« Last post by jamieyello on May 09, 2016, 06:53:54 PM »
Hi, thanks for your reply. I've decided since this is a big complicated project first things first I just want to get a feel for creating things and just working with metal.
I have started looking at Maxon motors, I think I will be getting one of those, I'm looking at the ones in the $100 range to just mess with for now.
« Last post by dbassman on May 09, 2016, 12:48:38 PM »
I'm hoping someone here might be able to help me… I'm looking for a High Torque 12V DC Motor. I've been unable to find one that meets advertised torque values.
#1. 12V DC
#2. 120 RPM or greater
#3. Body diameter approximately 25mm
#4. Shaft diameter 4mm
#5. Torque (Sustained) >4kgf-cm; (Peak) >8kgf-cm
Any help would be greatly appreciated!
« Last post by bdeuell on May 07, 2016, 10:26:05 PM »
Selecting the best motor for an application depends on many variables. If you are looking for precision dynamic control with the best weight to power ratio servo motors are probably the best choice.
Note: s servo motor is technically any motor with a feedback control loop. RC servos to fit this definition but are likely below the performance requirements you are looking for. in industrial applications a servo typically refers to a brushless motor with encoder feedback... this is more along the lines of what you need.
Maxon offers some very powerful motors for their size.
Also your mechanical construction will be key to smooth operation, any backlash will result in jerky motion.
First, thanks again for the information on where to find replacement screws for the RAD series. I have a load more questions for this project so I thought starting a build log thread would be better than posting them all seperately. This is just about the most complex project I've ever started so any advice, suggestions or ideas would be much appreciated.
Patrick started off as a RAD v1.0 robot found at a car boot with no battery or controller. Some quick testing showed all the motors are still working fine, so my plan is to replace the head and driver board with a completely custom made unit, and add an 11.1v lipo battery for power. The head is partially based on an earlier project, a sensor-equipped timelapse camera, but this is intended to be more of a "do-anything robot", so the list of gadgets includes the following...
Raspberry Pi 2
Arduino Mega 2560
NoIR camera fitted with a selectable IR-cut filter and CS-mount socket.
TCS34725 RGB sensor
PIR motion sensors
MLX90614 non-contact thermometer
SR-04 sonar ping sensor
DHT-22 temperature and humidity sensor
TGS2600 air quality sensor
mosfet based electric charge detector
Adafruit ultimate GPS reciever
IR LED illuminator
12x neopixel ring light
5 inch HDMI display
USB WiFi adapter
7 port USB hub
The speakers should be able to play music for a group of people outdoors. I have a pair of 3 inch, 4 ohm, 5 watt speakers that are powerfull enough, and was going to wire them to an Adafruit 20w d-class amplifier board connected to the Raspberry Pi's audio out, the I2C bus for volume control, and the direct 11.1v feed from the LiPo for power. Is there a better way to handle the audio in terms of power to weight & size than this? All the 20w speakers I've found are too big for the head, but if I drive the 5w speakers, will I just have to software limit the volume to stop them blowing? I don't have the amplifier board yet but was going to go with it based on it's size and efficiency, and the fact that it can be put into standby mode when not in use.
2) Power management
I'm trying to make the 5AH battery last as long as possible by giving the Raspberry Pi the ability to turn off anything that's not in use, but I'm really not sure how to go about this. I was going to put a latching relay between the 5v output of the BEC and the input of the USB hub, controlled by 2 GPIO pins of the Pi (the arduino is powered by the hub, so this would shut down most of the gadgets for a soft sleep mode). The HDMI screen backlight is another large drain though, it has a tiny toggle switch for controlling the backlight that I might be able to solder to an optocoupler, or I could use another relay to control the 5v feed to it. The power supply so far uses 2 buck circuits, a BEC, and a MoPi board, so I'd like to avoid adding too much more complexity but I don't want it to waste too much power either. Any suggestions?
Those are the main points I'm having difficulty with so far. At the moment I'm just trying to design and print a stack of mounting plates to hold all the boards inside the head. Anyway, thanks in advance for any advice and I'll post back when there's some progress.
« Last post by jamieyello on May 06, 2016, 07:35:34 PM »
Hi, I'm looking to get into my first robotics project, I'm fluent in BASIC and less so in c++, I want to be able to control a robotic arm that will mimic my own with as little latency as possible, along with a couple other functions.
I want to run a portion of the software on windows, where the position of my arm among other data will be determined (I will have kinect sensors along with other motion tracking peripherals), that will be sent over to a raspberry pi, which will serve as the motherboard for the robot. Two cameras will be set up to the pi, the pi will send the footage over a video transmitter like this onehttp://www.supercircuits.com/high-powered-weatherproof-2-4-ghz-wireless-transmitter-receiver-dvl24wr
(It looks like it wasn't designed to work without a line of sight, but it also seems to have very little latency)
The footage I will combine into video that can be sent to an HMD such as Oculus Rift. The Oculus will supply camera position as well (to be sent to the Pi).
The robot arms, I will not be going the traditional route with. Normally when someone would think to build a robotic arm they would go, "Ok, servo motor at the elbow and other servos wherever the arm needs to be turned." However I feel like that's not as efficient as it could be, what I want to do is replicate the way a human arm works, and hopefully replicate that same naturally designed efficiency, doing this by copying the leverage obtained by muscles and tendons using wires. I also want it to be fluid like a human arm would be, when I see robot arms they're always jerking around, usually barely keeping up. I feel like work could be done on the software end to take into account the inertia and smooth things out. I also feel like that's the work of servo motors.
So you get where I'm going? Tl;Dr, robot avatar with single arm. Future additions will include legs a body head and if I'm feeling ambitious maybe another arm.
So bottom line, the software, I'm fine with. Hardware? I'm still ok. I've never done any soldering, no engineering of any kind. DC motors vs Servo motors, do servos require more power? Are they less efficient? Less torque? I can't find any solid answers to any of these questions. What kind of materials and tools will I need to create the arm? A saw of some kind? It might seem like I'm not knowledgable enough to take on an ambitious project, but then I am self taught in most things I know. If nothing else I hope I amused you with my ramblings.
One more question, what is the strongest motor available to the public around the size of a bottle, does anyone know? I've wanted to know that since I saw Jamie from mythbusters use the supposed motor to make a grappling gun.
I realize this is an old topic, but I'm new and I can't hold this in…
Besides I, Robot
being my #1 favorite, my next two favorite robot movies are CHAPPiE
and Big Hero 6
. Right now these three are at the top of my personal list. Of course I have more, but to avoid writing too much here, I can share my other favorites and my thoughts on those later...
But, I can't leave out The Iron Giant
has a special place in my heart.
« Last post by bdeuell on May 04, 2016, 07:57:33 PM »
It's almost as funny as the Car-Drive thru prank.
Lmao, even the most basic puns are hilarious coming from a robot
« Last post by jjsylvestre on May 03, 2016, 07:18:27 PM »
I have a question in regards to using P-channel MOSFETs as a switch to power a relay coil. I am trying to implement this sequential circuit below. When this circuit is first turned on, 'fault' is 1 (indicating everything is good) and current will flow up until the second MOSFET. When you press the set/reset button, the relay latches on and will stay latched until the fault is 0. Once fault returns to 1 0, the relay won't be powered until you press the reset button again.
My question is... will a P-Channel MOSFET work as the switching device here or should I use something else like a solid state signal relay? I have only used MOSFETs when the source is at a constant voltage so I'm not sure if it will affect the MOSFETs behavior if the source is changing.
Note: The indicator light circuitry is also whats in the schematic. That shouldn't affect the circuitry of concern.
« Last post by munim on May 03, 2016, 01:27:30 PM »
there is some information in the tutorials section regarding calculating motor torque. search the posts and web as well.
a couple notes:
- what units is your speed in?
- you will probably need a gearbox (or some other speed reduction)
- what acceleration do you need?
- what surface is it driving over?
- does it need to go up any incline if so how steep?
- how much will the robot weigh (i'm assuming the weight provided is for the payload)?
speed is in km/hrs.
surface is ceramics/home tiles/wood floor.
i think i will be able to control acceleration with my aurdino.
not using robot for inclined surfaces yet.
total weight will be 10 kg of robot and the payload
Thanks for replying appreciate ur concern.