Announcement

Collapse
No announcement yet.

Some P1AM-PROTO Boards I'm Working On

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

  • #16
    I decided not to stuff the RS-485 board shown above in favor of a new and improved RS-485 board that uses the ADM2582E isolated RS-485 transceiver. It'll be full duplex with IO on screw terminals. Similar to the MKR RS-485 board but with a chip that runs from 3.3V and has 3.3V IO on the CPU side. Boards are being fabbed right now. Hoping to have boards by late next week then test and document over the weekend.

    Comment


    • #17
      Originally posted by bikerglen View Post
      I decided not to stuff the RS-485 board shown above in favor of a new and improved RS-485 board that uses the ADM2582E isolated RS-485 transceiver. It'll be full duplex with IO on screw terminals. Similar to the MKR RS-485 board but with a chip that runs from 3.3V and has 3.3V IO on the CPU side. Boards are being fabbed right now. Hoping to have boards by late next week then test and document over the weekend.
      something for thought... http://www.plctalk.net/qanda/showthread.php?t=124180

      Comment


      • #18
        Originally posted by kewakl View Post
        Haha. Let's call it a board that supports both full-duplex RS-422 and half-duplex RS-485 communications depending on how the four position screw terminal block is connected to the outside world.

        Schematics and board design and BOM here: https://github.com/bikerglen/p1am-pr...s485/boards/v1

        Comment


        • #19
          On the Halloween Prop controller shield, I got it playing CD quality audio and sending DMX-512 simultaneously but the ArduinoSound library spends about 2.5 ms of every 3 ms inside an ISR reading the next block of data of the uSD card for the I2S DMA output. The green trace is my 50 Hz task that launches DMX packets. The yellow trace is the DMX packet. The blue trace is the time spent inside the ISR while reading audio data from the uSD card. I had to mess around with the NVIC IRQ priority levels to get the DMX data to transmit while the sound was playing.

          Since the main loop code is blocked by the uSD/I2S IRQ for 2.5 out of every 3 ms, there's quite a bit of jitter as to when then 50 Hz task executes too. Unfortunately, I don't think there's a DMA version of the uSD/SD library for the Arduino. Still need to check out the Adafruit Zero DMA library and I2S libraries.

          A version of the P1AM-100 with a Cortex M4 and a QSPI uSD card interface would be killer about now.

          Click image for larger version

Name:	scope_4.png
Views:	99
Size:	37.0 KB
ID:	129779

          Comment


          • #20
            Moving to the SdFat library reduced the ISR time to 1.75 out of every 3 ms. I think I can live with that.

            Comment

            Working...
            X