Portable Personal Heater (Rev 2)

Introduction: Portable Personal Heater (Rev 2)

Previous Instructable

This is the second revision of my first Instructable - Portable Personal Heater. That is where I looked at some characteristic data from manufacturer (http://www.carbonheater.us/). I continued to design my own heater using a piece from an old hand towel. Unfortunately, it didn't get nearly as warm as I had expected. Therefore, I am creating a new Instructable to show how I got it to work the way I want.

What's new in Rev 2?

I scrapped the exercise armband idea. I opted for an adjustable version so it can be more like a one-size-fits-all project. I also did my own testing to see how hot different lengths get and how much current is drawn when powered using a 5V supply. This is a sanity check on the assumptions I made in my Temp vs Volt/Length graph. Since I now have a working prototype, I'll also be able to test my battery life of my battery bank and show the temperature response too.

Step 1: Temperature and Current Vs Length


Before doing anything, I want to test my filament to see how the temperature and current behave using a 5V supply. Even though 5V is quite common nowadays (thanks to USB), http://www.carbonheater.us/ includes test data for 3.7V and 7.4V (based on nominal voltage for Li-Ion cells) - they don't have results for 5V though!


First, I cut a 24 inch length of carbon fiber filament. I stripped a piece of wire and wrapped it around the fiber twice. I repeated this every 4 inches and twisted the ends together so they wouldn't unravel (picture). In order to secure them in place, I sewed them on. Rather than using thread, I used a single piece from a stranded wire instead of regular thread (picture). This helps reduce the contact resistance and is recommended by the manufacturer. To keep the wires from unraveling, I tinned them with solder. See this pic for one of their recommendations.

To make it even better, you could add some conductive epoxy to cover the wire and filament. I did not do this for this experiment, but I recommend it for the final design.

Test Time!

I used a fancy datalogger to measure the voltage, current, and temperature and a power supply to provide the 5V. That is overkill since you can accomplish the same thing with an Arduino and ADCs. I have access to this datalogger at work, so that's what I used.

I connected the power supply and logged the response. Like the manufacturer did in this pic, I wrapped the filament and thermocouple between the folds of a rag towel. This will be more similar to the final application and will help the temperature to reach the final value faster. I connected the power supply across two terminals and logged the response for about 30 minutes. I kept testing like this every 4 inches up to 24 inches total.


All of my test results are summarized in this table. I also made graphs of the average current and the temperature for each length.

Step 2: Assemble the Armband

Determine the Filament Length

The battery bank I have can only output 1A maximum. The 8in filament drew 985mA which leaves very little headroom from my battery bank. It will also drain the battery faster. So the filament should be a little longer to reduce the current draw. The temperature at 12in would be okay for most applications, but I'd rather get just a little warmer. I decided to make the filament 10in which would leave me wtih a 25% safety margin since the current draw will be about 750mA.

Attach the Filament

Using the same method as the test, I wrapped some wire around the filament twice. Using a small strand of wire, I sewed the carbon fiber filament down. I marked the armband with a green dot where I wanted to center the heater. Using pins, I secured it in place. I lifted the overlap that I had previously sewed down and slid the end of the filament and wires underneath. This way they will be protected and not scratch any bare skin. I routed the wires under the overlap also to keep them out of the way. Using regular thread, I sewed the filament down by the edges. I also reinforced the wires where they came out of the armband for strain relief.

Attach the Velcro

I found a couple scrap pieces of velcro that will be perfect for holding this together. I wrapped the band around my arm and determined the position for the two pieces. This allows for the armband to fit different sizes.

Add the Loop

The idea was to pull the armband through a loop and attach to the velcro. It kind of works, but make sure your piece is long enough for that. My armband was just a little too short to work reliably. I used Google SketchUp to design a small loop and sent it to my 3D printer.

Most people don't have 3D printers, so another way to get it to work is to use an old Bic pen and some cordage. After taking the pen apart, feed the cordage through the plastic tube and then sew the cord to the towel. Or you can not worry about the loop and just wrap it and velcro it down.

Connect the Wires

I snipped the end off from an old USB cable. I added a 1A fuse in series with the +5V signal to help protect my battery bank as well as my arm from burning in case of a short circuit. The ground wires were simply soldered together. Simple!

Step 3: Operation

So here is the interesting part. Since I did my own testing, I am much more confident that this will work.

Test Setup

Since I don't want to wrap this around my arm and sit next to the data logger for a long time, I had to find a suitable alternative. I used the rest of the hand towel and wrapped it up as tightly as I could. Then I tightly wrapped the armband heater around that. I inserted a thermocouple to be touching the filament right in the middle of it. Similar to the first test, I configured the data logger to record the voltage, current, and temperature.


I also graphed the voltage, temperature, and current over time. Those aren't as interesting, so I summarized everything into this table. I was surprised to see that it got so warm and lasted for so long!

Step 4: Closing Notes


I have worn this against my bare sink for a few minutes. It does get too warm for my comfort at 54°C. But it feels very nice when wearing it over long sleeves. I would warm up faster if I could have the element directly on my skin. Plus, having long sleeves over it will add more insulation to help keep more heat where I want it. For these reasons, I want to make my next heater with a slightly longer heating element to drop the temperature. This has the added benefit of also lowering the current draw that ultimately extends the battery life.

Future Improvements

After building this armband heater, I thought of a few ways to improve it.

  • Start with a longer armband so that it can reach all the way around your arm and also loop back.
  • As I put this armband on, sometimes the loop goes crooked. Next time, I will add small holes to sew it in place and prevent it from bunching up. I might also use the Bic pen and cordage idea I mentioned earlier.
  • Change the position of the rough-side velcro so it can't scratch me.
  • This gets really warm - about 54°C. It's a little too warm for me to wear directly against my skin. Next I will try a 12 in long heater which will also help extend the battery life.

Additional Features

One way to extend the battery life even more is to pulse the heater. By having the heater run at 100%, I know it will last for 11 hours. Ultimately, I want to use this when I go on cold weather camping trips to keep me warm at night. Since I'll be asleep, this doesn't need to be on all the time. But I can pulse it on for 10 minutes and off for 10 minutes just in case I wake up too cold. By running the heater at 50%, I could double the battery life to about 22 hours! That is definitely enough to keep me warm while camping for 3 full nights.

Be the First to Share


    • Game Design: Student Design Challenge

      Game Design: Student Design Challenge
    • For the Home Contest

      For the Home Contest
    • Make It Bridge

      Make It Bridge



    6 years ago


    You could consider adding a thermostat (PWM controlled by a NTC resistor would do fine), to get just the minimal heating needed.

    For an armband (bracelet), you could use two loops of memory wire (the jewellery kind, not "real" memory wire like NiTiNol ;), with one sown into each side seam (if you get the idea), then it will need nothing further to stay on your arm

    Have a nice day.