Announcement

Collapse
No announcement yet.

Update on USB Disconnect Issues (unable to detect USB modules during scan)

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

  • I kind of wish a legal combination was:

    - Phone directly to legacy controller (no hub)

    - HiTechnic motor controllers, lego sensors, including the sensor breakout board.

    - DC power wiring as it was last year, but emphasize how cool power poles are.

    It's small and simple to wire, lets you use the Android devices as the controller (and the new software platform).

    /Mitch.
    Mitch Lichtenberg
    Technical Mentor, Saratoga High School Mechanical Science and Engineering Team (M-SET)
    FRC 649
    FTC 6165, 7641, 7390

    Comment


    • Originally posted by Mitch View Post
      I kind of wish a legal combination was:

      - Phone directly to legacy controller (no hub)

      - HiTechnic motor controllers, lego sensors, including the sensor breakout board.

      - DC power wiring as it was last year, but emphasize how cool power poles are.

      It's small and simple to wire, lets you use the Android devices as the controller (and the new software platform).

      /Mitch.
      If they abolish the PDM, then that will be legal since there is nothing in the rules that says you have to use a COTS USB hub. A COTS USB hub is permitted but not required.

      Comment


      • Dissapointed

        Originally posted by Cheer4FTC View Post
        I'm heartbroken to say that I also agree with everything said by Mitch and hexafraction. It is difficult to express how disappointing the rollout of the new technology, its robustness, the testing done before rollout (or lack-thereof), and the responsiveness of the responsible parties (or lack-thereof) has been this season. IMHO, the entire FTC season and program has been put at risk.


        Experienced teams might decide the same.
        I 100% agree. I took a year off as a coach and came back this year with a new team and am having to deal with new technology. Thank goodness for the team members eagerness to learn, or I'd be ready to throw the towel in. Having to tell the kids it is "black magic" to get the controllers to be recognized is crazy when Mitch has clearly identified a work around. And having to send a PDM in for re-work one week before competition is crazy too. Yes, I just sent one in Tuesday, and we have a competition next weekend. I don't expect to get it back until next week. Samantha looks pretty darn good right now...

        Comment


        • For what it's worth, I just sent a PDM in last Friday and it arrived back here today. I used USPS priority mail - the smallest standard box.
          We'll see how the refurbished PDM works out.

          Martin Haeberli
          Mentor, FTC 7593 TigerBots

          Originally posted by FTC3525 View Post
          I 100% agree. I took a year off as a coach and came back this year with a new team and am having to deal with new technology. Thank goodness for the team members eagerness to learn, or I'd be ready to throw the towel in. Having to tell the kids it is "black magic" to get the controllers to be recognized is crazy when Mitch has clearly identified a work around. And having to send a PDM in for re-work one week before competition is crazy too. Yes, I just sent one in Tuesday, and we have a competition next weekend. I don't expect to get it back until next week. Samantha looks pretty darn good right now...

          Comment


          • Originally posted by mhaeberli View Post
            For what it's worth, I just sent a PDM in last Friday and it arrived back here today. I used USPS priority mail - the smallest standard box.
            We'll see how the refurbished PDM works out.

            Martin Haeberli
            Mentor, FTC 7593 TigerBots
            We have one on its way as well. Hoping that there's a new hardware revision out there that fixes these bugs. If not, I'll try to design some extra protector boards with TVS and polyfuses to at least have some safety when doing initial tests of a new wiring scheme.
            FTC6460 mentor (software+computer vision+electronics), FPGA enthusiast. In favor of allowing custom electronics on FTC bots.
            Co-founder of ##ftc live chat for FTC programming--currently you may need to join and wait some time for help--volunteer basis only.

            Comment


            • Originally posted by mikets View Post
              Owning the system for a little over 2 months and doing a few dozen connects/disconnects and the connector on the cable is worn? I wouldn't say that's a "considerable amount". I would expect that after at least a year of frequent use. I hope the replacement cables I am getting will do better.
              The micro-USB connectors are designed for phone users, not robots. The design goal actually is an easy disconnect as we sometimes walk away with a plugged in charger cable and in those cases we want the cable to disconnect and not tear the guts out of the phone or make us drop the phone.

              One simple solution could be a 90 degree mirco-USB connector and a rubber band that holds it in place. Or a 3D printed case to hold a right-angle connector in place.

              Comment


              • Originally posted by FTC5064 View Post
                The micro-USB connectors are designed for phone users, not robots. The design goal actually is an easy disconnect as we sometimes walk away with a plugged in charger cable and in those cases we want the cable to disconnect and not tear the guts out of the phone or make us drop the phone.

                One simple solution could be a 90 degree mirco-USB connector and a rubber band that holds it in place. Or a 3D printed case to hold a right-angle connector in place.
                http://www.thingiverse.com/thing:1096699
                http://www.thingiverse.com/thing:1099628
                2015 FTC World Champion - Valley X Robotics 2844 - Founding Memeber

                Comment


                • Originally posted by GLDad View Post
                  By the way, my comments about this , being a battery issue, is me coming from a background where we ( RC car racers ) would use 2400 mAH 7.2 volt battery packs and completely drain the 2400 mAh pack in 4 about 4 minutes and 20 seconds. Over the years, as we moved into NiMh batteries and got up into 3000 mAh packs, we could dump them in 5 minutes. My point is that any decent modern NiMh cell could easily be drained at 28 amps or more and not sag bellow much bellow 1 volts per cell. Assuming a 10 cell pack, draining at 20 amps , it should not drop anywhere near 9 volts. I'll get my old Competition Electronics Turbothrity (charger/discharger) out and do some testing to know for sure, but I REALLY think it's bordering on absurd to attribute this type of problem to batteries.

                  If the USB power is dropping bellow it's needed 5 volts, and it's somehow related to voltage from the battery, then there needs to be a better way to build the robot electronics, because it should not be a problem.
                  Hi GLDad,

                  The problem is a little bit different here. The digital devices on board the robot (the micro processors in the modules) require a steady 5V signal to operate properly. These are essentially mini computers that reside in these devices. The robot rules require that the devices all be powered from a single 12V battery. This means that both the DC motors, the servos, and the electronic modules all be powered from the single battery. The situation is slightly different than from a typical RC car or plane. in order to provide a steady 5V DC power source to the USB devices, the PDM is equipped with a voltage regulator that steps down the power (in an efficient way) from the 12V level available from the battery. This voltage regulator has a fairly low dropout voltage (this is a good thing) of 1.5V or so. This means that it will continue to provide reliable 5VDC power to the modules even if its input voltage drops to 6.5 Volts.

                  A normal 12V battery has enough capacity to reliably run the DC motors, servos and devices found on a typical FTC robot. When I talked about the 9V battery, I was specifically testing the case of a Matrix 9V battery powering the 12V motors. Although a 9V RC battery (rated for 2400 mAh, but this was an older battery and it might only have a practical capacity closer to 1900 mAh) can power the RC motors for a long time without any problem, an event (such as both motors moving from 100% full speed ahead to 100% full speed backwards) can cause a very short but large draw in current on the battery. We suspect that this can cause a momentary state where the voltage of the battery drops to below the dropout voltage of 6.5 Volts of the voltage regulator. This temporary drop in voltage (which might not be easily seen/detected, since the time frame is so short) can cause a brownout condition in the digital microprocessors, which require a steady 5V DC power supply, even with the capacitors in place (which are designed to smooth out power and filter out noise on the lines).

                  These robots are a little different than a typical RC car. For example, the legacy module and the device interface module allow you to connect all types of sensors and devices to them and the Android phone can communicate with them digitally and send information back and forth from/to these devices. Similarly the Android device can talk to the motor and servo controllers to get information about the position of the encoder or servos, the input voltage to the motor controllers, etc. These digital components require a steady 5V power supply.

                  You should also note that when you power a robot using a 12V 3000 mAh battery that's in decent shape and reasonably charged, this battery has enough capacity to run the robots properly, even if there are large spikes in current draw from the devices on the robot. The 9V test that I did was to see if the smaller 9V Matrix battery (which is a legacy component used in previous seasons) had enough capacity to power the larger 12V motors and controllers. Please be careful when interpreting these reported results.

                  Tom

                  Comment


                  • Originally posted by GLDad View Post
                    I continue to have issues , I've secured my cables, done everything I can to eliminate the cables being loose and causing communications problems .

                    I have to question the comments about batteries being a cause of this problem or even contributing significantly to it. The specs on the Power Distribution module list Power : 9 – 15 volts DC. 20 amps max. If a person using one of these with a 9 volt battery, can't operate four motors reliably , under very little load ( driving on a flat smooth surface with the worst load being a sudden change in direction ), there is a problem with the design of this product. If a person is using a 12 volt battery, and can 't reliably work under moderate to low load, there is a design problem.

                    Anyway, I look forward to more detailed response from FIRST and/or Modern Robotics on this matter, but what's been said so far is a bit of a let down.
                    GLDad,

                    If you are still having problems with your modules, then I suggest you contact Modern Robotics to troubleshoot. I am preparing a more detailed response on the matter, but in the interim, I suggest that you reach out to MR and call their support line for help. If you are sure that your wires are secured and your battery is fresh, then you should check with them to determine if you potentially have any hardware issues. They are more than happy and capable to help do some quick diagnostic tests with you and determine if your modules are working properly or if they need to be serviced. Note that one of their engineers is an FTC alum and mentor. They also have an efficient RMA process in place to help you get parts to you in an expedited manner.

                    Tom

                    Comment


                    • Originally posted by ViperMentor View Post
                      Yes and no. Both times I was handed blown PDMs, only the regulator chip was blown. However, for a temporary fix, I replaced the blown regulator with a standard 5V regular (which fixed the PDM). Unfortunately, the second time this happened, I did not realize that the reason the team had blown the PDM was because when they had replaced the connector with Powerpoles, they had connected the two halves backwards (as opposed to the first blown PDM which was wired correctly but then connected to a battery from last season with reversed Powerpoles). So naturally, when another team tested the "repaired" PDM they promptly blew it up again (because again they ignored the fact that if one is reversed, you can only physically connect red to black and black to red). This time that capacitor did blow up.
                      Thanks for the information ViperMentor!

                      Comment


                      • Nicely done R2D2!

                        Comment


                        • Originally posted by Alec View Post
                          Thanks Mitch! It will awesome if this fixes the MR core module scan issues.



                          Hi emeflag. Please note that Tom can reproduce the PDM failure with a simple bot with 2 motors:



                          The problem is the condition of the battery, and the load on the motors, will determine how quickly your battery will go from 13.7V to 12V (or to 9V), and be at a point where a power surge from the motors will knock out the PDM, sending your bot tumbling down the RES-Q mountain taking out everything in its path. On the positive side, it'll be a spectacular sight, and the video may go viral. But is that the type of publicity we want?

                          Disallowing the PDM is a better solution in my opinion.
                          Hi Alec,

                          The problem that I described is different from the one Mitch is referring to. The problem that I was testing was using a 9V battery to power a 12V set of motors. This is not what the battery or motors were designed for, but we wanted to test it to see if it could run properly. The situation Mitch described is related to power up of the modules. The test case I described was driving the motors (they powered up properly).

                          A 12V motor in good condition and that is charged has enough capacity to reliably run a typical FTC bot for more than a match. The case of the simple K9 bot was really just designed to see if we could run a 12V system with a 9V battery (the answer is that it is not recommended).

                          Tom

                          Comment


                          • Sorry, I would like to clarify something here - I understand that there are two problems being discussed, but one appears to be resolved by placing a capacitor on the new MR DC motor/servo controllers, right? Has anyone tested to see if the problem occurs when using ONLY the Legacy module and the OLD DC motor/servo controllers?

                            Comment


                            • Originally posted by Tom Eng View Post
                              Nicely done R2D2!
                              I can only take credit for the PDM one. Looking at doing a few other ideas too
                              2015 FTC World Champion - Valley X Robotics 2844 - Founding Memeber

                              Comment


                              • Tom Eng,

                                I realize that current FTC robot setup, with it's Android phones and USB based communications between devices is a little different then your typical RC vehicle. However, they are similar in many ways, and I strongly believe that the FTC/Modern Robotics electronics should be considered to be flawed in some way, if they can not operate at their current specs for power.

                                FTC rules for robots to use up to 8 motors and 12 servos. The Modern Robotics Power Distribution module, Motor Controller , and Servo Controllers are all specify a operating voltage from 9 to 15 volts.

                                I don't believe it's impossible for a insufficiently charged or worn battery to contribute to problems with the USB communications systems in these Modern Robotics devices that communicate with the Android phones. However I do think this is highly unlikely to be the cause for most of us . IMHO, the Modern Robotics electronics should filter out voltage spikes, or dips and continue to operate with in the confines of their power specs, without failure of USB communications related to voltage spikes or drops. It should be assumed that there will be voltage spikes and/or brief dips and the electronics should filter these out, as it should be understood in the design that these things will occur.

                                In my case, I'm confident that my battery is neither worn, nor is it operating anywhere near the extremes of the of the Modern Robotics voltage specifications or total power consumption specs. I have been holding off on calling Modern Robotics, because I'm still trying to gather as much information as I can and also trying to make sure my cabling is secure and not causing the problems. My robot is not going to be in competition, I built it for my own education and tinkering, in hopes that over the next three years of my daughter's involvement in the FIRST FTC , I might be able to contribute to the local robotics club.

                                Comment

                                Working...
                                X