Recent Posts

Jaime, Please provide more details.  If you accept, then you and I will work closely to create the GUI that controls this robot.  I have experience with python to create multi-threaded client/server style GUI's, but never have done live video streams.

What language did you use?
How does the robot receive/transmit commands? Please be very specific.

So we are on the same page.  I was thinking that RACO GUI should be developed with:

• multi-threading
• multi-client/server connections
• UDP thread for video
• TCP thread for commands
• all on different ports. duh
• simple serial communication, possibly UART cause it's simple
• MAYBE bluetooth module to transmit commands through my phone that has good web connection
• initial login screen that makes the password part of the transmitted packet
• this packet is parsed at RACO's end and if the password/key does not match then trash the packet
• video display
• sensor/motor display
• if electrode sensor sleeve is being used to control the robot then display electrode sensor status
• Please post more ideas!

Explanation of protocol: So we need TCP packet transmission for the commands. Why? Because we need it to be reliable!(remember little kids. lol) Since we do not want to receive garbage on the receiving end of RACO and quite possibly spit out some random ASCII characters(garbled binary code), that can enable an odd arm movement and might just damage a kid or worse RACO, we will adopt TCP!

The cool thing of the GUI is that we can also receive data directly from RACO and provide us with the robot state such as battery status, axis rotation, sensor reliability, video, thresholds(electrodes) going off and the such.  This is turning out to be quite an interesting project! Thanks for contributing Jaime!

Login/Password Screen



Main Screen... this is where you will see what the robot sees. The textfield was supposed to be to write in things the speech program in my computer was going to say. Not sure if you want to use this with your robotic arm... This is also where you will be able to use keystrokes to control the robot... (not done yet...)




Settings Box

ok thanks for the advice.I will do.
what do you think about the formula pag 10. Lenth of the wire is not included. do you agree with it?

I would suggest that you use a power source most commonly found in a lab. The main reason is that a homemade hack can be very dangerous. It takes less than 1 amp to stop the human heart. So please be careful when working with electricity and always have someone else around even in a lab setting.

Exactly right now I am working on a GUI to control a telepresence robot... I'm not quite done yet...I still need to remotely access the webcam through the internet... but everything is turning out right. I could help out with that!

Thank you Drako 1011,
What do you sugest me as a power supply?does it have to be an expensive LAB power supply?
I do have another question, in the paper of the heat treatment there is formula that's help us to calculate the Current needed to achieve a desireble temperature page 10 where the temperature of the wire depends only on the wire diameter(mm) and the current(Amps). is that correct? what about the lenth of the wire?and the voltage? does not have any influence?
Thanks to clarfy the issue.
Thanks again

I've seen this approach before and it does work as long as you control the current. An AC laptop power supply is gonna give a sinusoidal current. Meaning that your current is not gonna stay constant. Without a constant current, you cannot achieve the heat required to train your nitinol.

Find a DC current that will remain at the required current when you add the nitinol to the circuit. That should work.

hi
thank you for the welcome and the advice.
But what about applying a elctric current. I foud this tutorial to how to train the nitinol.
http://www-personal.umich.edu/~btrease/share/SMA-Shape-Training-Tutorial.pdf
I have just tried once with a Power laptop AC wich is 19V and 4.75A, but it does¡n seem to work.
It semms simple and looks mucho more quickes en eficient, why do you think it does not work a priori?
I ppreiate your help.

Hi mccali, I suggest you do use an oven.  It worked for me.  If you have a harbor freight store, get a digital thermometer and use that to see if your oven (or your neighbors) can reach 500 Celsius.  I was surprised that mine reached that high temperature.  Although, the best temperature to train NiTinol wire is 540, but 500 will suffice.  Be careful because it is VERY HOT!  It started melting my thermometer casing!

I left it on the stove for 10 minutes (place only when stove has reached 500-540 Celsius)

http://www.harborfreight.com/non-contact-laser-thermometer-96451.html

Post some pics for us to see!

By the way, welcome to the forum

hi
I'm a newbie in Nitinol, I discovered this such nice material recently. I'm trying to found out about the diferent ways to train the nitnol. I see that using a electric stove it did work, is thta right? how long did you leave it in the stove? I'm trying to train a 0.02" nitinol wire , can you sugest me way to train this material ?(I do not have a super powerl hoven), what about using electricity is it possible?
thanks for you help and time