Announcement

Collapse
No announcement yet.

Servos with new Modern Robotics Hardware

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

  • #31
    We are running into this problem of our servo, HS-485HB, not going the full 180 range of motion using MR Servo controller. We need the full range. Has anyone found a solution using HS-485HB or what other HiTec servo to buy that gives the full 180 range using the MR Servo controller.

    Thanks,
    Cesar
    Team 5073

    Comment


    • #32
      Originally posted by FTC5073 View Post
      We are running into this problem of our servo, HS-485HB, not going the full 180 range of motion using MR Servo controller. We need the full range. Has anyone found a solution using HS-485HB or what other HiTec servo to buy that gives the full 180 range using the MR Servo controller.

      Thanks,
      Cesar
      Team 5073

      We'd be very interested in hearing about any solutions or work-around as well. We spent the first half of our season fighting the well documented control issues, and now that we have that (hopefully) solved, the team is concentrating on game solutions, and the servo issue is pretty daunting. Basically, and yes we're "preaching to the choir", it's pretty difficult for the kids to design any working mechanism based on the publish servo specs when you can't predict with ANY level of certainty what range the servo may/may not perform at.

      Any findings, tips, etc. are appreciated,
      Redfish Robotics
      FIRST Tech Challenge Team 9958
      http://www.redfishrobotics.com
      https://twitter.com/RedfishRobotics
      "We're Hooked on FIRST"

      Comment


      • #33
        I hesitate to post this because I'm not 100% sure we aren't sending you down a black hole.

        About a month ago, we switched all of our servos over to HiTec digital servos. We didn't want to go to the Legacy controller and keep the old terminal-block wired HiTechnic servo controllers. The nice thing about the digital servos is that they are programmable. You can by a programmer device for $25 and set all kinds of fun and interesting things on the servos, INCLUDING their range of motion. Basically, you can set Left point, right point, mid point and fail-safe point and the servo scales the incoming pulse width to correspond to the range of motion you give it.

        We happily installed the servos and it worked just like we wanted ... until it didn't. All of a sudden our servos started twitching. Sometimes a little, sometimes violently. We entered a 2 week black hole 3 weeks before our first competition. We did everything we could possibly think of to isolate the problem, thinking it was likely a unknown coding error. But we discovered that the situation was worse than that. As soon as we write a value to initialize any servo position, all the servos became subject to this chaos ... but not every time. Eventually we stripped all the code out except initialization and saw no change to the randomness. Then we started replacing servos and going back to Analog one by one, desperately grabbing servos off old robots to get enough.

        We replaced enough of the servos and switched to the Tetrix controller and the problem went away. Was it the last servo we replaced? No. We could replace that one and put a different digital one in and it still worked. There were about 4-6 servos on each controller and the servos had a minimal current draw when idle. We haven't had a chance to go back and try the digitals with the Tetrix controller. That will happen soon. We did have one high torque digital still on the robot, just not all of the smaller ones.

        So... you could try digital. I suspect there is nothing wrong with the servos. I'm not as confident that the electrical noise the digital servos produce doesn't completely whack out the servo controllers.

        Another warning ... the digital servos died when on our robot. Whatever was causing the twitching must have been cooking them slowly internally. They would get very hard to turn ... as if something melted inside them. It happened to 5 of the 8 we bought (ouch!). Again, no idea why this was happening or if it is a cause or effect.

        Good Luck... I'd be interested to know if anyone else is having success with HiTec digital servos and the new Modern Robotics system.

        Mark Keup
        Coach of the Supposable Thumbs - FTC 4106

        Comment


        • #34
          I know that the HS-645MG 180-degree servos do indeed go 180 degrees. I do not know what other ones do.
          Programmer for Team 4997 Masquerade -- 2012 World Champions, 2014 - 2016 Division Finalists
          Founding Member of Team 6433 Neutrinos -- 2015 World Champions

          Check out my intro video to the new tech platform
          Check out my team's Robot Reveal for Res-Q

          Comment


          • #35
            Originally posted by Varun Singh View Post
            I know that the HS-645MG 180-degree servos do indeed go 180 degrees. I do not know what other ones do.
            I'm glad you mentioned this, because the ServoCity page for the HS-645MG says it needs a 2400 microsecond pulse width to get the 180 degrees. I wouldn't have thought to have my team try it out based on this page.
            John McDonnell
            Co-Mentor, Team 5873
            https://www.facebook.com/Team5873

            Comment


            • #36
              As you can see this is not a simple issue!
              The HS-645MG will NOT turn 180 degrees on the MR controller. It needs a signal of 600-2400 msec to turn 180 degrees, the same as the HS-485HB and other analog servos. The MR controller only does ~750-2230 msec. This works out to be 145 to 150 degrees. Servo City will 'modify' any analog servo by installing resistors in them.
              This is NOT allowed in this competition!!!

              The HS-5485HB is the digital version of the HS-485HB and can be programmed to turn 180 degrees.

              Digital servos have a much greater standing toque than analog ones, up to 5x. This is great if you need more holding toque but it also means that if you stall them they will burn up 5x as fast. When this happens the motor comes apart internally and puts an excessive load on the controller (I assume it could damage the controller as well). All the servos including the analog ones on that controller will twitch as Mark described. This can be caused by just one bad servo per controller. The ones that are hard to turn are dead.

              We used the HS-5485HB servos with no problems (after burning one of our first ones up this way). No servo likes being stalled out for a long time, but digital ones are far easier to burn up. We use their full toque to push and hold things but are very careful not to stall any servo into fixed stops. We clip the set point value in the software to just touch the stop.

              It worked well at regionals so we will see how it does at state. Google FTC10432 to see videos.

              Hope this is helpful-
              Tim

              Comment


              • #37
                Hi Tim,

                So what you are saying is that if if there is a bad digital servo (as in drawing high current?) the controller will cause all servos to twitch ... or is it the controller that goes? Does any servo drawing past the max output of the MR controller cause the same thing? We are curious to start debugging this again.

                What is odd is that you mention stalling the servos out. Some of ours that died were installed supported with with bearings and Actobotics Servo Hubs and never had any arms or anything attached to them yet, but they still failed. I'm starting to wonder if using the programmer to change the end points somehow contributed to the problem. At $35 a pop, I'm only going to be able to afford a few more tests! I'm also wondering if we burned one out and then the twitching started and the twitching caused the other servos to burn out, even sitting there with no loads on them. Have you seen this behavior on the legacy servo controller?

                I'd hate the kids to get into superstition mode --- all digital servos are bad -- because they would lose a lot of interesting options if they do that.

                Mark

                Comment


                • #38
                  Originally posted by FTC10432 View Post
                  Digital servos have a much greater standing toque than analog ones, up to 5x. This is great if you need more holding toque but it also means that if you stall them they will burn up 5x as fast. When this happens the motor comes apart internally and puts an excessive load on the controller (I assume it could damage the controller as well). All the servos including the analog ones on that controller will twitch as Mark described. This can be caused by just one bad servo per controller. The ones that are hard to turn are dead.
                  We have a digital servo too but it gave us nothing but trouble. May be you can shed some light on what is really happening. We bought the expensive digital servo because of its greater torque. When it arrived, it worked pretty well for a few minutes. But soon, it's very hot to the touch. Then, when we set it to a certain position, it tried moving to the position but very jerkily. At times, we have to "help" it to overcome gravity to a certain angle. We have a "servo tester" made out of the obsolete NXT Mindstorms and the HiTechnic servo controller. Today, we connected the digital servo to this tester, it worked very well, no jerkiness. So we suspect the Modern Robotics servo controller. We have two Modern Robotics servo controllers on the robot. The other one happened to have one spare channel left. So we connected the digital servo to the other Modern Robotics servo controller. It behaved the same way, very jerky and couldn't quite get to the set position unless we "helped" it. Then just for the heck of it, we connected a HiTechnic servo to the legacy module of our robot and changed the servo name to the first channel on the HiTechnic sero controller. Interestingly, the arm went to the forward position very swiftly without hesitation but it couldn't retract. You can feel the digital servo was holding its position at the extended angle, but the code was commanding it to retract. Just to make sure the code is actually working correctly, we unhooked the digital servo and hooked up a regular servo and we can turn the regular servo in both directions but hooking the digital servo back to the same HiTechnic servo channel, it can only extend but not retract. I have no idea what's going on with the digital servo. It literally has a mind of its own and doesn't always listen to our command. Is the Modern Robotics servo controller damaged? But why is the other Modern Robotics servo controller behaved the same way? Are both damaged? If so, why are there no problems with other regular servos on the same controller, only the digital servo? I am out of ideas on what else to try.

                  Comment


                  • #39
                    Hi Mark,
                    Yes, one bad servo causes all of them to go nuts. We never damaged a controller but did not let it operate like that very long.
                    The servos typically got fairly hot. They do not any more.

                    Remember that a '90' degree servo uses a 'normal signal' of 1100-1900 microsec PWM signal but the MR controller can overdrive it to ~145 degrees with 750-2230 msec (HT controller goes even farther but this is our first year so we do not have one....). If you set the digital servo to go 180 degrees on 1100-1900 msec. the MR controller can drive it into it's internal stop, not just a stop on the robot.

                    Our programmer set up a program to increase the signal slowly with the game pad and display the controller set point on the driver station. This made it easy to find the end points to set up clipping the output to that servo. Perhaps this is overkill but I do not think so. If this was a standard people would not have to be as concerned with the total stall current of the servos and the controller rating.

                    I think the digital servos work great now with no issues.

                    Tim

                    Comment


                    • #40
                      Originally posted by FTC10432 View Post
                      The HS-645MG will NOT turn 180 degrees on the MR controller.
                      It works well on our robot. We have had no problems at all with it.
                      Programmer for Team 4997 Masquerade -- 2012 World Champions, 2014 - 2016 Division Finalists
                      Founding Member of Team 6433 Neutrinos -- 2015 World Champions

                      Check out my intro video to the new tech platform
                      Check out my team's Robot Reveal for Res-Q

                      Comment


                      • #41
                        Apparently there is a change to address the servo range issue in the new 2.0 firmware for the Modern Robotics servo controller: http://www.modernroboticsinc.com/firmware-update

                        Comment

                        Working...
                        X