M]

I'll reply to the rest of the comments later, but I just wanted to point out that I am in high school already. I do have a tech program that is pretty good, but the thing is that the teacher hates me and thinks I am stupid.

I do think we have PCB making thing, but I doubt they'd ever let me touch it. Afterall, I am stupid, right?
I do have a soldering iron, BTW, and I can solder
I will reply to the first part when I get home.

Home: Thanks, I will search Google for some timer information, wont be too hard I don't think. The reason I was asking about the the potentiometers was because I am going to get a drive to ACTIVE and so I was thinking of picking up one of their grab bags.

Now, the reason I like PCB proto boards is because they are very neat and tidy. If I can get a few, I can permanently solder the power regulator circuit to one board, some motor driver stuff into another. I think this is good because it saves me having to recreate the circuit again and again when I want to create a new robot.

What I was wondering was, what is the duty cycle range I am looking to get, so that my voltage output is about 1.5V - 3V?

It seems you're quite under estimating me, I know my teacher thinks I am stupid, but I know a few things here and there .

 

[Hank Fletcher]

What I was wondering was, what is the duty cycle range I am looking to get, so that my voltage output is about 1.5V - 3V?

That depends on what the motor supply voltage to the h-bridge will be. If the supply voltage is 5V, then you'll want a duty cycle of 50% to get 2.5V.

I am stupid, right?

I wouldn't have said that. In general, make sure when you ask someone a question you listen to and understand the answer before you ask the same question again. It sounds like you have some respect issues to work out with your tech teacher. I'm surprised you're looking for answers and equipment online when you should be getting it in your classroom. Take the initiative in solving your problem, and talk it over with your teacher. While I think it's unlikely that your tech teacher hates you, if that really is the case then you have a problem you ought to take up with the principal or vp. If you're getting good grades in tech class, then your impression that you're stupid is unfounded. If you're not getting good grades, pull your socks up, because it sounds like tech studies are of interest, value, and otherwise important to you.