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

  • #31
    Originally posted by Tom Eng View Post
    Hi Folks,
    ...

    2. Battery Voltage - People have reported this to us, and we have seen this in our pilot testing and demos over the past year, but if the voltage on your main 12V battery drops too low (below 12V) we suspect that this can cause USB module disconnect issues. In my personal experience, the battery has to be fairly low before this seems to occur. The Power Module has a built in voltage regulator that takes the input voltage and steps it down to 5V for the USB modules. According to Modern Robotics, the Power Module can drop down to a fairly low value (at least 9Volts, which would be extremely low for a 12V battery) before the Voltage regulator would not be able to power the USB ports. However, we suspect that if your voltage level on your main battery is low, then there might be situations where a large current draw (for example, when the DC motors switch direction from forward to reverse) could drop the input voltage low enough that the Power Module fails to reliably power the USB ports.
    ...

    Tom
    So what you're saying is that if six or more motors and servos are running during end game, climbing mountains and doing pullups and such, the USB hub on the PDM is guaranteed to lock up or disconnect. In other words, the new platform, and the PDM in particular, is fatally flawed.

    Comment


    • #32
      Originally posted by Alec View Post
      The PDM serves no practical purpose in the first place. The PDM costs a lot of money for the sole purpose of taking up a lot of valuable real estate on the robot and for making cable management a nightmare.
      This hadn't dawned on me, but really is the PDM just a usb hub and power splitter? Are there any other electronics in there?

      Do the rules allow for a robot without one?

      Comment


      • #33
        Originally posted by korimako View Post
        This hadn't dawned on me, but really is the PDM just a usb hub and power splitter? Are there any other electronics in there?

        Do the rules allow for a robot without one?
        Apparently the PDM has a voltage regulator to deliver 5V to the USB hub (when it can). But really the USB hub should be powered from a different battery than the battery that's powering the 8 motors and 12 servos.

        Perhaps the ZTE phone battery is sufficient to power COTS USB hub. If not, then we can power the COTS hub with a dedicated battery.

        Comment


        • #34
          <RE01> Robot control modules are constrained as follows:
          • a. Exactly two (2) Android devices MUST be used in FTC competitions:
            • i. Exactly one (1) Android device must be used as the Robot Controller.
            • ii. Exactly one (1) Android device must be used as a component of the Driver Station.
            • iii. The only allowed Android devices are the ZTE Speed and the Motorola Moto G 2nd Generation

            phone. No other devices may be used as Robot Controllers or in Driver Stations in FTC
            competitions.
          • b. The Robot Controller Android device USB interface may only connect to the Core Power Distribution
            Module.

          Comment


          • #35
            Originally posted by mlwilliams View Post
            <RE01> Robot control modules are constrained as follows:
            • a. Exactly two (2) Android devices MUST be used in FTC competitions:
              • i. Exactly one (1) Android device must be used as the Robot Controller.
              • ii. Exactly one (1) Android device must be used as a component of the Driver Station.
              • iii. The only allowed Android devices are the ZTE Speed and the Motorola Moto G 2nd Generation

              phone. No other devices may be used as Robot Controllers or in Driver Stations in FTC
              competitions.
            • b. The Robot Controller Android device USB interface may only connect to the Core Power Distribution
              Module.
            Yes, but the rules are subject to change where necessary.

            Comment


            • #36
              We are having trouble where the phone cannot communicate with the CORE Distribution Module. Is there a way to determine if we have a cable or port issue? We can communicate with the computer when the USB is plug in, but not the robot. We also tried to create a new Configuration File and the phone indicated it did not see any hardware. We just finished charging all the batteries.... So "Dead in the water"...

              Thanks for your support!

              Comment


              • #37
                Originally posted by Alec View Post
                The PDM serves no practical purpose in the first place. The PDM costs a lot of money for the sole purpose of taking up a lot of valuable real estate on the robot and for making cable management a nightmare.
                Let me try to understand what you are saying. From my point of view, the PDM provides the following:
                - A power switch.
                - A powered USB hub.
                - A power distribution panel
                Without the PDM, I would need to get a powered USB hub, a separate power distribution panel which splits the 12V battery into several connectors, preferably fuse protected. If the power distribution panel doesn't have a power switch, I need to get a separate power switch just like in the NXT days. I could crimp daisy chained Anderson connectors but daisy chain has a lot of issues (e.g. if one connector failed, all down stream modules lose power. Also have to be careful on wire size being larger on up stream than down stream). From my estimation, it may take more combined real estate than the PDM if done safely and properly.

                Comment


                • #38
                  Originally posted by mikets View Post
                  Let me try to understand what you are saying. From my point of view, the PDM provides the following:
                  - A power switch.
                  - A powered USB hub.
                  - A power distribution panel
                  Without the PDM, I would need to get a powered USB hub, a separate power distribution panel which splits the 12V battery into several connectors, preferably fuse protected. If the power distribution panel doesn't have a power switch, I need to get a separate power switch just like in the NXT days. I could crimp daisy chained Anderson connectors but daisy chain has a lot of issues (e.g. if one connector failed, all down stream modules lose power. Also have to be careful on wire size being larger on up stream than down stream). From my estimation, it may take more combined real estate than the PDM if done safely and properly.
                  1. The Tetrix power switch costs about $5 with the FTC discount. Many teams already have these. Most teams will probably need them even with the PDM as the power switch has to be accessible to the FTAs and it may be tough to put your PDM in an accessible location.

                  2. You can get a variety of power distribution panels for $20-40, and more advanced ones with individual fuses for more money. I don't think we need a fused distribution panel as the battery has the same sized fuse already (and FTC was always run with only the battery fuse in previous seasons). Or as you say, you could make your own daisy-chained powerpole distribution, as teams have always done, as was recommended by FTC in previous seasons, and probably as many teams already have lying around from previous seasons.

                  3. Does the hub have to be powered? There seem to be many options here, including the simple 4-port unpowered hub that is distributed with the driver station to various powered ones on Amazon.

                  Regardless, I guess I share Alec's perspective: why are we all forced to use this and only this solution, which has never been stress-tested on FTC robots and which seems to be the source of a number of reported problems? It seems like there are many different off-the-shelf possibilities for achieving the same goal. If teams want to use the PDM, fine, but if teams want to use another solution, why should it not be allowed too?

                  Comment


                  • #39
                    I discovered today that the USB part of the PDM is nonfunctional for us. We had been having on and off problems for the past couple weeks, but today it was completely dead. I was able to use a Micro-USB hub to connect two modules at a time to the robot controller. I was having some trouble with the ZTE Speed, so I switched to my personal device, a Samsung Galaxy S5 running Lollipop. Samsung had fixed the USB issue before it released the update, so it worked fine. I still had problems recognizing the modules, even with stress-relieved cables and no moving parts (the same problems I had when I first got the stuff, and like many others are having), so using the ZTE Speed or the S5 did not really fix or create more problems. This shows that there is also a software problem involved. By bypassing the PDM, I was able to prove that there was a hardware problem, and by switching robot controllers and also by not getting any consistent results with either, I was able to show that their is a software problem. In some case, I would get the error that a particular device could not be found, in other cases I would get an error saying that the particular serial number was not a motor/servo controller, even though it was, and in a few cases I would get a long complicated message mentioning something about the USB (I don't remember what it was, but I'm pretty sure others have seen it as well; it is like 4 lines long or something). If this is not a software problem, and the software is perfectly written, then that means that Modern Robotics has produced a bunch of bad devices, and they don't know what they are doing. Our first tournament is October 31st, and with so many connection problems (taking around 10 or 20 times to get it to work correctly), it will be a long and arduous competition, as basically every match at least one team is guaranteed to run into errors, creating lots of delays or lots of forfeits of matches. With so many problems with the current system, I believe that the rules should be amended to allow for USB hubs and Anderson power pole splitters and whatever phone/OS a team wishes to use, if it makes the system work better, since the current technology is clearly not the best.
                    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


                    • #40
                      Low batteries unlikely to be main culprit

                      Originally posted by Tom Eng View Post
                      2. Battery Voltage - People have reported this to us, and we have seen this in our pilot testing and demos over the past year, but if the voltage on your main 12V battery drops too low (below 12V) we suspect that this can cause USB module disconnect issues. In my personal experience, the battery has to be fairly low before this seems to occur. The Power Module has a built in voltage regulator that takes the input voltage and steps it down to 5V for the USB modules. According to Modern Robotics, the Power Module can drop down to a fairly low value (at least 9Volts, which would be extremely low for a 12V battery) before the Voltage regulator would not be able to power the USB ports. However, we suspect that if your voltage level on your main battery is low, then there might be situations where a large current draw (for example, when the DC motors switch direction from forward to reverse) could drop the input voltage low enough that the Power Module fails to reliably power the USB ports.
                      Today's experiment for me was to get some data on this battery voltage issue.

                      I built two specialized Anderson powerpole jumper cables: one that inserts a small dedicated voltmeter across the line, and a second that inserts a potentiometer on the positive rail. I wired these in series from the 12V battery (pot first, then meter). I ran a simple bench-bot legacy motor controller opmode that started two motors going. Finally, I slowly cranked down the voltage going into the Core Power Distribution Module until things went awry.

                      The result: every thing was fine until just under 5.5v. This seems quite repeatable (n ~= 5). To recover from the stoppage, I had to both restore the potentiometer (of course) but also power-cycle the CPDM. Plugging and replugging the OTG connector at the phone, for example, was insufficient.

                      Now, I'm not here reversing the motors to try to cause big spikes in current draw, but I was surprised how low I could push the voltage before problems occurred. That leads me to think that 12V battery voltage levels are not the main cause of USB connectivity issues people are seeing, as the level required to cause failure is so low that a real bot would be rendered motionless from lack of motor power long before the USB connectivity would be affected.

                      Comment


                      • #41
                        Originally posted by mlwilliams View Post
                        <RE01> Robot control modules are constrained as follows:
                        • a. Exactly two (2) Android devices MUST be used in FTC competitions:
                          • i. Exactly one (1) Android device must be used as the Robot Controller.
                          • ii. Exactly one (1) Android device must be used as a component of the Driver Station.
                          • iii. The only allowed Android devices are the ZTE Speed and the Motorola Moto G 2nd Generation

                          phone. No other devices may be used as Robot Controllers or in Driver Stations in FTC
                          competitions.
                        • b. The Robot Controller Android device USB interface may only connect to the Core Power Distribution
                          Module.
                        I'm at least thinking I can save the ~$100 on the CPD if I want a second platform for 8 pairs of hands to work on... Jittery about purchasing more than necessary if there may be hardware issues yet...

                        Comment


                        • #42
                          There is a fatal flaw in the design of the PDM. Under normal FTC match scenarios, where during the course of the match the battery level has dipped below 12V (or below 9V in the case of a MATRIX battery), the USB hub on the PDM can lock up as a result of a normal voltage spike of the DC motors. When the USB hub locks up during the match, all the other core modules are disconnected from the Robot Controller phone until you manually power cycle the PDM after the match.

                          After power cycling the PDM, you may or may not be able to get the Robot Controller to re-connect to the core modules before your next match. Supposing you somehow managed to get the Robot Controller to re-connect to all of the core modules just in time for your next match. This leaves you with no time to do all the other things you needed to do to get ready for your next match.

                          A COTS USB hub (powered by a COTS 5V battery pack if need be) is immune to voltage spikes of the DC motors. A solution is to disallow the use of the PDM.

                          Comment


                          • #43
                            Varun,
                            So, to deconstruct and summarize this:
                            a) you used a micro-USB hub, not the PDM to distribute control from a phone (S5 or ZTE Speed). You still had intermittent problems with the robot controller phone consistently recognizing the attached Motor Controllers, etc.
                            (by the way, what micro-USB hub did you happen to use?)
                            b) I'm not sure I understand your point about proving there is a software problem. Can you please explain?
                            Thanks,
                            Martin
                            Mentor, Team 7593 TigerBots
                            ----
                            Originally posted by Varun Singh View Post
                            I discovered today that the USB part of the PDM is nonfunctional for us. We had been having on and off problems for the past couple weeks, but today it was completely dead. I was able to use a Micro-USB hub to connect two modules at a time to the robot controller. I was having some trouble with the ZTE Speed, so I switched to my personal device, a Samsung Galaxy S5 running Lollipop. Samsung had fixed the USB issue before it released the update, so it worked fine. I still had problems recognizing the modules, even with stress-relieved cables and no moving parts (the same problems I had when I first got the stuff, and like many others are having), so using the ZTE Speed or the S5 did not really fix or create more problems. This shows that there is also a software problem involved. By bypassing the PDM, I was able to prove that there was a hardware problem, and by switching robot controllers and also by not getting any consistent results with either, I was able to show that their is a software problem. In some case, I would get the error that a particular device could not be found, in other cases I would get an error saying that the particular serial number was not a motor/servo controller, even though it was, and in a few cases I would get a long complicated message mentioning something about the USB (I don't remember what it was, but I'm pretty sure others have seen it as well; it is like 4 lines long or something). If this is not a software problem, and the software is perfectly written, then that means that Modern Robotics has produced a bunch of bad devices, and they don't know what they are doing. Our first tournament is October 31st, and with so many connection problems (taking around 10 or 20 times to get it to work correctly), it will be a long and arduous competition, as basically every match at least one team is guaranteed to run into errors, creating lots of delays or lots of forfeits of matches. With so many problems with the current system, I believe that the rules should be amended to allow for USB hubs and Anderson power pole splitters and whatever phone/OS a team wishes to use, if it makes the system work better, since the current technology is clearly not the best.

                            Comment


                            • #44
                              ZTE Speed Phone KitKat system update science experiment

                              FTC Coaches,

                              I have had similar USB Connectivity as everyone else. Today based on another thread I decided to accept the system update for KitKat 4.4.4 that keeps wanting to be installed.

                              My phones worked but it is difficult to determine with limited testing whether or not this has made a significant difference. I did notice, however, and you may too that the driver station
                              did not display the two white gamepad icons automatically. I did note that after initializing them by the gamepad start button and A or B press that the gamepad ICONS then appeared.

                              My conclusion is that the USB functionality is impacted by the system update. So, are there any other takers on accepting the system update and then feeding back to this thread whether
                              or not they think it has helped?

                              Best regards and let's get this darn thing working together.

                              Marc Center, FTC8927 (veteran) and FRC10426 (rookie) head coach, ROyal Oak Bulldogs, Royal Oak, MI - Post #1
                              Marc Center FTC8927 Head Coach / Eng Mentor Royal Oak, MI

                              Comment


                              • #45
                                I keep my phones in airplane mode and do not normally connect to a wifi network. However both phones appear to be running KitKat 4.4.4 so they must of updated unbeknownst to me.

                                I've only see only have seen the USB problem 2% of the time. So that said, I personally can not correlate to Phone version, jostling, wireless vs wired code download, or a bad Power module.

                                All I could say is that I try to be consistent with how I update the RC code :
                                1) leave Power Module switched on and RC App running and waiting for an Opmode selection from the DS
                                2) swap the OTG usb to the RC phone with the computer usb cable and download the new code
                                3) wait for the old app to disappear and the new app to open on the RC phone
                                4) swap back the OTG cable and issue a "restart robot" command on the RC
                                5) select an opmode from the DS ( don't just hit the init button) start with an opmode selection

                                There have been a few times that I've experience the USB problem that caused me to reboot the phones and power off the power module.
                                My impression has always been that I may have deviated from the above steps. However I cannot tell if that was a coincidence or not.

                                What I can tell you is that after step 3, the RC always shows a USB connection problem on the screen. But of course it goes away after I swap back in the OTG cable and issue the restart as described in step 4.


                                Hope this helps,

                                David

                                Comment

                                Working...
                                X