next topic »

[on:]

Hi there,
I am a complete beginner with Arduino and I trying to start with my first big arduino project. I am trying to create a remote controlled robot that goes inside a pipe and checks for defects in the pipes inner coating using a holiday detector. I am only in the design stage and I have decided that the robot would need two dc motors, which would have the same control (e.g. forward, backward) , one stepper motor to control camera rotation and one small dc motor to rotate the holiday detector lead. I am thinking of having an independent power supply for all the motors. The small dc motor kind of rotates around the robot as it will be attached to a rod in the middle. Can I use the Arduino motor shield for this? I think it is possible because it says it should be able to control 2 dc motors (out of 3 as one dc motor doesnt have independent control and just copies the first one) and one stepper motor.

Also I would like to use the xbee modules for wireless transmission, preferably the ones which have a long range and a high data rate. I dont need video as such more like one frame per 2 seconds. So I am thinking of using two xbee modules onboard the robot, one to control the robot and another to send the image data and instrument data. There will also be two xbee modules at the base, one for controlling the robot and another for receving the data. I know that I need some processing power because of the image compression and transmission so I am thinking of going for the mega 2560. Should this setup work? I am pretty new and I have limited funds so I would like to make sure all my bases are somewhat safe before I dive in.

[Reply #1 on:]

most bots that go down pipes have cables attached so that they can be pulled out when something does not go right and they break or get stuck. Google for terms like "inspection robot" for previous similar projects.

[Reply #2 on:]

As zoomkat stated, there are practical advantages of having physical cables on this type of robot.  Another would be that wireless communication isn't going to be of much use inside of most metal pipes...  Usually metal pipes either come into direct contact with the actual ground some where along their length, or are attached to enough metal and other material for it to act like a ground.  A metal container electrically connected to a ground is one of the best ways to block electromagnetic waves, including radio frequencies like Xbee.

In regards to your ideas on the rest of the robot, your general approach seems to be reasonable.  However, pipes often have bends or junctions, even the striaght pieces can be vertical or at an extreme angle.  So you probably want a drive system that can traverse all these different configurations, as well as a horizontal length.  So on a related note, I also agree with zoomkat that you should do search online.  Not just for other DIY projects, but the professionally made pipe inspection robots, especially when it comes to overall structure and drive mechanisms. 

[Reply #3 on:]

Thank you for helping me out here.

Yeah, I was thinking about attaching a rope to it incase it gets stuck, by putting a pulley at the end or something similar. I will do some more research on it 'inspection robots' first. The robot will be used to enure that the coating inside the pipe is good enough. So after painting, we will put the robot in and see if it has been done right and then go on with the next one or redo the painting. So I will be using it on straight pipes and not ones installed under the ground.

I initially thought about using a wired connection as that would make things easier but I was told that after painting, the robot would be going in at a high temperature (100 celsius approx although I could just put it in when it cools down to something cooler) Also the wires could damage the new coating. Even when we use the rope, we have to make sure it doesnt damage the coating. I thought since the pipe would be on level ground, I could put maybe a xbee signal 'booster' at the pipe opening, just to make sure I get the signal.Would xbee still not work?

[Reply #4 on:]

If the pipe is on top of the ground and straight, I'd just use a setup that allowed you to push/pull the cam in the pipe via rod/pipe sections. Make a light circular frame with soft supporting pads to support the cam, and mount the cam on a rotating circular disk with the lens close to the pipe surface. push the cam thru the pipe, rotating it via the push pipe. I've used 1" 10' pvc pipe to break up clogs in sewer lines. I connected the pipe sections by glueing a coupling to one pipe end, and the sticking another pipe end into the coupling with a bolt hole thru that end of the coupling/pipe for a connecting pipe. the area was confined, so the long pipe needed to be lengthened one section at a time, and then disassembled as it was pulled out of the sewer pipe.

[Reply #5 on:]

Yeah, thats kind of what we do now. We use a mirror and sometimes a camera to examine the pipes . But the whole point of using this robot is to examine pipes that are about 300 metres to 1km long.

[Reply #6 on:]

Also sometimes a camera isnt enough to detect a problem. Thats where the holiday detector come in. right now we use a long rod with a sponge at one end and connect it to the holiday detector at the other end and then put it in. We cant do this for 300m to 1km so we thought a robot on wheel which has a camera and holiday detector would be nice.

[Reply #7 on:]

I should be careful about using a holiday detector along with an arduino. It is likely that the spark it produces will upset the processor and cause it to reset. Also it will cause severe radio interference.

[Reply #8 on:]

The present design has the holiday detector about 50cm away from the arduino. Also I am using a brush lead holiday detector that blinks an led when there is a problem. Would that help or do I need to keep it a bit further off ? I hacked the holiday detector by using the signal it gives to its led and am trying to make it a digital input.That is still a work in progress.  I also wanted to get enclosures for the motors and arduino so that there is no em interference with the xbees . Would that be enough?

[Reply #9 on:]

Would that be enough?

It is hard to say, it is one of those things that you are going to have to try, but just to give yourself an idea take an AM radio and see how far away you have to be until you can't here the spark.

[Reply #10 on:]

Thank you Mike. That should give me an idea. I will give it a try and let you know what the results are. How much of an effect would the motors and holiday detector have on the xbee though?

[Reply #11 on:]

It should not be too bad because the Xbee protocol repeates a package that has been corrupted.

[Reply #12 on:]

If electrical interference turns out to be a problem, I imagine that you could electrically isolate the control electronics from the holiday detector, and put them inside a screened enclosure. I'm not totally sold on the idea of trying to make this a self-contained system, though. If I was doing this then at least the MK 1 version would have the minimum hardware needed to detect and repair flaws on something that could be pulled through the pipe manually, with hard-wired controls. Replacing the manual pull with a tube crawler and replacing the hardwired controls with remote control can come later if necessary.

[Reply #13 on:]

My first thought, when reading the subject line, was that the Arduino was going to detect holidays, based on decorations, I guessed. I thought it might do something different if it detected Christmas decorations vs. Halloween decorations.

I see that that is not the case. learn something new every day.