PDA

View Full Version : Please help me!



Judge
05-30-2008, 01:34 PM
I'm new to this forum and fairly new to the electronics aspect of robotics. The information online is very confusing, maybe someone can help clarify it for me. Will my 400 watt 12 vdc battery be able to power 4 x Banebots 64:1, 36mm Planetary Gearmotor, RS-540 Motor http://www.trossenrobotics.com/store/p/4908-64-1-36mm-Planetary-Gearmotor-RS-540-Motor.aspx? I don't know how to understand amp calculations. I had thought a 12vdc 400 watt battery would provide only 33 amps, but this website says that a 3.3ah battery can provide 60 amps continuous; I don't understand http://www.robotmarketplace.com/products/BPK-3300N-12.html. My rover will be about 20 lbs., so if I can't use those motors with my 400 watt battery, what should I use? Also, how much torque should I have to move 20 lbs.?

Adrenalynn
05-30-2008, 01:54 PM
First, welcome to the forum!

Second - your battery is, I assume, 400 watt HOURS, right? Or specifically, 33Ah, right? Amps = Watts/Volts Watts = Amps * Volts (roughly, we're assuming no resistance for the moment).

The measure of Amp Hours or Watt Hours is a little touchy. The basic concept is: "How long can this thing deliver power to my device, what point will it fall to a point that it won't operate anymore".

Assuming that you have enough torque with four of those motors to operate at a kind of medium current draw, we see that the motors will draw somewhere between 1A with no load, and 42A if the motor is stalled (which you really don't want to do).

Let's just pick a number and say we want to put 20A into each motor. That's 80A total. A 33Ah battery would power the motors for somewhere in the neighborhood of 20 mins.

You didn't note what kind of motor controllers you're using, and you didn't note what technology battery.

A 33Ah gel-cell can deliver almost 400A into a direct short to ground (which is basically what a stalled motor is) in my experience. And a LiPo can burst into flames and burn your house down quickly. It's ALWAYS important to fuse them under what the stall current is!

Anyway, if you can give us a bit more information - what size are your wheels, are you doing any other gearing, are you open to doing any other gearing, what kind of motor controllers, what kind of battery technology, ... then we can probably help out more!

Judge
05-30-2008, 02:06 PM
Thank you for the prompt reply. I intend to use 5" wheels. Here is the website for my battery http://www.batterygeek.net/product_p/spps-12-444_batterygeek.htm. My rover will use BaneBots Motor Controller (45A peak) if I use the Banebot motors. How much torque is required to move a 20lb. robot? What type of battery should I use and where can I buy it? I want enough power to last for hours of use. Thanks

Adrenalynn
05-30-2008, 02:50 PM
I'm swamped, and hopefully someone else will come along shortly.

I'm going to write a tutorial on this as soon as I have a few hours. I promised Alex I'd do it at some point...

[I'm just going to bang some rough numbers and assumptions and deal with explnation later]

0.0052083 * in-oz = lb-ft

About 11ft-lb at peak (not stalled) per motor
44ft-lb total.

After running the numbers, I would guess that your four-motor design on a 20lb bot with well inflated 5" rubber tires would be able to climb a 20 degree incline max on pavement, would have a top-speed of 3 ft/sec, and would get there at 0.5ft/sec^2. If you need less climbing, we can regear it and get some more speed or acceleration...


Your motor controller isn't big enough. You would need four of those motor controllers.

How are you going to turn - slip-skid or servo-based steering car-style?

Hours of battery? More of those batteries (ouch) or increase the robot weight and add cheaper battery technology...

Sorry - brief answer, I know. I'll put my numbers out there so everyone can check 'em this evening...

Judge
05-30-2008, 03:01 PM
OK, so 4 x BaneBots Motor Controller (45A peak) and 4 x Banebots 64:1, 36mm Planetary Gearmotor, RS-540 Motor. Do you think those motors are overkill for a 20lb robot? I certainly can't spend more than $1000 on battery power, so what would you guys recommend using? My issue with all this is that this is a house bot. I want long lasting, lightweight, and efficient. If we don't count batteries and motors, my bot will weigh only 8 lbs. There must be a simpler way to move around such a bot. I'm not as experienced as you guys. :veryhappy:

Adrenalynn
05-30-2008, 03:13 PM
What do you need the bot to accomplish? My little competition bot weighs 7lbs with batteries, runs for 1hr 15mins at full power, can climb a 45deg incline, and moves about 1.5mph. Total out-the-door build cost is about half of one of your batteries, including batteries.

4mem8
05-30-2008, 03:49 PM
M'mmm this is an interesting post and could go on for a while, Before any of this can be calculated as Adrenalynn has stated you have to provide all of the information that you want your bot to do, This is very important in deciding how to approach your requirements Judge, But it sounds like you and Adrenalynn are in control of this now. Good luck with your project.

Adrenalynn
05-30-2008, 04:19 PM
Please, by all means jump in, M'!

I found another quantum singularity in my garage. Beating it back is occupying all my resources.

Where'd I put that spare warp-core? Gonna need it to power those superconducting magnets...

Judge
05-30-2008, 06:16 PM
Well, my goals for this robot are mainly accademic. Physically, I want him to drive around on a level surface indoors. It's simply a rover with a brain. However, I don't want him to last for an hour or even 5; I want him to last all day like 12 hours. My original plans (using the 444wh battery and 4 x Lynxmotion motors) would have lasted for many hours, but then I realized I didn't have enough torque to move an 8lb bot and 9lb motor :genmad:. I could use my old gel-cell wheelchair battery to power the Banebots motors and a 200 wh battery from BatteryGeeks to power my sensors, etc., but then I might just have a reoccurring problem. Heavier batteries require stronger motors that need more current and bigger batteries. Where's the perfect balance?

Adrenalynn
05-30-2008, 06:30 PM
12hr runtimes? Sometimes reality must impact our desires... ;)

Maybe it's time to start thinking about how often you're willing to chase it down and drop in more batteries?

Judge
05-30-2008, 06:34 PM
You think 12 hours is impossible?! My electric wheelchair lasts me 15 hour days at Disney (miles of driving). I didn't know this was just a forum for people building toys.

LinuxGuy
05-30-2008, 06:44 PM
You think 12 hours is impossible?! My electric wheelchair lasts me 15 hour days at Disney (miles of driving). I didn't know this was just a forum for people building toys.
I don't think what we build can be called toys, considering how much money and time we sink into them.

These are real robots, and they move in all sorts of different ways. It's what research is all about. :veryhappy: :veryhappy: Just because we might deal on a different scale from universities and government agencies like NASA, it does not mean we can't make real breakthroughs and contribute to robotics.

Right now I am into researching autonomous robots with manipulators, and the arm I have designed is quite different from what you would expect to see. I have never seen an arm with a double joint. My robot project has been in development for almost two years now. I don't expect W.A.L.T.E.R. to be finished anytime soon. I still have to work on the "L" and "E" parts.

8-Dale

Judge
05-30-2008, 06:52 PM
Perfect then, now all I need is a battery that can power 4 motors that can in turn carry the battery and an 8lb. bot for 12 hours (not constant use). Any suggestions? Btw, the toy comment was just sarcasm for the guy who implied that my plan is impossible.

Adrenalynn
05-30-2008, 07:00 PM
Naw, I don't think it's impossible. If you've got the money the design can be done, certainly. I can gettcha months of operation with a $50,000 fuel cell...

If I were you, I'd scale the project size down and start looking at low-amperage motors. Figure a 3lb robot and a couple <1A Solarbotics motors. Or scale it up and start looking at batteries with real current delivery.

As far as "toys" go, my current top project would run over your wheel chair, chew it up, and drop it in its own self-contained garbage can... My batteries could power those motors for days - but the motors you're looking at couldn't move their weight...

Adrenalynn
05-30-2008, 07:02 PM
Incidentally, that "guy" is lacking a few parts that would qualify her for such title and secondarily, that "guy" is the one that does calculus in her head to compute the motor specifications for your project. Just something to think about...

Judge
05-30-2008, 07:08 PM
Ok, chica. What are your batteries?

Adrenalynn
05-30-2008, 07:30 PM
I'd expect that a 10lb bot running your battery proposal (which is actually pretty darned efficient weight::power) would run for pretty close to your 12hrs, climbing a max 10deg incline, but would only do about 2'/sec, or roughly 1.5mph, utilizing four of these: http://www.trossenrobotics.com/store/p/4261-Gear-Head-Motor-7-2vdc-30-1-291rpm-6mm-shaft-.aspx Shrink your wheels down to 3" or so.

As the 'bots weight goes down the speed will go up, and if it's substantially lighter, it can be regeared...

As far as what I'm running - 80Ah gel-cells x 6. That's a couple hundred pounds driven by over two thousand dollars in motors and about the same cost as your battery when it's all said and done. Less weight::power efficient, more cost efficient for large applications. It has trouble with knocking off the corners of walls at the moment - about 3.5'x2.5'. ;)

4mem8
05-30-2008, 08:12 PM
Adrenalynn, That's some serious robot, What electronics are you using ? CPU,Sensors ect. or ITX?.

Dave
05-30-2008, 08:38 PM
OK, so 4 x BaneBots Motor Controller (45A peak) and 4 x Banebots 64:1, 36mm Planetary Gearmotor, RS-540 Motor.

I have a few comments. Food for thought.

Calculating battery run time for robots can be kind of tricky because there are many variables that come into play when your bot is stopping, starting, spinning around in circles, bouncing off walls, etc. I ran four BaneBots 36mm 64:1 RS-545 (similar to the 540's) motors continuously at full throttle for about an hour and a half on a 7.2 Ah gell cell battery. Depending on the surface it's rolling on and what you have stacked on it, and also taking into account that the motors probably won't be running full throttle the whole time... your mileage may vary.

In the past, we've split the output of 1 BB 45A motor controller to drive two motors similar to the one's you'll be using. It worked fine, but we didn't run it that way for an extended period of time.


You think 12 hours is impossible?! My electric wheelchair lasts me 15 hour days at Disney (miles of driving). I didn't know this was just a forum for people building toys.

My advice is based on the gear that I use. Personally, most of my experience has been with the motors that what we sell, which are basically RC car motors with gearboxes. Undoubtedly, your chair has a much beefier motor and some serious batteries. Also, I won't take offense at the "toy" comment, because I've never built a robot that performed any useful function beyond my own amusement. ;) However, I can't speak for some of our friends...


As far as "toys" go, my current top project would run over your wheel chair, chew it up, and drop it in its own self-contained garbage can... My batteries could power those motors for days - but the motors you're looking at couldn't move their weight...

Easy there, killer. :p

Adrenalynn
05-30-2008, 08:39 PM
I don't wanna hijaack this thread - I'll blog more later. At the moment 2xPicoITX 2xArduino, 3x custom IO board, 3x SSC-32. That's the going plan right now, anyway. I'm sure it will grow. ;)

Adrenalynn
05-30-2008, 08:45 PM
>> full throttle for about an hour and a half on a 7.2 Ah gell cell battery

33/7.2 ~= 4.58 which we can call our scaling factor
4.58 * 90mins = 412mins / 60 = 6.87hrs < 12hrs.

As you note, that's not taking into account starting which is a momentary stall and a huge current sucker. Stopping, with a good regenerative motor driver, can actually help offset that.

I like to calculate for near worst-case. I was also running the motors at 74&#37; of max load in my calculations, bursting to 95% max load. Sounds like yours were [virtually] unloaded?

Given your experiences, I'd still like to live with my initial derived numbers. But, as always, I *love* to hear of your experiences (and others) because it helps me build better real-world models - so a hearty THANKS!