Pecan Pico 5 is in the Works

My friend Sven (DL7AD) has recently made some improvements to my original Pecan Pico 3/4 design and we were mailing our plans back and forth. Since my PCBs are already running low I decided to also prepare a next PCB order according to this new layout. Since low voltage designs also require a low voltage programmer I thought it might be a good idea to fill in the rest of the available 10 cm with a custom USBasp programmer (see After programming, this part of the PCB can be simply broken away and used for later projects or for sniffing at the serial port for debugging. The PCB order is not yet sent out, so if you see mistakes, please send me a notification. The change in layout needs a few pins to be re-mapped, but in general it should work with the original Pecan 4 software already. The GPS has no longer its own voltage regulator. We’re planning to switch it off in software. Also the MAX6 or 7 GPS can be optionally connected to the serial port or to the I2C bus. I’ll upload this to my github as soon as I know that the design works.


PecanPico5.sch PecanPico5.brd



10 thoughts on “Pecan Pico 5 is in the Works

  1. FYI, there is a *bug fix* in your Si4463.cpp code that handles 1 byte commands. This used to exist in the A1 and B0 versions of the chips, but all the chips I’m seeing lately and the ones that I got all are B1 versions. Not sure if this was fixed for not, but you may be able to remove that couple of lines of code (altho only one command you’ve implemented is a single byte). Also, following along the new hardware version. I guess you know if you can find them that a 4063 has been made that has no RX and is about $3 cheaper… at a fast past, it appears to be pin/function compat with the 4463… Nice work… what happened on the launch yesterday, saw that the -3 came down at approx 6p local…

  2. Thanks for your suggestions. At the time when I started the Si446x development, the Si4063 wasn’t yet available, but you’re right, that should be a pin compatible drop in replacement and it should work with just the same library code.
    The launch yesterday was very unlucky. First Tom (K5SAF) has rented a cylinder with big letters “HYDROGEN” on it, but after filling the first foil balloon we realized that it contained propane instead of hydrogen. Therefore we’ve aborted the launch of the big 1600g Hwoyee and we just organized two cans of helium for party balloons. When we started filling the foil balloons my precision scale was broken. Some wires in the power supply were ripped off in our rush packing activity. Therefore we had to fill the balloons rather approximately. I believe that at least 2 balloons must have burst because of over-filling so that it slowly came down. We’re still planning on launching the Hwoyee when we get real hydrogen and the predictions and weather is good.

  3. I found a little issue with your pecanpico4 firmware. The Si446x driver has the setup wrong for the GPIO on the SI chip..,. it actually works because you only ever check for the CTR to come back which is an 0xFF, but the setup of the SDO pin is actually wrong, it’s configuring it to be the DATA_RX function, not the SPI MISO… I think there was a decimal to hex conversion issue. You have it programmed to send a 0x11 to that register and that needs to be a 0x0B… The instructions for that register say that *dec* 11 is used to make the pin be MISO (SDO) and hence I think why it was entered at 0x11…

    is should be..
    const char gpio_pin_cfg_command[] = {0x13, 0x02, 0x02, 0x02, 0x02, 0x08, 0x0B, 0x00}; // Set all GPIOs LOW; Link NIRQ to CTS; Link SDO to MISO; Max drive strength
    SendCmdReceiveAnswer(8, 8, gpio_pin_cfg_command);
    //Serial.println(“LEDs should be switched off at this point”);

    Just figured I’d help you out with that change… I’ve just about got the code up and running on an STM32, (how I found the issue, I was getting only FF’s back from the SPI after executing the version above with an 0x11 where the 0x0B should have been 🙂

  4. Thomas,
    Was just looking at your new skinny design… you might want to check that TPS61200 circuit, the FB (feedback pin) isn’t the output I don’t believe, it goes ultimately to ground. The Vout pin is the output and I think you have those 2 swapped?

      • Hmm, not sure what you’re looking at. PecanPico5 (here) or my preliminary PecanPico6 ( ?.
        Both boards have the FB pin (pin 10) of the TPS61200 is connected to a voltage divider at both PCBs so that you can adjust the output voltage to any desired value.

        Note that both boards are a preliminary version. I have made some changes in the mean time and I’ll post them as son as I have he boards in my hands. One mistake that I fixed was the voltage divider to measure the battery voltage was connected to Vdd instead of Vbat.

  5. Thanks Thomas, I must have been looking at an interim version I just re-checked the pdf of the schematic and indeed it is correct… what might have happened is that at a glance I saw that the Vout pin (pin2) was terminated at ground at then end of the line… but when I looked at the circuit, you used a net to bring it to the FB pin, so all is well… Best of luck, looking forward to your 6 design… I’m kinda thinking of doing an STM32L1 design just for kicks now that I have aprs and 16 tone mfsk running on my stm32f3 board

  6. Recently, I am investigating the possibility of using ATMega328P, GPS, and Si4463 for my animal tracking project. The Google search leads me here. Thanks for sharing the circuit design of your work.
    I find a possible error in your schematic diagram (both PecanPico5 and PecanPico6). The Rx (Pin 0) and Tx (Pin 1) of ATMega328P are connected to Rx and Tx of the GPS module, respectively. I think they should be swapped (Tx to Rx and Rx to Tx).

    • Yes, this is correct. The RXD and TXD to the GPS are swapped. The PCBs for Pecan Pico 5 and 6 are just in testing that’s why I haven’t published the full design here on the website. You may have found some temporary designs somewhere on the web when I was exchanging some details with others. However there are a few more problems currently with them: The ADC input doesn’t measure the battery voltage, but the regulated voltage (which is nonsense) and also I have problems with the PLL of the Si4464 which doesn’t want to lock for some reason. Therefore please don’t just copy those designs and expect them to work. 5 + 6 are not ready yet!

  7. Hi Alan,

    Thanks for sharing your design. I’ve been looking at my own Pico tracker solution based around the TI CC430 on the 70cm band and was looking for a design on 2m, I found your design with the si4463 very informative. I also simulated the low pass filter to verify the results if your interested.

    Jason Ball (VK2VJB).

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s