Announcement

Collapse
No announcement yet.

Reset pins on the 7" Diablo

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

  • Reset pins on the 7" Diablo

    Is the reset pin on the programming header the same as the GPIO one on the 7" Diablo display? When I connect (via a 1K resistor) to the programming header the module resets continuously on high state but if I connect to the GPIO reset it works.

    Also, after a reset, the module does not mount the uSD card. Any ideas why?

  • #2


    From the datasheet It looks to be almost the same, are you sure you have the right pin?

    What do you mean by after reset the module does not mount the uSD card? Are you saying the mount command fails?
    Mark

    Comment


    • #3


      I think it was low a low power level from the USB connection through the host controller (that's how the screen was powered). When I connected a higher powered source the problem disappeared. I have no idea why it worked on the GPIO reset pin and not on the programming header though.

      Yes the mount command fails on the second attempt after the reset. I played around with some start up delays and finally got it working, but if the host resets via the USB (this causes the reset pin to go low for 100ms), the screen does not mount the uSD card. Weird...
      It could be voltage stabilisation. I haven't got my digital scope handy so I can't test this, but I will when I get hold of it.

      Comment


      • #4


        Now I am getting intermittent resets and funny touch behaviour. I really need to get the scope on this screen!

        Also the screen NEVER mounts the uSD card after a download when it is plugged into the programmer.

        I will try a different uSD card (the one you supplied) and see if that improves things.

        Comment


        • #5


          The 7" display draws pretty close to the maximum allowed on a USB2 port (500ma).

          It doesn't help that some computer manufacturers try to increase their 'green' credentials by limiting their entire USB 'farm' to 500ma, sounds more like a copout.

          The display will work at a lower voltage than the uSD card will, so if the uSD is failing, the supply voltage is probably marginal.

          Maybe a powered hub will help you?

          But for sure, check the 5v at the display end.
          Mark

          Comment


          • #6


            Hi Mark,

            I'm using a powered hub.

            I'm sure it's a power issue now. I connected my bench top 5V supply rail up to the system (3A) and it works fine.

            I think that USB is definitely on the edge supplying power to this 7" screen. It looks like the start-up inrush current prevents the uSD card from initially starting and for some reason prevents it from starting on subsequent mount requests. If the card is taken out and re-inserted it boots fine (while connected to the USB port). Putting in a delay to allow for voltage stabilisation (1.5 seconds) does not seem to let the screen mount the uSD either. That's what's really weird. I would have expected that to work. Perhaps if the card is inserted, there is a delay in the firmware allowing stabilisation on marginal voltages?

            With the 5V bench-top supply connected it boots every time, however if the bench top supply is connected and the screen is re-flashed via USB it DOES NOT mount the uSD card. I have to pop it out and re- insert it then it works.

            I am not actually going to be using this screen in the finished product so some issues booting the card are not really issues for me, but you guys might want to check this behaviour out because it is a little bit odd.

            Comment


            • #7


              Hmm, my 70DT running on my laptop doesn't display those issues.

              What brand and size of uSD card are you using?

              Can you measure the voltage and/or power consumption of your 70DT?
              Mark

              Comment


              • #8


                Brand of card is Trancend 2GB. I have also tried the brand supplied by you guys. The strange, weird, odd, and totally mind numbing thing is it started to mount BOTH cards now after a compile and upload?

                Voltage and Current.

                5.144V on the Screen.

                Current is a bit strange
                Inrush is about 900ma - 1A (that explains why it does not want to work wit the USB)
                Settles within about 6 seconds to 190mA.

                Isn't that a bit long?

                Some of the button behaviour (from the standard genie generated code so we can rule out code errors) is strange as well. At first I thought it was the power supply but I just tested it with the 3A supply, genie generated code and the uSD supplied by 4Dsystems. The buttons flicker on and off rapidly when pressed (momentary action user buttons). Weird or what? Maybe the 4wire resistive circuit is not working properly.

                Could this be a duff screen?

                Unfortunately my digital oscilloscope is in the lab which is a long drive from where I am now otherwise I could log this for you (I measured this with a Fluke so it's quite accurate).

                Comment


                • #9


                  OOPS!

                  Just checked the current again with auto range off and the meter set to DC not AC (schoolboy error)!

                  Ok
                  Start up is 527mA
                  Card loading peaks at 557mA
                  Runs at 536-537mA

                  Sounds more reasonable but perhaps a little bit on the high side?

                  Comment


                  • #10


                    Brand of card is Trancend 2GB.
                    Not a good brand, they don't work reliably in my Windows SPI card reader, and users report lots of issues with them. To Windows they 'brown out' (Appear to be removed and reinserted) when used for more than a few ms here and there.

                    If the power supply is inadequate you can often see the display resetting several times before it 'finally' starts up. If you use a good power supply the display comes up 'first time' and is 'ready to go' much quicker.

                    A USB port is supposed to be able to supply 500ma, so you may just be over the limit. However, some hubs use series resistors to limit the 500ma, so if your Hub is doing that then this may be contributing to the issue.

                    The AR1021 does some 'calibration' of the Touch Panel during power up, so if there is dodgy power during this time I can imaging that it might come up with the wrong values. Please check that your touch chip does have AR1021 written on it, although I'd be very surprised if it wasn't.

                    When I power my 70DT by a USB port in my computer the backlight will come on and off about 7 times before settling to on, this indicates marginal power and multiple display resetting. Despite that Touch works fine, as does the mounting of the uSD. If I power it from a USB wall wart the backlight comes on and stays on.

                    So I really feel your issue is power supply related, try ditching the HUB and powering directly, also make sure that the USB cable isn't one of those ultra thin ones, 500ma just wont get through one of them
                    Mark

                    Comment


                    • #11


                      Hi Mark,

                      I agree its almost certainly a power problem except for the touch behaviour...

                      As noted below, the touch behaviour is the same whether it is run from the USB hub or a bench top power supply capable of "clean" (unswitched) 3A of 5V. It does it with the transcend card or the card supplied by you guys. This is with "standard" unmodified genie code to eliminate any of my programming errors.

                      Touch chip is AR1021.

                      Funny thing is after one of the powered resets I did on the USB powered hub it seemed to work fine. The majority of resets came up with the "bouncing" button behaviour though.

                      BTW 2GB uSD's are getting hard to find and windows 7 have made it really hard to format them as Fat16 if they are more than 2GB. Any suggestions where I can get some "good" ones?

                      I will send a little video of the "bouncing" buttons to your email address along with the code from the other post.

                      Happy new year BTW.

                      Comment


                      • #12


                        Which PmmC version are you using?

                        Looks like I need to find some of the Touch utilities I did for the AR1020 and get them going for the 70DT and AR1021 to see whether anything can be gleaned from them.

                        Win7 and up will format 4gb as FAT (or FAT16), the default is FAT32, but it is easily changed in the 'File System' combobox.

                        For above 4GB you can use RMPet (in the tools menu) to partition the uSD card into 4GB (or whatever) and the rest.

                        Happy new year to you too
                        Mark

                        Comment


                        • #13


                          PmmC version is a good question I'll check and get back to you. I assumed it would be the current one because I only just got the screen.

                          I'll have a look at the utilities as well, but shouldn't it just work if I'm generating code from Genie?

                          I'll post the button video later today because I have to step out for a while.

                          I don't know why but the RMPet tool does not work properly when I partition the drive with it. I've read the stuff on it as well. Windows 7 just hates it when I format cards using it and I get errors galore....

                          The question is more substantial anyway because I will have to find a large supply of these cards (100's or 1000's) and an inexpensive automated system to rollout whatever ends up on them.

                          Comment


                          • #14


                            The question is more substantial anyway because I will have to find a large supply of these cards (100's or 1000's) and an inexpensive automated system to rollout whatever ends up on them.
                            Why don't you contact sales at 4dsystems dot com dot au and they can get you the right cards as many as you need.
                            Atilla

                            Comment


                            • #15


                              PmmC version is a good question I'll check and get back to you. I assumed it would be the current one because I only just got the screen.Just want to make sure it is correct.

                              I'll have a look at the utilities as well, but shouldn't it just work if I'm generating code from Genie?
                              I don't think the utilities I'm thinking of are publically available. Yes it should just work. The issue you are describing is on that could happen on the AR1020 in certain circumstances. But it is not supposed to happen on the AR1021. On the AR1020 it is fixed by a 'factory reset', but I have never needed to do a factory reset on the AR1021 (and the procedure is different to the AR1020)

                              I don't know why but the RMPet tool does not work properly when I partition the drive with it. I've read the stuff on it as well. Windows 7 just hates it when I format cards using it and I get errors galore....
                              Can you perhaps make a video of that, or post screenshots so we can get a feel for what is happening on your computer?
                              Mark

                              Comment

                              Working...
                              X