×

SparkFun Electronics will be closed on November 26, 2020 and November 27, 2020 for the Thanksgiving holiday. We will begin normal operations at 9:00 AM Mountain Time on November 30, 2020.

×

Please see all COVID-19 updates here as some shipments may be delayed due to CDC safety and staffing guidelines. If you have an order or shipping question please refer to our Customer Support page. For technical questions please check out our Forums. Thank you for your continued support.

Dana Myers

Member Since: October 26, 2010

Country: United States

  • Where's the schematic link for this? Thanks!

  • Just looking at the HT16K33 datasheet, I note that the controller/driver is specified for 4.5-5.5V operation (typ 5V). So operating at 3.3V may be problematic; also I note that the input threshold is 0.7 * Vdd, so operating at 5V, a 3.3V pull-up may not be reliable, 5V pull-ups are required and the I2C drivers need to tolerate this.

  • The COM outputs can drive the 7 segment inputs, sure. But each digit is split into two because they're 14-segments, so it strikes me that each digit will required 2 COM outputs.

  • I've lashed the AS3935 'Ding Dent' breakout to an ESP32, using I2C and ESP-IDF, and have yet to see the I2C interface stop working. How long does it take in your experience for the I2C interface to stop working? I'm wondering if there's something specific to the Arduino IDE.

    If the MCU in use has a pin that can be used as a counter input and interrupt, then you can automatically calibrate the antenna at start-up, which is what I do with the ESP32.

    Out here in the SF Bay Area, we don't see a lot of electrical storms, but I appear to have accurately detected a storm passing-over near this morning; interestingly, there was a cluster of disturbers immediately before the lighting - which makes me think disturbers can also be natural electrical activity that isn't a ground-strike:

    5/30/20 3:18:02 as3935: disturber
    5/30/20 3:43:11 as3935: disturber
    5/30/20 3:47:21 as3935: disturber
    5/30/20 3:48:20 as3935: disturber
    5/30/20 3:49:48 as3935: disturber
    5/30/20 3:51:06 as3935: disturber
    5/30/20 3:51:57 as3935: disturber
    5/30/20 3:53:07 as3935: disturber
    5/30/20 3:53:25 as3935: disturber
    5/30/20 3:55:48 as3935: disturber
    5/30/20 3:57:17 as3935: disturber
    5/30/20 3:58:18 as3935: disturber
    5/30/20 3:58:22 as3935: lightning: 5
    5/30/20 3:59:48 as3935: disturber
    5/30/20 4:00:09 as3935: lightning: 5
    5/30/20 4:00:40 as3935: lightning: 5
    5/30/20 4:00:41 as3935: disturber
    5/30/20 4:03:47 as3935: disturber
    5/30/20 4:06:03 as3935: lightning: 5
    5/30/20 4:15:28 as3935: re-calc: 20
    5/30/20 4:16:13 as3935: disturber
    5/30/20 4:17:12 as3935: re-calc: 24
    5/30/20 4:23:05 as3935: re-calc: 63
    

    So - I'm most interested in what the I2C issue is, is it a known erratum for the AS3935?

    Cheers, Dana

  • Technically, this isn't over-clocking, since the chip datasheet has an FMAX of 320MHz. The PLL can produce up to 384MHz, but I don't have high hopes that'll be stable. It's also possible to run the part at a very low clock rate, with proportionately less power consumption. I've tested it at 2MHz.

  • Given that the FE310-G002 datasheet specifies FMAX of 320MHz and my Red-V (R8 100ohms) appears to be stable at 320MHz, perhaps the product page could replace "150MHz" with "320MHz"?

  • With R8 changed to 100 ohms, it's easy to change the clock rate from the HFXOSC value of 16MHz.

    #include <metal/machine.h>
    
        long new_cpu_clock = metal_clock_set_rate_hz(&__metal_dt_clock_4.clock,
          320000000L);
    

    Though this function appears to hang every now and again - I believe it's possibly because the metal code is not observing the required 100uS delay before checking PLL lock in the Freedom Metal code.

    The chip warms up quite a bit when running at 320MHz :-)

  • I'll give it a try as soon as I have the 'fixed' Red-V (and also re-work the 'first run' one). Fortunately I've figured-out how to get Freedom Metal to program the PLL (it bricks the 'first run' one until I jumper across R8 temporarily to program it :-))

  • The SparkFun 'Features' for this board list clock rate of 150MHz but the -G002 datasheet lists FMAX of 320MHz - I don't see anything to indicate these chips are binned into speed grades. Is the FE320-G002 on the RedBoard capable of operating at 320MHz?

  • That's pretty much how pad-per-hole protoboards work. I don't think SparkFun sells wiring pencils (http://en.wikipedia.org/wiki/Wiring_pencil), but something like a Verowire pen works well for me for digital and low-current analog wiring. Discrete components are tack-soldered into the holes with a "solder tail", and I use the wiring pen to run a few turns around each desired pin/lead, then solder them down to make them permanent.

No public wish lists :(