Loading ...
Sorry, an error occurred while loading the content.

IV-17 Kinda working - part 2!

Expand Messages
  • shklaw75
    Hi Chris/group, It s been a while since i last killed the PIC processor i had on the breadboard, testing the IV-17 smartsockets, but i have finally got my act
    Message 1 of 8 , Dec 9, 2011
    • 0 Attachment
      Hi Chris/group,

      It's been a while since i last killed the PIC processor i had on the
      breadboard, testing the IV-17 smartsockets, but i have finally got my act together and got some prototype PCB's made for the IV-17 smartsockets.

      It coming together nicely, but still having a few issues, some new, some the same as when i had it setup on the breadboard.

      The decimal points are working, i messed up with the schematic and
      left a couple of resistors out, so have had to botch them onto the PCB to get the DP's to work properly.

      Static and scrolling messages are generally been displayed, but sometimes it takes a few send commands from the terminal program for them to come up.
      Another funny that is happening when i send a long message, (eg, the quick brown fox.....) the last character gets duplicated and a couple of extra characters (not part of the message) also get displayed. Is there a limit to the length of the message been scrolled?

      The other thing that sometimes comes up is when a static message is sent, instead of the new message been displayed, only one of the old letters is kept on the display and is a lot brighter than normal.
      You can see a video of what is happening here. http://www.flickr.com/photos/mrlaw/6455798555/in/set-72157625204679347/
      In the video, when i say A, i am sending ABCD and when i say W i am sending WXYZ.

      I have checked the UART settings in windows and the terminal programs, both are set to 38400, 8N1.

      The clock/date function don't work properly, Chris you mentioned a while ago this feature was not implemented in the version 13 firmware, does the chip you are selling have this functionality working?

      With regards to buying programmed uP's, are you selling the uP in the
      surface mount package as well?
      The boards i have had made up use the PIC18F25K22-I/SO, i was making the board as small as possible so when you chain them together, they don't become too much of a monster!

      Great work Chris on getting the IV-17 smartsockets out there, cant wait to see some pictures/videos of peoples kits been used!

      Cheers.
      Simon
    • fixitsan2
      It s good to hear from you again Simon, well done for persevering ! I can give you a few answers, hopefully it is good enough. Firstly I had been having issues
      Message 2 of 8 , Dec 10, 2011
      • 0 Attachment
        It's good to hear from you again Simon, well done for persevering !

        I can give you a few answers, hopefully it is good enough.

        Firstly I had been having issues with data transmission at 38400 baud. the only reason I went to that speed was just an intellectual challenge in order to reduce latency. I wanted to do away with CR/LF characters and all, and the high speed just give me too many challenges. I reduced the data speed down to 19200 and then 9600, rewrote some other parts of the interrupt handler and achieved very low latency with a data speed whch appears to be more tolerant of noise than before.

        I've set the baudrate so slow because some older uC's and PC hardware, as well as some GPS modules etc, all work at 9600max baud. by choosing that slow speed I still have headroom to speed things up if required.

        Suspect the stuck display yoiu sometimes see is due to the speed issues relating to the use of 38400baud with that code version. I have never had any issues since bringing the speed down.

        The clock function in the latest software works. In response to the clock display command the tubes display HH:MM and the seconds are displayed on the middle six DP's in binary format. The date is displayed as DD:MM, I assume we know what year it is most of the time !

        I can sell the 18f25k22 in SOIC format. However what you need to be aware of is that two pins of the three wire interface to the HV5812 need to be exchanged if you use a board designed for a previous version of this code, but that is just a case of cutting 2 tracks on your board and adding a couple of bridges.

        Hehe, now I come to think of it I have sold all my SOIC based boards so I don't know how I'll program another single IC yet, but that's trivial, it is doable. But I don't have any in stock. If you would like to post them to me I am happy to return them programmed, or i could buy some locally.

        Thanks for sticking with the project. I'll upload the latest user manual/build instructions to the files section so you can see what differences there are between the latest version and the version you have now.

        I think the dynamic scrolling is new to you (where you can scroll across just a limited number of the tubes and leave the rest showing the existing characters). You can now also address each group of four tubes collectively, such as setting the font of all tubes to be the same with a simpler message "GF1" will set all tubes of the first four tube module to font 1. (Used to be something like $F1111/CR/LF). it isn't a big saving on four tubes, but when you have 8 or 12 tubes it is a useful shortcut.

        You can also now force an enumeration cycle. Previously the tubes enumerated at power on, but if you have a large array you might need two 5V supplies. If the seperate supplies didn't reach the pic's minimum voltage at the same time the enumeration routine, which ran *only* at power up, could fail. Now it doesn't matter how or when each unit gets powered up, you just send "ID" and enumeration occurs, and the tubes display their position in the array, EG 1234567890123456 for 16 tubes. If you want to force an enumeration but without the display, just send "IN" (index, no display)

        I'm up to about 20.5k of the 32k of memory, so I hope to add some more effects now which work on a 'per group' basis. I once tried to make a fancy routine for effects so that no matter what size array you had yoiu could implement a sweeping effect which stayed synchronised. I might go back to looking at that, but in the meantime it should be easy to add effects which synchronise across fixed length arrays, like 4 tubes (making fancy FLW's and so on easy to implement)

        I've just about sold out of the first batch of kits now, I am very short of HV5812's so must order some more quick. I think I have plenty of boards and other components though.

        I'll be selling some kits _with_tubes_ this week or next if anyone is interested.

        Thanks again for the news Simon, I think you should be pleased with the software upgrade.

        Chris
      • shklaw75
        Hi Chris, Thanks for the update on the new software, its sounds like it will fix a lot of the issues i am having at the moment. The new features also sound
        Message 3 of 8 , Dec 10, 2011
        • 0 Attachment
          Hi Chris,

          Thanks for the update on the new software, its sounds like it will fix a lot of the issues i am having at the moment.
          The new features also sound cool, look forward to trying them out.

          Having the baud rate at 9600 sounds like a good idea, better for compatibility and more robust. That will also make it easier when i update the text to smartsockets program, the UART settings can stay hardcoded in the program, instead of the user having to select the rate (depending on whether they are using the IV-17 SS or older SS's).

          I have had a look at the schematic in the new assembly manual and compared it to the old schematic and from what i can see, the connections between the pic and HV5812 for the strobe and clock pins need to be swapped around.

          Another change is pin 24 (RB3) used to go to pin 13 (blanking) of the HV5812, now its not connected to anything and the blanking pin is connected to GND. Is this correct?

          Cheers.
          Simon


          --- In smartsockets@yahoogroups.com, "fixitsan2" <fixitsan@...> wrote:
          >
          > It's good to hear from you again Simon, well done for persevering !
          >
          > I can give you a few answers, hopefully it is good enough.
        • fixitsan2
          ... That s correct, although I connected the blanking pin of the HV5812 to RB3 pin in previous versions, it was never used. I had anticipated using blanking to
          Message 4 of 8 , Dec 11, 2011
          • 0 Attachment
            --- In smartsockets@yahoogroups.com, "shklaw75" <shklaw75@...> wrote:
            >
            > Hi Chris,
            >
            > Another change is pin 24 (RB3) used to go to pin 13 (blanking) of the HV5812, now its not connected to anything and the blanking pin is connected to GND. Is this correct?
            >
            > Cheers.
            > Simon

            That's correct, although I connected the blanking pin of the HV5812 to RB3 pin in previous versions, it was never used. I had anticipated using blanking to control display brightness by PWMing the blanking pin at one point.
            At the moment there is no blanking but i didn't like the strobing effects I was getting PWMing the blanking pin as a result of beat frequencies related to the muxing frequency
            i have always maintained the idea that I would like to add brightness control, and if I do, it will probably be a software implemented routine because I will be able to adjust the muxing time-space ratio to dim the display in a manner which i hope won't cause any nasty flicker or other artifacts.

            Chris
          • shklaw75
            Thanks Chris. Are these two features implemented in the current firmware? http://groups.yahoo.com/group/smartsockets/message/691 My PCB has pin 14 tied to GND,
            Message 5 of 8 , Dec 13, 2011
            • 0 Attachment
              Thanks Chris.

              Are these two features implemented in the current firmware?

              http://groups.yahoo.com/group/smartsockets/message/691

              My PCB has pin 14 tied to GND, if the features are not implemented will this need to be left floating instead of tied to GND?

              Cheers
              Simon

              --- In smartsockets@yahoogroups.com, "fixitsan2" <fixitsan@...> wrote:
              >
              > That's correct, although I connected the blanking pin of the HV5812 to RB3 pin in previous versions, it was never used. I had anticipated using blanking to control display brightness by PWMing the blanking pin at one point.
              > At the moment there is no blanking but i didn't like the strobing effects I was getting PWMing the blanking pin as a result of beat frequencies related to the muxing frequency
              > i have always maintained the idea that I would like to add brightness control, and if I do, it will probably be a software implemented routine because I will be able to adjust the muxing time-space ratio to dim the display in a manner which i hope won't cause any nasty flicker or other artifacts.
              >
              > Chris
              >
            • fixitsan2
              ... Hi Simon, there was no need for these functions once I spun my own boards and bought in ready made 35V PSU s, so they are defunct. Pin 14 was a test mode
              Message 6 of 8 , Dec 14, 2011
              • 0 Attachment
                --- In smartsockets@yahoogroups.com, "shklaw75" <shklaw75@...> wrote:
                >
                >
                >
                > Thanks Chris.
                >
                > Are these two features implemented in the current firmware?
                >
                > http://groups.yahoo.com/group/smartsockets/message/691
                >
                > My PCB has pin 14 tied to GND, if the features are not implemented will this need to be left floating instead of tied to GND?
                >
                > Cheers
                > Simon

                Hi Simon, there was no need for these functions once I spun my own boards and bought in ready made 35V PSU's, so they are defunct.

                Pin 14 was a test mode feature, if you're tied it to ground I guess you don't even use it ! Display testing is now done by sending either a single 'V' (version display) or "ID" (enumerate and display)

                The information about the current operational modes and the schematic is in the manuals in the IV-17 files section.

                Chris

                Chris
              • fixitsan2
                ... Having said no, I do have a version of code which does use these features if you want to spin your own supply. The code is good version 13 code which I
                Message 7 of 8 , Dec 14, 2011
                • 0 Attachment
                  --- In smartsockets@yahoogroups.com, "shklaw75" <shklaw75@...> wrote:
                  >
                  >
                  >
                  > Thanks Chris.
                  >
                  > Are these two features implemented in the current firmware?
                  >
                  > http://groups.yahoo.com/group/smartsockets/message/691

                  Having said no, I do have a version of code which does use these features if you want to spin your own supply. The code is good version 13 code which I produced for another group member recently.

                  Chris
                • shklaw75
                  Thanks Chris, I will stick with the latest version of code without these features and power the sockets with a separate supply. Cheers. Simon
                  Message 8 of 8 , Dec 14, 2011
                  • 0 Attachment
                    Thanks Chris,

                    I will stick with the latest version of code without these features and power the sockets with a separate supply.

                    Cheers.
                    Simon

                    --- In smartsockets@yahoogroups.com, "fixitsan2" <fixitsan@...> wrote:
                    >

                    > Having said no, I do have a version of code which does use these features if you want to spin your own supply. The code is good version 13 code which I produced for another group member recently.
                    >
                    > Chris
                    >
                  Your message has been successfully submitted and would be delivered to recipients shortly.