Sembazuru

Member Since: October 17, 2012

Country: United States

  • Nice. Good science there.

  • One thing that I noticed in the video was the track jumping as a result of the centrifugal effect created when the car was going through the loop. Did you have any problems with false triggering by the track jumping up and crossing the trigger gates?

  • More mounting locations helps prevent rotation of the board w/o using stops on either side.

    Because the physical board is bigger, you are really paying for a more expensive piece of FR4. It looks to be roughly twice the size, so the PCB cost in the BOM is (roughly) twice that of the smaller stick. (Remember, the main contributor for the cost of a PCB is square footage.)

  • Another method for calibrating the thermocouple is to use known physical models in the environment. The simplest one is check it out in both ice water and boiling water. (Granted, you should probably look up the freezing and boiling point for your actual altitude and use that instead of the “sea level” temperatures of 0°C and 100°C…)

  • Typo-o in the first paragraph. “TO-220 7804” should be “TO-220 7805”

  • Sounds kind of like how a CCD works.

  • I attended a glass etching class at my local makerspace and the teacher there used a Silhouette Portrait for cutting her vinyl transfers. The software allowed importing any bitmap and converting it to a vector cutting path. (I’m not sure if the software supports importing vector diagrams). Here is a review from Make: on an older version of Silhouette’s flagship model, Cameo. I just checked Silhouette’s web site and found that the Cameo is up to Cameo 3. I haven’t put in the research to see what is new though. But, I think all the models use the same software…

  • Usually, sign cutting vinyl cutters use a metal blade for cutting. The problem with a laser cutter for vinyl (or paper) is the laser burns the medium. Ever get a match near many sheet plastics? The plastic curls back on itself away from the flame like it is alive. The same would happen on either side of a laser cut obliterating very fine detail.

  • Might be simpler, but if you can multiplex several functions onto one component, that is less components you have to purchase and helps reduce your full project BOM cost.

  • Well, there is an opportunity for enhancement. Program in a “no internet” cache system that stores data until internet is restored and then uploads the data. This will require you to keep track of time on the data collection unit and in addition to your data (soil moisture, soil temperature, battery level, etc.) also include a date/time stamp for when the data was collected from the sensors. Then when plotting the data from your cloud service of choice use the date/time stamp that you uploaded with the data instead of the automatic date/time stamp that the cloud service assigned when it received the data. Since it is a nominally internet attached project, use an NTP service to set the on-board clock, periodically resynchronizing to adjust for any onboard clock drift. (I know there is version of the Arduino RTC library that automatically takes care of synchronizing to an NTP server. I don’t know if Particle has something similar.)

    See how simple projects can get complicated quickly? ;-)