Problems and inspiration.

Today I went to a FRC. It was fun, and I joined a team only 3 weeks earlier, and hopefully things will turn out good. So far they want me to be their lead programmer, and it sounds like fun. Today’s match was a ton of fun and it is a great place to meet people that love robotics. I think these robots are more fun than your typical robotics, and it beats the FLL by 100 times better. These robots make your Lego NXT sets look like toys [well, lego’s are toys aren’t they?]

They seem to have a lot of fun and it included having the best food I’ve had at a robotics event – a group was fundraising for their robotics team and was selling these great burgers that were amazing [partially because they were cooked on a propane stove] and it was just a fun experience.

 

On the bad side, I found out that port A on my newer NXT no longer works – the motor output is fried. The rotation sensor seems to work OK though… Probaly something I should report to Lego.

I could probably take it apart and find the problem, but there’s no point. I’m probably better off making it into a full rotation port and build a rotation sensor so that I can do it without a huge motor. I looked up the specs, and it’s a simple trigger system

1 degree equals this en-emulated into the two signal wires:

0 0

0 1

1 1

1 0

And opposite order for the other direction.

I tried making one with light resistors, but I can’t get it to work. Probably better off using them as light sensors. I feel that light sensors reflecting off an object is more accurate at telling distance than the Ultrasonic, which works best in port 4 than any other [partially by the raised transfer speed/communication]

I’m thinking I’ll work some more on the printer.. make a skid steer out of the other robot for now, at least it only takes 2 motors, and I could use the prototype board to make a third motor port…

 

There were a lot of cool ideas at the FRC and am considering what to build out of the ideas I saw there,n I could really improve my robot with their ideas.

Advertisements

Printer update! Surprise Enclosed!

OK. So a friend of mine and I got together last weekend and we totally rebuilt the printer. It’s design is very similar, and I’m sure we used many the same parts, but now we have a more functional robot. While it is practically impossible to get it to feed paper into it correctly, we finally got it to work. We gave up dots and complex programs and it paid off. We can draw lines on the paper now instead of lousy dots that the stupid ink-jets that take three hours to print. We can only draw one line at a time, but it works much nicer. Found the camera so pics to follow.

Other project:

Centrifugal Force Air Compressor:

This is a cool project. I was actually trying to make my own GluideWheel or whatever it’s called to connect PF motors as if they were NXT servos with all the rotation sensor stuff. A great idea, but I wanted to build one first, since I have the equip.

Really its not that hard to make a lego NXT motor port rotation sensor. All it is is a group of LED’s and light reactive switches [whatever you call those useful things] and a correct hookup. I hooked up the PF to a NXT port and found that the 2 amps out of the NXT was much better than the 200mA of the battery box, and decided that this motor is practically unstoppable, which when connected to a NXT port, it’s practically that. So I decided to gain centrifugal force and use it to run an air pump geared up from the normal motor, and it worked fine, in fact it went all the way to 47 PSI before it couldn’t kep up with the leaks, and just stayed even. But under the regular battery box hooked up the way it is it would stop at about 10 PSI, hardly enough to move a small piston.

This was a great idea that works amazingly. More tomorrow.

Preview for Solar Battery Backup Device using Lego Power Functions

There will be photos and video of the project coming soon, but I can tell what I have so far.

I’m taking a shoe-box and converting it into a battery backup. Although I’m only using a power functions battery box filled with 6 rechargeable 2400mAh batteries, creating a total of 7.2v and 14400mAh, it’s only powering LED’s, which use about [according to my research] 10-20mAh. That would amount to [with 20mAh] about 720 hours of light in one charge, with one LED. If I ran LED’s for 8 hours [overnight], then I could run at least 30 LED’s. That would be plenty of light for a room or two… and during the day I could charge them with the solar panel, and use a small power regulator, maybe the NXT, but could be as simple as running a motor while charging in the sun to keep the rechargeable batteries from overcharging. I could run the NXT all night easily on a rechargeable battery pack that I bought for it from Lego Education [a site full of bulk Lego’s!], but I could only trust the power pack and a wall socket with the best power-strip with all sorts of protection for that.

The great thing about Power Functions is the plug. It has 4 wires, as follows:

positive [0v]

variable [the direction of the swich on the battery box or the separate switch uses this to run the motors and lights]

variable [reverse charge of the one directly above for return voltage]

negative [full battery charge, for me, its 7.2v]

Using the two outer ones, I can charge the battery without impairing the ability to run the motor in either direction. This is really useful, although I don’t like the design for the external switch because it doesn’t work independently of the battery pack to send a voltage to the motor or led’s – only the battery pack can use the outer wires, and the motor and switch can only use the inner ones [but the switch, regardless of position, will always send voltage through it on the outer wires. I assume this was used for the Infrared receiver and other parts to work with it, independent of the position of the switches. The good thing is that when you switch the battery pack to the off position, it shuts off all the wires, including the outer wires. This in some places may not be extremely useful, but the use is good here because I can control it better from the external switch anyways… If you really didn’t like it, I’m sure it’s not too hard to take apart the battery box and edit that problem, but I like it in it’s “vanilla” [original] form.

I hope to have pictures out soon, maybe a video too.

power functions port wiring by jared

Printer Project

This is an ongoing project, which has been in development for many months.

Original Air Date: April 3, 2012

As seen on my Google page,

I wanted a robot that could print. This project is still in progress, and is over 3 months old. The printer works OK, but I have had it working better before. I will be returning to the project as soon as some other smaller projects and limits have been removed.


This is the best video of it so far, although its still in beta [really alpha with all the help it needed] and right now its in beta. Final version will have no arrays unless I can make it dependable…




The robot right now is built and meant to do a black and white image of an object. I place a marker in the pen slot, and then feed some paper into the print spooler. After that i open the program on my computer, design the image via hundreds [or soon to be hundreds] of little check boxes. You then click print. After this, it goes into a few algorithms. One, it prints the file to the paper, measuring the exact distances from one place to the next, and when it’s supposed to print a dot, it lowers the pen assembly onto the paper making a dot. it can take anywhere from 5 minutes to half an hour to print, depending on the size of the paper. While it does this, the computer receives real-time data of:
  • X position of carriage on track, each number is a pixel, with up to 20 pixels wide
  • Y position of the print spooler, a rotating drum to push a sheet through the press
  • Up or down position of the pen
  • Sensor data, mainly the data from the sensor on the end of the track for the X axis, to stop the pen assembly from falling off the edge and a calibration device.

The physical printer is in perfect condition, but the program has many bugs. I need to edit how the printer is fed information. Right now the data buffer is through an undependable array, which will have to be changed.