what dc motors to buy with what bridge?
hi all,
once agian Im looking to·buy some more stuff for my hobby. the thing is I have no experance at all with DC motors. so I could use some help. the problem I have had in the past is not asking the experts before I buy something so I end up with something that doesnt work the way I wanted it to. so.. heres the specs or goals..
1) 2 DC Motor drive system, Im thinking about 6 inch wheels or so. tracks are cool but there are just so many problems with them at the speed I want to go······ plus they· are· usually· expencive
2) 1 or 2 H bridge's , or some other bridge if theres a good reason
3) I want the bot to be able to move as fast as a normal humans walking speed
4) the platform will be about 12 to 14 inches wide·and 18 to 20 inches long
5) the total payload I want it to carry is 12 to 14 lbs. (total meaning bot + payload)
6) course money is always tight, so something cheap and good quality
7) the bot will be going outside as well as indoors.
right now im planing on a 12 volt drive system and I will be using a bs2 to start with.
I could use advice for the wheels, motors and bridge. that being said I dont need something that looks like its going to run in the baja 1000.
thanks for the time·
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Muahdib
IC layout designer
Phoenix Arizona
once agian Im looking to·buy some more stuff for my hobby. the thing is I have no experance at all with DC motors. so I could use some help. the problem I have had in the past is not asking the experts before I buy something so I end up with something that doesnt work the way I wanted it to. so.. heres the specs or goals..
1) 2 DC Motor drive system, Im thinking about 6 inch wheels or so. tracks are cool but there are just so many problems with them at the speed I want to go······ plus they· are· usually· expencive
2) 1 or 2 H bridge's , or some other bridge if theres a good reason
3) I want the bot to be able to move as fast as a normal humans walking speed
4) the platform will be about 12 to 14 inches wide·and 18 to 20 inches long
5) the total payload I want it to carry is 12 to 14 lbs. (total meaning bot + payload)
6) course money is always tight, so something cheap and good quality
7) the bot will be going outside as well as indoors.
right now im planing on a 12 volt drive system and I will be using a bs2 to start with.
I could use advice for the wheels, motors and bridge. that being said I dont need something that looks like its going to run in the baja 1000.
thanks for the time·
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Muahdib
IC layout designer
Phoenix Arizona
Comments
· I can’t recommend motors, but if you are dealing with a 12V system with a stall current no greater than 25A then the HB-25 would be a great choice for a motor controller. It has an H-Bridge and PWM circuitry built-in and operates with the simplicity of a servo by using servo compatible pulses. I hope this helps. Take care.
http://www.parallax.com/detail.asp?product_id=29144
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Chris Savage
Parallax Tech Support
My recommendation goes for the H-Bridge IC.... I am currently using the LM18201 with great results... this is a single chip that can drive dc motors at up to 3 amps... the logic of the IC is adaptable... it has a "Brake Pin"... "Direction Pin" and "PWM Pin"...· but.... it also has a feature that allows you to control the Speed and·Direction with a single pin... for that to work you will have to drive this pin·with a Pulse Width Modulated PWM· type of signal.... so... a signal with a 50% duty cycle·means no movement.. but variations of the dyte cycle percentage·means movement in one direction or another (lets say 49% or less is forward and 51% or more is reverse)· so.. the lower or the higher the Duty Cycle percentage goes.. the slower or faster the motor moves... it is important to mention that this pulse is not servo type of pulse.... you will have to play with your code in order to obtain the correct frecuency and duty cycles..... I love this feature because simplifies the connections, the number of microcontroller pins used for controlling the motors... and it also simplifies the code... I am using it to control a couple gearhead motors and in 50% duty cycle it actually serves as a brake... so... it looks like you can control speed, direction and braking with one pin.. neat... (I am not DC motor expert, but Pittman Gearhead·motors are high quality, and if you are patient enough you can get them on ebay for cheap, I've got two Series 9000· for $18 each about a month ago)
Another nice feature of the LM18201 is the fact that if you need to drive a motor over 3 amps... you can actually connect two or more in parallel to double or triple the Amp rating, but you need to follow certain simple rules in the PCB design, here is more info about this possibly less known feature of the LM18200/LM18201 IC:
http://search.national.com/iphrase/query?query=lm18201%20parallel
If your power needs are above this... you should consider Parallax's HB-25
...· also.. in its most simple configuration it actually requires only a ceramic capacitor for Vcc bypassing (very important)·that's it.. no sense resistors,·no charge pump capacitors and diodes·(well, only if you are planning to switch the motor·at frecuencies above 1Khz... but even then it is just a couple ceramic capacitors for the charge pump, no diodes required)....· the LM18201 doesn't provide a easy way to measure Amps through a sense resistor... if you require this option you should use the LM18200· instead...
I hope this helps..· Happy Roboting!!
Post Edited (Joe "Bot" Red) : 1/19/2007 8:31:28 PM GMT
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Muahdib
IC layout designer
Austin Texas
http://www.robart.com/wheels.aspx
I particularly like the plastic wheels with the light, foamy rubber tire... but I guess it depends on your budget, the aluminum wheels look super cool.. but they are expensive.. I hope this helps.. happy roboting!!!
1 in your fellows opinion's is it better to drive wheels directly from a geared motor? or use a chain and sproket?
2 what rpm motor shoudl I be looking for for a direct wheel drive? this would be for a 4 or 5 inch wheel. and remember I want this to be able to move at about the same pase as normal walking speed.
3 same question for chain and sproket. I realized this is somewhat adjustable for overall speed of the bot.. but a starting point?
direct drive would be cheaper.. so thats the way im leaning. the down side is that the motors can get hung on things because of the posission they would have to be at.
thank again for the time
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Muahdib
IC layout designer
Austin Texas
http://www.lynxmotion.com/images/html/wheelspd.htm
Some links about wheel and chassis construction:
http://www.societyofrobots.com/mechanics_chassisconstruction.shtml#wheels
http://www.infernolab.com/projectfiles/wheels.html
You should not drive the wheels directly from the motor unless the the robot is really light... otherwise you should use a drivetrain or something similar to this: http://www.lynxmotion.com/Product.aspx?productID=229&CategoryID=6
I hope this helps!!..·
heres a dumb question.. what do you think is normal walking speed in inches per second?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Muahdib
IC layout designer
Austin Texas
at 42 inches per second you need a : 160 RPM motor
at 60 inches per second you need a: ·229 RPM motor
Please remember that the DC motor speed is very dependant on voltage, so, it is important that you make sure the voltage and motor specifications are a close match... in any case, you can choose a motor with a higher RPM and then finely control the speed via software.... one thing I like about the Pittman motors, besides quality, is the fact that they have a web page with all the technical datasheets, including, size, gearhead ratios, voltage to RPM relationships, etc. Here:
Main Page:··················· ·http://www.pennmotion.com/quick_index.html
Technical Datasheets:···· http://www.pennmotion.com/pdf/lcg_bulletin.pdf
BTW, I am going to start working on exactly the same idea as yours, I need to assemble the motors and wheels, but is going to take some time since it is not my main project, in my case, I am going to try to build the drivetrain myself with direct wheel drive through a homemade drivetrain, I've got a couple new ball bearings and I am thinking on using short threaded rods as the shaft/transmission between the wheels and the motor, I will keep you posted with any updates...
BTW.. what is the purpose of your project?... mine is a three wheled robot which soooomeday can possibly become a Balancing robot...· :-)
Post Edited (Joe "Bot" Red) : 1/24/2007 3:04:19 PM GMT
I was going to guess 30 inches per second because that is the step you learn to take in basic training for the army/airforce/marines. the navy for some reason marches to a 28 inch step. dont know why.
I want to build a platform that I can do many thing with. while I really like my bs2 and boebot.. its just to small and really doesnt have enough bandwidth to do larger things, like video, speach, moving things that are larger then a few onces.. boebot is a good learning tool and test bed, but it's limited hourse power is making me build something with a real pc onboard. so, to that end I got one of the new AMD geode "all in one boards" for about 90 bucks, I have some old sdram laying around and and old laptop hard drive. the footprint is about 9 inches x 9 inches and the cpu only puts off 15 watts. soo. i starting thinking aobut building a bot around it..
there is really nothing out there for a reasonable price that has this much computing power.
I will most likely still use the bs2's( I have 2 of them) to do something onboard.
so anyway, thats what im starting out to do, if I find a combonation of hardware that works well together, I will post it here so that people hopfully learn from my mistakes
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Muahdib
IC layout designer
Austin Texas
You posed the question:
"I was going to guess 30 inches per second because that is the step you learn to take in basic training for the army/airforce/marines. the navy for some reason marches to a 28 inch step. dont know why. "
My guess is that if you asked a Navy man about that stride difference, he'd probably say that "It's more difficult to walk on water." <ducking>
Regards,
Bruce Bates
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
<!--StartFragment -->
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Muahdib
IC layout designer
Austin Texas
the pittman motors that were suggested above (Series 9000 ) I cant seem to find in anything but a 24v version. also.. will 2 of them be able to move 12 lbs?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Muahdib
IC layout designer
Austin Texas
http://www.lynxmotion.com/Product.aspx?productID=93&CategoryID=11
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Muahdib
IC layout designer
Austin Texas
http://www.linengineering.com/site/resources/onlinecalculator.htm
Be careful with the direct wheel drive idea, the gearhead motors should never go over their peak torque load, otherwise the gearhead assembly can destroy itself (well, this·can only happen if·your batteries can provide enough mojo to the motors to go over the peak torque and the robot gets stuck and cannot move the wheels)... ·I am thinking for my drivetrain to have the motor shaft attached to the drivetrain through some kind of strong plastic round piece, if the robot gets stuck, the force of the gearhead motor will grind the plastic way before the gearhead gets self-destructed.
If you are not in a hurry to get the motors.. I suggest you take the time to do some search... like I said.. I've got Pittman Series 9000 (model 9236S018) for $18 on ebay... if you get lucky you can do the same..
I went through the Parallax HB-25's Manual...· this is a really nice motor controller!!.. it can handle a lot of power and it is really flexible.. they also have an offer right now:· Two· HB-25s· for $79.95,· it is normally around $100 for two of those... nice.....
Talking about robotics.. here is my current project!!!:
http://forums.parallax.com/forums/default.aspx?f=25&m=161756&g=162095#m162095
It is a work in progress, just today I finished adding a I/O Port Expander, it is a MAX7300, it is a I2C bus controlled IC.. my goal is to use it to sense switches at the tip of the legs in order to determine when the leg is touching the ground...
Well.. I hope this helps and keep me posted!... C-ya
Post Edited (Joe "Bot" Red) : 1/25/2007 3:35:23 PM GMT
that bot is cool. I have seem them before but have never known anyone that built one. do you think you can make it go up and down stairs? I know you would have to modify some things. that would be cool tho. good to know that lynxmotion is good stuff.
about the HB-25's. dont you think these are overkill for what Im wanting to do? I dont know, but 25 or 30 amps. that seems like a lot to when the motors will only be pulling 2 to 4 at the most. I was under the impression that using something like that would make the bot harder to control at lower speeds? they are also HUDGE.. I guess I will have to go read the whole data sheet[noparse]:)[/noparse]
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Muahdib
IC layout designer
Austin Texas
again joe, thanks for your time an input. Im really like your advice
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Muahdib
IC layout designer
Austin Texas
Post Edited (Rontopia) : 1/25/2007 9:24:52 PM GMT
About the Hexapod going uptairs... you actually read my mind!.. that has been my goal for some time... but I needed to have the switches at the bottom of the toes working... so..I got the MAX7300 working.. and now I can read the switches... so much fun, so little time...
Yeah, the HB-25 is overkill, but I got excited about the offer, well, they might be big for a small robot, but I really like their features, 25 amps is a lot... but as you progress in your robot idea, I am certain one day you will require more amps, and well, you will be already familiar with the HB-25 and it will be fairly easy to upgrade, and you won't have to spend more money on a more powerful motor controller... just a tought.. anyway, you can go cheaper with an H-Bridge design, but it can be frustrating at times to get it working, not to mention that if you don't have a logic circuit in front of the H-Bridge to control it.. you can potentially shortcircuit one side of the bridge, and boom!! burnt batteries/circuits/robot, also, I simple H-Bridge (just the MOSFETs) requires 4 pins per motor to control it, and that translates to 8 pins for a two wheeled motor, that is a lot, specially when you compare against the LM18201 2 Pin or the HB-25 1 pin solution......... you have the LM18201 alternative as well, but remember that it won't handle more than 3 amps, unless you parallel it with another LM18201, then you can handle 6 amps, and so on... but I think that sometimes when we take that road of experimenting with alternatives, at the end we spend more money in failed components and ideas, not to mention the waste of time!!.. sometimes is better just to get the finished product, the HB-25 is quite a beautiful design, you can even daisy-chain two HB-25 and control both motors with one pin!!!.... sweet!
You said that you were under the impression that a stronger motor controller will make harder to control the motors at lower speeds, well, the lower speed control is really related to your code and how fast your microcontroller is, if your microcontroller cannot react fast enough to the sensors and generate the pulses for the H-bridge, then that would be the problem, remember that the DC motor Amps comsumption is directly proportional to the voltage applied... it doesn't really matter if the motor controller can provide more amps... I wonder if this was your concern, or the harder to control concern was about the type of input the Hb-25 requires (PWM type of pulse)
BTW, which one is your preferred microcontroller??..· my choice is the Parallax Propeller, tons of power, pins, and the Spin/Assembly languages are both powerful and beautiful...
Back to work... ughhh.... keep me posted!.. c-ya!!
Post Edited (Joe "Bot" Red) : 1/25/2007 9:34:29 PM GMT
your right the HB is the way to go.
right now I have 2 BS2's, I was just looking at the Propeller a little while ago. I dont know much about it and I know even less about Assembly. I think I would rather program from a higher level.
but remember that what Im really wanting to builed is a PC based bot. im sure it will be awhile till i understand all the programing involved but I started studying C# a few months ago and am going to a community college in a few weeks to take some more C# and maybe some Java..
have you ever used a pc to talk to a propeller? other then uploading your program? im wondering if that would be hard to do? I could use the propeller to run the motors and sensors and report to the pc when there is an event. I can use the PC for video and internet access, as well as a host of other things. ever done anything like that? is it possible to pair a propellor and a PC? or even a BS2 and a PC? hmm I understand that is a silly question the way it is phrased there, but you know what I mean? how do you make a propeller talk in real time to a PC?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Muahdib
IC layout designer
Austin Texas
Glad to hear about the HB-25, I believe is a really good choice... About the AMD geode "All in one" boards, interesting approach, I thought about the Mini-ITX approach before, but they are not cheap, and definitely overkill for many robotic applications, tell me more about your particular AMD geode Board...
About talking from the Propeller to the PC and back, it is super simple, the great guys at Parallax come up with all sorts of Libraries for the Propeller, and there is one that allows you full duplex serial communication, I am using it for debugging right now at 115200 bps.... now, to achieve wireless communication I am using a Radiotronix Wi232DTS module, it can transmit at 115200 bps, pretty cool, there are some other options in the market, like ZigBee that are very flexible and powerfull too... my Hexapod project intent is also to develop a Visual Basic based PC main program that talks wirelessly to the Hexapod....
Back to your project... what did you decide on your motors???... look at this!!!
http://cgi.ebay.com/2-NEW-PITTMAN-DC-GEAR-MOTOR-GM9212C594-250RPM-15vdc_W0QQitemZ330079938486QQihZ014QQcategoryZ71400QQrdZ1QQcmdZViewItem
These are two Pittman 9000 series 15.1 Volts running at 250 RPM!!.. and you can buy both now for $25 it cant get better than that!!..
uhmm... those motors are tempting!!.. :-)....···
I found these on Ebay Express
http://item.express.ebay.com/__Motion-Control-Drives-Motors_PITTMAN-500-RPM-24V-DC-ROBOTIC-GEAR-MOTOR_W0QQitemZ170065917134QQihZ007QQptdnZMotionQ20ControlQ2cQ20DrivesQ20Q26Q20MotorsQQcmdZExpressItem
http://item.express.ebay.com/__Motion-Control-Drives-Motors_Pittman-Gearmotor-DC-12-v-motor-Robotic-Hobby_W0QQitemZ130064010952QQihZ003QQptdnZMotionQ20ControlQ2cQ20DrivesQ20Q26Q20MotorsQQcmdZExpressItem
These are also very good choices..
ok fella... keep me posted...·
Post Edited (Joe "Bot" Red) : 1/26/2007 4:10:21 PM GMT
heres the geodo link
http://www.amd.com/us-en/ConnectivitySolutions/ProductInformation/0,,50_2330_9863_10837,00.html
I dont see the one I bought at newegg anymore, but im sure there are a few out there. its not a mini itx form your looking for its a mini atx( 9x9 inches or so) i think is what its called. they are putitng the geodo in the mini itx form factor more now heres a link to one of those
http://www.newegg.com/Product/Product.asp?Item=N82E16813153052
as far as overkill in processing power. I know that most people in robots are looking at or building bots that use some kind of wireless (blue tooth or what have you) link to a "base pc" and having the base do a lot of the work. there are many reasons for this like saving power, saving money, saving size. I also know this what Mr. Microsoft believes is the future, which fits in really well with his buisness model, hmm isnt that nice? I agree with him that robotics is the next Microsoft or the next home pc.. but I dont agree with the idea that computing power will be offloaded to a base pc. I think the future is not going to be offloading data but rather monitoring data. at that, monitoring it from a distance over the web or through some other network. to do that, I think one needs to have more computing power then the current micro controler out on the market today.
really the next big step in robotics is going to be low cost vision. its actually what i do at work right now( im a computer chip designer).. I do r&d for a company here in austin that does work with scaning lidar or ladar. in the next few years, we will see a lot of advances in how robots will be able to "see" the world. once that happens you will need more then a micro controler to process all that data. thats why I chose the pc onboard approach, so that when we have a prototype in a year or two, I might actually know what im doing with my platform and be able to "plug" in this new vision system. if not, I still have a really cool bot[noparse];)[/noparse]
so.. back to what Im working on or getting ready to work on. the reason I want a pc onboard, is so that I can log into it form anywhere. I can use its sensors and even a simple webcam. as I learn more about programing(still a novist in oop) I have a lot more resources onboard to play with. easy to manage, easy to change lost of power. the pros far outweigh the cons IMHO[noparse]:)[/noparse]
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Muahdib
IC layout designer
Austin Texas
http://www.quixant.co.uk/hardware.php
http://www.axiontech.com/prdt.php?src=FG&item=75735
http://www.shop.com/op/~Biostar_M7VIG_400_Socket_A_mATX_MB_w_AMD_Geode_NX_1750_CPU-prod-37381640-49882062?sourceid=298
you know what.. I just did some reading and I guess this processor is a lot more powerful than I thought. the geodo NX I have is a real 1ghz cpu. cool
http://en.wikipedia.org/wiki/Geode_(processor)#Linux_on_Geode
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Muahdib
IC layout designer
Austin Texas
Post Edited (Rontopia) : 1/26/2007 5:10:24 PM GMT
Thanks again...
BTW.. did you see the Pittman motors on Ebay Express? I like Ebay Express because they are trusted online shops...
Ok.. stay in touch!!.... I will try to work on my motors next week.... c-ya