PWM speed controllers cheaper than Victors?

[ProjectZero]

Hi guys,

This summer I'm working on an independent project to build a hexapod robot. I've all ready built the frame, attached the motors, etc. I was using 6 Victor 884s, powered by a 12V battery and controlled by an Arduino Uno. Problem is that when I leave school for the summer I'll have to give back the Victors, and they're waay to expensive for me to buy on my own. Does anyone know of any cheaper pwm motor controllers that work similarly? Thanks alot!

[MrForbes]

How much current do they really need to handle?

[thefro526]

This doesn't really answer your question but, have you tried contacting any local teams to see if they have any used Victor 884's or 883's? I know some teams have quite a few sitting around that they may never use.

[Chris Hibner]

In my first few years in FIRST, teams used to use Tekin speed controllers for RC cars. We used to use Bosch drill motors which had pretty similar power characteristics as the CIMs. They might work. The problem is that they weren't very reliable.

[nighterfighter]

1 Amp drivers-
http://www.sparkfun.com/products/9457

15 Amp-

http://www.sparkfun.com/products/9107

[StevenB]

Both RobotShop and Robot Marketplace have a pretty wide range of speed controllers. Just make sure you know what current you're going to be pulling.

http://www.robotshop.com/brushed-motor-controllers.html

http://www.robotmarketplace.com/products/speed_controllers_main.html

[DonRotolo]

It all comes down to the current requirements. And the cash available, I suppose. :rolleyes:

Sparkfun has a motoduino shield that handles two 2A motors for about $25.
You can build one that handles 2 channels of 2.5A each for about $12 - I have the schematic and a PC Board design if you want.
And there are more options, depending on whether you have an electronics junk box or not.

[ProjectZero]

Woah, thanks everyone. Guess I should have done a bit more research first. I tested today and even when stalled, the motors don't seem to pull more than 0.5 amps, and the typical no/minimal load current is 40 mA. I'm looking to go with nighterfighter's sparkfun suggestions - as basic (and cheap) as option 1 is, it looks like it will do what I need it to. As long as I make certain the Arduino can output a PWM signal in the correct range to control it, we should be good.

[RyanCahoon]

Hi guys,

This summer I'm working on an independent project to build a hexapod robot.

I tested today and even when stalled, the motors don't seem to pull more than 0.5 amps, and the typical no/minimal load current is 40 mA.

Just out of curiosity, what design are you using? Most table-top hexapods I've seen use RC servo motors, which shouldn't need external speed controllers, and you get position control for free (unless part of the project is writing the control loops yourself). Larger ones (e.g. able to carry humans around), I imagine, use discrete speed controllers, but those motors are going to pull more 0.5A...

--Ryan

[nighterfighter]

Sparkfun has a motoduino shield that handles two 2A motors for about $25.
You can build one that handles 2 channels of 2.5A each for about $12 - I have the schematic and a PC Board design if you want.


Here is the Ardumoto he is talking about- http://www.sparkfun.com/products/9815

I have one, very easy to use.

Don-

Can you post that PCB design and such? I'd be interested in it.

[dudefise]

Sparkfun has a 50A model if you need that kind of power - but its $59..still cheaper than jags/vics...
http://www.sparkfun.com/products/9668

[RyanCahoon]

Sparkfun has a 50A model

I've been tempted several times before by the controller you linked to, except that, as near as I can tell, the only control method are the pots that come attached to the board, so some circuit bending would be required, at the least. Also, it appears that it's uni-directional, so the applicability to robotics may somewhat limited, unless you're building a shooter wheel or something similar.

--Ryan

[billbo911]

W... As long as I make certain the Arduino can output a PWM signal in the correct range to control it, we should be good.

Yes, the Arduino can do PWM. The thing to keep in mind is what exactly you mean when you say "PWM".

When we use the term "PWM" when we are referring to FIRST robotics, it is usually understood that we are talking about "hobby PWM" specifications. Hobby PWM usually runs at ~50Hz with a pulse width of 1ms to 2ms. This is usually the signal that feeds a speed controller, a servo or a "servo motor" like a Vex motor.

The speed controller, servo etc. contain electronics that interpret the "PWM" signal and then modify the output to the motor accordingly. For example, Victors and Jaguars both put PWM signals directly to motors. The Vic at 100 Hz and the Jag at 15Khz IIRC. The key is, the outputs to the motor go from 0% duty cycle to 100% duty. They also reverse the polarity of the output to control the direction of rotation of the motor.

The Arduino is capable of generating both hobby PWM and 0% to 100% duty cycle PWM. In fact, the later is done natively in the Arduino.

[dtengineering]

...even when stalled, the motors don't seem to pull more than 0.5 amps, and the typical no/minimal load current is 40 mA. ...

Why not just use an L293D motor driver chip? $3.50. http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=296-9518-5-ND

Or save a buck and use the roughly equivalent SN754410 http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=296-9911-5-ND

If you need more power, check out the L298D.

I've been using these for years on mini-sumo robots based on the Tamiya Dual Motor Gearbox and Twin Motor Gearboxes.

Jason

[DonRotolo]

Can you post that PCB design and such? I'd be interested in it.Sure, give me a day or three to hunt everything down.

[Matt Krass]

Sure, give me a day or three to hunt everything down.

If you're interested, and you have the PCB artwork in Gerber format, you can upload it to the folks over at BatchPCB (which is run by the guys at SparkFun) and they'll make it available for ordering on demand.

Matt

[DonRotolo]

OK Folks, procrastination time is over.

This circuit is based on the "Test Circuit" on Page 5 of the LMD18201 Datasheet (http://www.national.com/ds/LM/LMD18201.pdf). You need to study the data sheet to figure out the input signals.

This is an H-Bridge. It does not decode R/C PWM signals such as used by Victors and Jaguars. Instead it takes a variable duty-cycle 'PWM' signal and amplifies it. There are several ways to convert R/C PWM to a duty cycle, including single-chip solutions.

The PC Board is simple enough to reproduce at home with stick-on tape. The set of 3 holes between IN+ and GND is for the 1200 uF capacitor; the extra hole allows for some component flexibility. The other 2 holes are for the 1 uF cap. Near the logic input side, the 2 sets of holes are for the 10 nF caps between pins 1 & 2 and 10 & 11. Pin 1 of the LMD18201 is marked with a square pad. You are looking at the COPPER side of the board!

You MUST use a heat sink with this chip, I suggest one about 2 or 3 inches tall and 1.5" wide, with fins, like this (http://search.digikey.com/scripts/dksearch/dksus.dll?pname&site=us&lang=en&WT.z_cat_cid=Dxn_US_US2011_Catlink&name=HS386-ND) (but you can find these cheaper)

You are well advised to also coat the thick power traces with a layer of solder, to increase the thickness and therefore current-carrying capacity. Not mandatory, but cheap insurance.

The image is about to scale - the board should be 2.5" long and 1.125 wide - but use the logic input pin holes (spaced exactly 0.100") as a reference.

Any questions, let me know.