Creative Commons images are CC BY 2.0

Description: This is a three axis accelerometer for the LilyPad system. Based on the ADXL335 MEMS accelerometer from Analog Devices, the LilyPad Accelerometer can detect joint movement as well as inclination and vibration.

The ADXL335 outputs a 0V to 3V analog signal on each of the X, Y, and Z axis. You will need to convert this analog voltage to a gravity amount and use trigonometry to calculate a true ‘angle’. If you are not looking for tilt, it is very easy to use this sensor for basic motion sensing.

LilyPad is a wearable e-textile technology developed by Leah Buechley and cooperatively designed by Leah and SparkFun. Each LilyPad was creatively designed to have large connecting pads to allow them to be sewn into clothing. Various input, output, power, and sensor boards are available. They’re even washable!

Note: A portion of this sale is given back to Dr. Leah Buechley for continued development and education of e-textiles.


  • 20mm outer diameter
  • Thin 0.8mm PCB


Recommended Products

Customer Comments

  • Usually the voltage used for the LilyPad system is 5V (because both the LilyPad LiPower and LilyPad Power Supply output 5V) - therefore, with the resistor still in place, this LilyPad accelerometer has been designed to accept 5V.
    If you want to use an input voltage of 3V, you can simply remove the resistor and solder a bridge between the pads.
    Note: If you are using the conductive wire to connect this to a power supply, be wary that the thread itself has resistance - so will need to take this into account when selecting an input voltage (due to the voltage drop across the conductive thread).

  • Can this accelerometer be used to calculate a jogger’s velocity? I know that some GPS are too inaccurate.
    Thank you in advance

    • Consider what an accelerometer actually measures; acceleration. So now consider how you get speed from acceleration - by integrating over time. Now consider the inaccuracies of the accelerometer, the integration methods and a host of other things (starting with the jogger not being an ideal directional vector, but rather something that’s going to make the accelerometer axes outputs bounce around pretty wildly), and the answer is typically “yes, but over small time periods, with a known reference velocity (usually: standing still), and with a fairly big and accumulating error (drift)”.

      You’re better off with a GPS since the inaccuracy tends to be relatively low and for calculations actually gets lower over distance and time (although you’re really measuring the average speed). GPS, on the other hand, is horrible at instantaneous events.

      If you still want to pursue it, just google ‘accelerometer velocity’ - there are some existing implementations and the sensor’s accuracy of course does matter. Just know that the general conclusion is that it is the wrong sensory input for the task :)

      Of course you could get the jogger to get into rollerskating instead and use encoders on the wheels… hm. ;)

      • thank you for your answer. I know that latest jogging watches use only GPS systems. As you already said GPS isn’t so accurate at instantaneous event, so if the accelerometer isn’t suitable for this aim, what should I use to create a velocity indicator with lilypad? Thank you again

        • Honestly, for a jogger, I don’t think there is a perfect system other than a measurement from a fixed reference (or reference with known velocity and vector); think speed traps or doppler. But that’s not going to work if the measuring device is on the person themselves :)

          GPS really should be ‘good enough’ for most joggers to get a sense of how fast they’re going, but if you wanted to try and improve things, you could always try combining the two. Use the GPS to get a speed measurement, then integrate the acceleration over time to get the delta over that calculated speed. It’ll still be fairly error-prone. On the up side, you can then offer position logs, a rudimentary pedometer (with the assumption that a ‘bounce’ is a step), etc. should that be an eventual interest.

          Is there a specific target application for this? There may be other approaches that work better.

  • Does this device integrate well with the regular arduino lilypad?

  • Just bought one of these. The installed resistor is 0 ohms, so apparently that means it’s setup for 3 volts these days.

  • Can this detect a heart rate if placed in the chest area?

    • You might be able to get some readings out from this, but more likely than not, it will simply look like noise in the data.

  • So what is the input Voltage for this? 5V? If so, what is the R1 value on the schematic? 14kOhm?
    Why can’t you use the empty connection for 3V input?

Customer Reviews

5 out of 5

Based on 2 ratings:

Currently viewing all customer reviews.

1 of 1 found this helpful:

Works like a charm

Easy to program and doesn’t use much memory. Will buy again.

Works well

This sensor is working well, and I recommend this documentation page for wiring diagram and example code:

Support Tips

Additional Example Tutorials

For more tutorials on using the ADXL335, try looking at the product page for the standard breakout for the ADXL335 =>