08/27/2018 at 15:33 •
Okay so for the first real demo in action we have the shirt wired up with vibration motors to alert the wearer of a fire coming up from behind them. Here is the video.
08/27/2018 at 13:23 •
Now that the system is build and the vibration motors are all working well, I have added serial control so that I can have an external device tell the arduino when to activate the motors. For this prototype I am going to add a Raspberry Pi Zero W which will communicate with the Arduino via serial and give it commands of when to activate the motors. My arduino is setup so when it gets an "R" , "C" or "L" on the serial port it will activate the right center or left vibration motor respectively. The arduino code is uploaded. A python example is shown below for controlling it from my windows machine.
import serial s = serial.Serial('COM4') s.write('R') #Activate Right Motor s.write('L') #Activate Left motor s.write('C') #Activate Center Motor s.write('RRCCLLLLCCRRCCLLLLCCRR') #Back Massage s.write('0') # Off
08/27/2018 at 13:14 •
Okay so here is the first test of the wearable system. This system has 3 peltier heater/coolers and 3 vibration motors. They are located on each shoulder and at the base of the neck. For this test it cycles through each motor and then finally the center peltier module is put into cooling for a second and then heating for a second. The code for this test is uploaded on the project page.
Since the peltier modules are in pockets, one second of heating and cooling is not obvious to the user. At this point, I am not sure if these will work for a user interface. Unlike the motors which are easily detected to be on and off in a fraction of a second, it takes some time for the peltiers to heat up and cool down. That time is even longer with a little fabric between the user and the module. Another issue with the peltier modules is that they will work much better when I give them a good heatsink on the back. For police, military, gamers and drivers having heatsinks on your shoulders is weird but seems do-able. For firefighters having a heatsink into your insulated gear doesn't help much and having a heatsink into a fire environment would be a really bad idea.
08/26/2018 at 04:25 •
Okay, so I got the first (very rough) version working. To control the Peltier heating/cooling modules I ended up going with a Arduino Mega 2560 that provides PWM to an Adafruit Motor control shield. Since peltier modules should have a constant DC current instead of PWM, I added some 47 uF capacitors to help smooth out the PWM signals. Next week, I'll take them back to the lab at school to check to see how much good the capacitors are actually doing. The motor control shield is handy because it controls 4 motors and does forward and reverse. For the peltier modules forward and reverse switch which side is hot and which side is cold.
Since the vibration motors draw about 100 mA at 5V, they can't be driven directly by the Arduino either. So I built a little board with 5 transistors on it that are used to switch on and off the vibration motors.
The whole apparatus is powered by two 18650 li-ion cells that I salvaged from an old laptop battery. A shot of the whole setup without the attachments to the shirt is shown below:
To attach them to the shirt, I hot glued pockets on the inside of the shirt for the peltier modules and vibration motors. I also cut little holes for the wires to exit.
I added straps to hold the Arduino on and a pocket to hold the batteries on the back of the shirt. When finally put together the shirt looks like this:
08/23/2018 at 02:58 •
Okay so for control and power here is what I am looking at:
For the vibration motors they will take between 2 and 5V with higher voltages causing more vibration. About 3 to 4 volts will work for my purposes. This will draw about 100 mA.
For the Peltier coolers, from my experimentation it seems like 5V @ 1.5 A will be sufficient. However, they will need to be controlled so that they provide steady cooling and then heating when needed for alerts. I am thinking it will not be as simple as turning it on and off or even using PWM. Based on the discussion hereit seems like I don't want to have it turn on and off using a thermostat like device because that is inefficient and causes thermal cycling. They also say that PWM signals will make it work less efficiently.
For the first version of control, I am thinking of using my Arduino along with the adafruit motorshield to control the peltier coolers. I will add in some capacitors to build a simple LRC circult to smooth out the PWM signals. As for the on/off vibration motor control, I'll just use some digital pins attached to transistors for switching.
08/22/2018 at 18:03 •
So now that I have all the parts, I have begun to experiment with them to see how they work by themselves. First I am going to have a look at the Peltier module. Peltier modules use the thermoelectric effect to produce a temperature gradient when given an applied voltage and can also produce a voltage when given a temperature gradient. For more info on how they work and a description of what their naming convention is, you can read this.
My module is a TEC1-12705 which is 40mm x 40mm. I first tried powering it with a Li-Ion cell, giving it about 4V. Later I tried using a 5V power supply that I have. All of the following results are with a 5V power source and drawing about 1 to 2 amps. For more details see the datasheet.
When holding the peltier, you can feel one side getting hot and the other side getting cold just few seconds after turning it on. However to really make it work, you need to install a heatsink so that the hot side doesn't become really hot and let heat back into the cold side. For my first heat sink I used a giant aluminum piece of scrap that we have sitting around the lab.
I originally just placed the module on the aluminum, but this didn't work very well. Then I applied a small amount of thermal paste and that didn't work much better. I applied a generous portion to coat the whole hot side of the module and that worked really really well. Here you can see that the hot side is 29 C and the cold side is -4.5 C. The cold side is below freezing while the hot side is only slightly warm! The aluminum block is doing a great job pulling away all the heat.
I also realized that you can just reverse the voltage to cause the peltier to switch which side is hot and which is cold. This is really great for my application and it means that I don't need to build in a separate heater (the conductive fabric) anymore. I can use the same peltier modules to produce hot and cold.
08/15/2018 at 16:54 •
Starting out I have a basic concept and the major components out and ready to play. Here I'll describe the core components that I am using. Moving forward, I am thinking about what electronics will be needed to control each output so that they can reliably send a signal to the user.
EeonTex High-Conductivity Heater Fabric - NW170-PI-20
I also purchased this from Adafruit, I am not 100% sure about what the model number is or the specs on these motors. You could get a controller for these and make them do all sorts of things. At this point, I will start off with just turning them on and off. The product page says they take from 2 to 5V. Increasing voltage causing greater vibration. I tested them out with about 4V and found the vibration to be satisfactory.
12705 Peltier Module
These devices are much more affordable on ebay and even cheaper if you can wait for them to be shipped from Asia. There are a couple of different versions on ebay with slightly different dimensions. The datasheet can be found online. For an initial test, I powered it with 4V and noticed that one side got noticeably cool and the other noticeably warm. These modules should be good for up to 14 V (room temperature) or 16V (when 50C hot). I haven't tried them to extensively yet, because I have not yet set up a heatsink for the hot side.