Popsicle stick bridge crusher

An automated bridge crusher that was designed to crush Popsicle stick bridges in a repeatable and consistent manor. Can measure up to 500kg.

Similar projects worth following
The third iteration of the project is what I am currently working on and consists of the following modules and improvements:
1. Common connectors on both motherboard and external device (RJ45)
2. Consistent cables allowing for easier repair (Cat 5e)
3. Shielded cables to allow for less noise on analog signals (STP)
4. Use interrupts for all digital inputs
5. Use properly documents state machines for all systems
6. Remove variable gain components from in the path of the analog pressure measurements

In the second iteration of the project (the one that the images are shown for)  the device was made under-budget and under the time constraint but with some major caveats. While the device had a working control system and was fairly reliable in performance. 

The major issues that we encountered were:

  1. The bridge crusher was not easy to be calibrated
  2. lost its calibration every time the device moved (or vibrated)
  3. wiring suffered from major EMC issues
  4. in order to take out a switch or other component wires needed to be de-soldered.
  5. software on computer was restricted to OS and required technical setup to use
  6. required external power supply to make the hmi and control system to work
  7. changing settings required recompiling software
  8. not consistent programming framework

For Review.pdf

First attempt at creating the new motherboard

Adobe Portable Document Format - 202.94 kB - 02/16/2020 at 06:41


  • Updating the Motherboard plan

    Joel Newman02/23/2020 at 23:06 0 comments

    While the first revision of the motherboard worked well there were a few things that needed to be fixed after working with the prototype. 

    Deficiencies noted:

    1. Inconsistent pin layout in RJ45 connectors
    2. many unused pins that could be used for other things
    3. board mounting holes are in odd location
    4. RJ45 connectors should be shielded in case shielded cables want to be used
    5. No way to tell if sensors plugged in or not
    6. GPIO extender for LED's was wrong footprint and to close to the jacks
    7. No good grounding point for a oscilloscope for testing
    8. No way to power the device without USB power
      1. Would be nice to power this from 12V and have on board converters
    9. LED lights should also be controlled from motherboard
    10. ADC had a bunch of noise on the input
      1. No can to prevent unwanted EMI
      2. No analog filtering to prevent aliasing
    11. Restricted use of pins on the microcontroller
    12. only external comms was though USB
    13. Soldering was substandard

    Objectives for next revision of the motherboard:

    1. Change the layout on the RJ45 pins to mach the PoE standard 
      1. Addresses deficiency 1
    2. Change the RJ45 sockets to be shielded to prevent EMI leakage
      1. Addresses deficiency 10
    3. On the sensor side of the circuit tie a pin from the power to a set pin (standard across all connectors) with a resistor to let the system know which device is plugged in. 
      1. It may be a good idea to include a resistor on each device so that the circuit can autodetect what kind of switch or sensor is plugged into a socket
        1. probably overkill for this project as it would make the programming much more difficult and not all jacks can support all types of signals.
      2. Need to include a GPIO extender to test all of the IO to see if its plugged in or not.
      3. Addresses deficiencies 2 and 5
    4. Analog Filter for input to ADC
      1. requires more thorough investigation on input topologies for high noise applications. Most likely going to use an example circuit from here ( to design a better bridge circuit.
    5. External power module so that usb or external 12V power can be used
      1. Addresses deficiency 8
    6. Use a standard microcontroller not a development board.
      1. Most likely a SAMD variant as the built in USB support makes the programming easier
      2. Addresses deficiencies 2,5,9, and 11. Possible 12 as well.
    7. Complete redesign of PCB
      1. Addresses all deficiencies

View project log

Enjoy this project?



Dan Maloney wrote 02/19/2020 at 17:39 point

I found it interesting that the Popsicle Bridge Challenge is so well-established. And that 500 kg load is actually needed for some of these designs.

I also saw that a lot of people test their bridges by just piling weights on top. Looks like this will do a point load in the center of the span, which seems like a more realistic way to test the structure.

  Are you sure? yes | no

Joel Newman wrote 02/23/2020 at 21:58 point

Thanks, in the past this bridge crusher has seen forces up to ~300kg when testing pieces of wood and about ~220kg when testing a bridge. The event that this was created for had about 40 or so bridges to test and the event organizers did not want to constantly be loading the bridges by hand. Also by having an automated test setup the results were more comparable. Another benefit by this was once the bridge reached its maximum stress the child could save the bridge if they wanted to.

One thing that was questionable is that the "calibration" of the system was done with a fish scale and was not the most accurate as the gain of the ADC was controlled by a small pot. This made vibrations change the gain over time.

  Are you sure? yes | no

Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates