Announcement

Collapse
No announcement yet.

Managing Electrostatic Discharge effects

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #16
    Originally posted by FTC7253 View Post
    ... Is the choke actually effective in reducing the impact of transient ESD spikes? I can absolutely understand how the low-pass behavior of ferrite chokes could improve integrity of the data signal by reducing continuous EMI noise (possibly caused by noisy power on the adjacent wires), ...
    I would be more concerned about the EMI from the DC motors and the servos. Also, there is an EMI component to a ESD event. So if the chokes are designed to reduce EMI, then yes it is a good idea to use them on data wires.

    Comment


    • #17
      Alec Ok. I think I can buy that the DC motors and servos can introduce EMI. I am also willing to entertain the idea that there is an associated short-term EMI component to and ESD event. Both of those bring us back around to what is the general frequency range of the EMI induced noise that we would like to filter out. I have not been able to find chokes on the common FTC supplier sites, and have not seen suggestions on specific chokes. But I do know that different chokes are suited for filtering somewhat specific frequencies. Any idea on the ideal choke properties and also on vendors?

      Comment


      • #18
        ESD has been an issue in FTC for a long time. In my very first year mentoring, our team and alliance partner both frustratingly disconnected in elims at our state championship, leaving them unable to play the game after a long season of time and expense. It was very very frustrating and surprisingly common. Since then, I've made it a priority to try to deal with ESD and disconnections as much as possible with teams I mentor.

        A few years ago, this document came out describing ESD issues and potential mitigation strategies. It described experiments where adding ferrite chokes clearly decreased the frequency of ESD events. It also led to anti-static spray becoming more widely used on FTC fields.

        When that doc came out, I bought a bunch of chokes, similar to these. My teams added them to quality USB cables and also did everything possible to try to mitigate discharges and disconnects: bracing USB cables and connections, wiping down the robot with anti-static spray periodically, shielding exposed metal to reduce metal-to-metal discharges, making sure batteries are "good" and fully charged to better deal with brown-outs, etc. This definitely led to much more reliable robots with fewer disconnection problems, and IMHO directly led to a surprising victory at a major FTC event a few years ago where other teams with much better robots had static disconnection problems while our team with our mediocre-but-reliable robot did not.

        We've never had a problem with the chokes, and seem to have fewer disconnection problems compared to other teams. I certainly recommend using chokes as described in both the old and new documents. So does one of the best teams in the world.

        4634 suggests that the Modern Robotics cables with chokes have not been reliable. I agree though I believe it is because they have looser connections that seem to wiggle and disconnect more frequently. I don't believe the issues with those cables are related to the chokes on the cables.

        Comment


        • #19
          How does FRC deal with ESD & EMI?

          Comment


          • #20
            We never have ESD & EMI issues with FRC. This is probably partly due to the fact that FRC electronics are a lot more robust than FTC electronics. The brain in FRC is RoboRIO which is specifically designed for robotics use whereas the brain in FTC is an Android phone which is definitely not designed for robotics. Especially, the communication link in FRC is CAN bus which is designed for the noisy automotive environment whereas in FTC, it is USB which is sensitive to noise and disconnects on wimpy micro/mini USB connectors. I am sure there is also a factor on the field that FTC is on foam tile which is known for producing static electric charge. There was a rumor that REV has a robot controller/Hub (https://ftcforum.usfirst.org/forum/f...or-next-season) could replace the Android phone in the future and hopefully is more robust to ESD than an Android phone. I have also been hoping that FTC will adopt CAN bus just like FRC so to make the robust components of FRC available for FTC, but it looks like the REV control hub is still using USB :-(

            Comment


            • #21
              Every so often the question comes up about FRC and why they don't have these problems. I think the real answer is that we don't know entirely. That being said I think there are a few key differences.

              1) Some failures are blamed on static that are really due to poor cables & connections. The USB cables are not as robust as the FRC connectors. I see a surprising number of phones sitting in a robot with only gravity and the USB cable holding it in place. I don't think I've every seen a RobotRIO that wasn't fastened down.

              Then again there are also many that I'm pretty sure are static related that are blamed on bad cables. I see people blaming cables when being asked by teams to spray down a field. Those teams that have clearly suffered from static. When a robot touches the side wall and it stops moving at that instant, that screams static to me.

              2) The tiles and wheels in FTC are static generators compared to the wheels & carpet of FRC. In a dry environment I constantly see students flinch when they pick up their robot - I presume that its from static shock. I don't see that happening in FRC events. (living in MN with dry winter air - static is always a possibility)

              I don't think the phones themselves are susceptible to static. They are built to take a lot of abuse. I don't know that I've ever seen a phone lock up from static. I have seen the Modern Robotics modules and sometimes the Rev controllers lock up. The Rev controllers appear to be better (That is admittedly subjective evidence - I don't have hard numbers). And going back in history we saw that the NXT was very static susceptible.

              Comment


              • #22
                Originally posted by mikets View Post
                looks like the REV control hub is still using USB :-(
                The Control Hub communicates to the Lynx PCB over a direct UART link via an internal pin header.

                Comment


                • #23
                  Chokes increase the inductance of the signal lines and inductors block changes in current (V = L di/dt), which is what ESD events are. If we consider the object applying the shock to the signal line as a capacitor at a very high voltage, we see that discharging the object through an inductor will spread the current flow over a longer period of time in comparison to discharging it through a wire with no inductance. Because the charge transfer is the same in both cases, the peak current is reduced when there is more inductance and you are less likely to damage your electronics. This leads to the conclusion that you want as much inductance as is practical, and roughly that bigger chokes should be better.

                  For practical reasons, you probably just want to buy chokes that fit nicely on your wires. If I remember correctly, the chokes I used in testing over the summer were 5mm clip on ferrite chokes, which fit well on various wires. AndyMark sells chokes that are comparable, but you could also look at DigiKey, McMaster Carr, Amazon, etc.

                  If someone wants to test some different options more thoroughly in the real world, I'd be curious to see the results and I'm sure the community would appreciate it.

                  Hope that helps!

                  -Eric

                  Comment

                  Working...
                  X