ATMEL Mega1284P evaluation board avalible

I believe the Library is wrong. I have the same socket top & botom, they mirror each other, both have CS in the wrong place.

:sleeping:
.. I ordered this (with an assumption I will use your eagle lib):

  1. the first picture shows "SD mode" 4bit
  2. the second "SPI mode"
  3. the third - maybe the corrected version (double check)
  4. new lib (the original one has the numbering wrong and the pads were count from left to right, top view) - double check as I am doing this while I sleep..
    PS: the case gnds are shown intentionally as 4,3,2,1 :sleeping:

PS1: BTW, that is not an error in the lib - top side in Boston means bottom side in Auckland.. :slight_smile:

microsd_spi.jpg

microsd_old_new.jpg

MICROSD_PITO.lbr (2.7 KB)

Thanks Pito, I'll see if I can get that in and get some new boards ordered today.

..wait a while before you order new boards.. it could be more to fix there.. P.
PS1:
-what I generally miss are the pullups on the sdcards pins like CS, datain,..
-pullups on the carddetect, writeprotected..
-add the sense (carddetect) on the microsd as well (and the pullup)
-add pullups on any RX, RX1 and alike lines which may float when the cable is not connected
-I would not use A7 for WP of sdcard1 but for cardetect (sense) of the usdcard
..maybe more
-try to load the power supplies with a few hundred mAps (when applicable) and see if it works stable (when inserting sdcards, when formating sdcards, etc)

  • I would add few 10-20uF ceramic caps to the sdcards vcc33

PS2: did you ever put a shield into this Mother of all boards?
If I were you I would assemble one more board and ask somebody to use it intensively for 1-2weeks in some harsh environment (various shields etc.). Then I would consider the second batch.. Moreover, you planned only one sdcard, did'n you? :slight_smile:

Pito, how about the internal pullups?

pullups- ..if the software sets them up then ok.. what if not?? I do not know, what I am suggesting to you is to think on all possible variants or potential issues, before you order new boards.. simply do use the board as-is, try various applications and you may see what is missing..
The new sdfat supporting more than one card will be not ready within 2 weeks so you do not need to have two cards there yet..P.

@pito,
I am doing a little re-arrangiing, wiring the top SD & uSD in parallel, adding a 2nd SD to the bottom & wiring it in parallel to the bottom SD.
Each side will have its own chip select, which is seperate from the shield slave select.

-what I generally miss are the pullups on the sdcards pins like CS, datain,..

The 2 chip selects have 10K pullups to Vcc33 for the SPI signal buffer chip. I am not putting any additional load on the MISO, MOSI, and SCK lines.

-pullups on the carddetect, writeprotected..,..

Will rely on the internal pullups for that. It is just a switch closure, no active signal behind it.

-add the sense (carddetect) on the microsd as well (and the pullup),..

I have removed the WP signal and instead used A6 and A7 as the top and bottom card detect/sense
I have read (perhaps prior Skyjumper posts?) that reading that the software can write the card even if it selected.

-add pullups on any RX, RX1 and alike lines which may float when the cable is not connected,..

I don't see the need for that. If your cable is not there, you will not get good data anyway. And then its just exrta load when you want to use that pin with a shield.

-I would not use A7 for WP of sdcard1 but for cardetect (sense) of the usdcard,..

See above

..maybe more
-try to load the power supplies with a few hundred mAps (when applicable) and see if it works stable (when inserting sdcards, when formating sdcards, etc),..

3.3V Reg is good fror 400mA, not worried about that one.

  • I would add few 10-20uF ceramic caps to the sdcards vcc33,..

Alread has a 100uF cap on the input, 100nF on the output, and 100nF at the buffer chip

PS2: did you ever put a shield into this Mother of all boards? ,..

Haven't used it with one (I don't have any) but we used a adafruit shield to hold the header pins in place, so alignment is correct.
I also used Blink to toggle all outputs, leading to the revised pins_arduino.c for this board.

If I were you I would assemble one more board and ask somebody to use it intensively for 1-2weeks in some harsh environment (various shields etc.). Then I would consider the second batch.. Moreover, you planned only one sdcard, did'n you?,..

Am assembling 8 or 9 more tomorrow, will then have some to play with besides the 2 I built to start for debugging.
They will not have the uSD shield. Rev 18 boards will have that option.

.. RX pullups - worst case scenario - imagine an application which, after a reset, checks (within a specific time interval) whether it gets in a character via usart, or not. If yes it jumps to a bootloader/monitor/shell, when not it continues to the application.. A logical 0 at RX means a new character comes in then.. ]:slight_smile:

Ok, here it is! The "Bobuino", an ATMega1284 based design which expands on the Arduino Duemilanove design with more IO pins, RealTime Clock with battery backup, Secure Data Card socket, and RS232 driver.

After a hard of day of soldering & soldering with my friend Skyjumper and his reflow oven and the "CrossRoads Electronics team" (my engineer wife, my son and I :smiley: ), I now have 10 Bobuinos ready for sale. Qantity is limited to the 10 pictured here; price for the first 10 is $80 each. If there's enough interest, I'll build some more.
For that $80, you get all these fanastic features plus USPS flatrate shipping to your US location (need to get some electrostatic bags too):

  1. Atmega1284P processor featuring 4 Kbyte EEPRPOM, 16 Kbyte SRAM, 128Kbyte Flash Memory; 2 hardware UARTS; 32 IO lines; and 40 pin DIP package, easily replaceable when you blow it up. Will be preprogrammed with a bootloader, and I'll also send the modified pins_arduino.c and Sd2PinMap.h files for use with the fat16lib's SdFat library. Use them as is, or cut & paste the relevant sections into your files.
  2. Female headers that accept standard shields. All shield pins are accessed normally: Vin, +5, +3.3V (50mA max), Reset, Gnd, D0-D13, A0-A3, A4-A5 (** see #24 below), and 2 addtional analog pins, A6-A7 (** see #25 below). D10-11-12-13 are the SPI pins. D10 is free to used as the Shield's SPI ChipSelect/SlaveSelect line. Activity "L" LED connected to D13 via series 1K resister (about a 3mA load when High).
  3. Jumper to select if RTC Square Wave output goes to D6 (INT2) or D31 (PD7). SqWave is not connected to anything when no jumper is installed.
  4. DS1307 Real time clock with 13.768KHz crystal oscillator, connected to dedicated I2C pins SCL/SDA, and with 10K pullup resistors, for access to Year, Month, Day of Month, Day of Week, Hours (24 hour format, or 0-12 with AM/PM indication), Minutes, Seconds. Square wave can be off; or the open drain output with 10K pullup resister can be software selected to output 1Hz, 4kHz, 8kHz, 32kHz. CR2032 3V Lithium batter is installed in a Tyco battery holder on the back of the board.
  5. 5.5/2.1mm Barrel jack for wallwart. Will include 7.5V, 1000mA switching regulator for an additional $5, or use your existing 7.5VDC to 12VDC wallwart. Input DC goes thru a reverse polarity protection diode to become Vin to the 5V regulator and to the header shield.
  6. 5V, 800mA linear regulator, mounted on expanded thermal pads (connected top & bottom).
  7. 3 holes (shown pictured with 0.025", 0.1" pitch pins for testing) to connect to external Vin, Gnd, +5V. Option: remove the regulator and install,or connect to, a leaded part with heatsink (MC33269T-5.0, TO-220 package) if planning to supply power from up to a 20VDC source, or your favorite switching regular to supply 5VDC.
  8. 3 pins with Jumper to select 5V from regulator or 5V from USB to supply the rest of the board. If no jumper is installed, 5V must come in from the Analog header (#2). 5V Power Indicater LED "PWR" located near the "L" LED.
  9. FTDI FT232RL chip to connect Rx0/Tx0 to USB connector. Rx0/Tx0 are also D0/D1 and connect to the shield. 4 holes adjacent are for 0.025"/0.1" pitch pins, intended to be used to "bitbang" a bootloader in using the adjacent FT232RL chip. (Pins shown pictured for testing only, jumper cable to ICSP header needed). Rx and Tx activity LEDs located adjacent to the chip.
  10. USB-B Mini connector, connects Rx0/Tx0 to PC via FT232RL chip. USB's Vbus (5v) goes thru a 500mA polyfuse.
  11. 3 pins with jumper to connect, or not, DTR from FT232RL chip to Reset capacitor.
  12. Reset switch.
  13. ICSP header pins for connecting AVR ICSP type bootloader, such as from MDFLY.com.
  14. Holes for connecting 315/434MHZ type Rx or Tx module (one or the other), plus an antenna wire. Up to user to confirm that signals agree with their part. If not, an adapter cable will be needed to match the pins (Vcc, Gnd, Antenna, and D9) to an offboard module. Intended to be used with VirtualWire.
  15. 3.3V, 400mA regulator with thermal pad area (top & bottom), supplies the SD Socket only.
  16. 74AC125 3.3V driver chip to convert ATMega SPI pins to 3.3V. D30 (PD4) is dedicated to the SD socket chip select.
  17. MAX232 driver chip for Rx1/Tx1.
  18. Jumpers to connect MAX232 IO to Rx1/Tx1. Rx1/Tx1 are also D2/D3 and Int0/Int1 and connect to the shield.
  19. FCI push-push SD card socket (micro-SD with appropriate adapter). Works great with Fat16lib's SdFat library. D30 must be set as ChipSelect.
  20. 16 MHz crystal oscillator and 22pF caps.
  21. RS232 header. Connect to DB9 Pins 2, 3, and 5 (Gnd) for RS232 operation. Requires jumpers (see #18) to be installed.
  22. JTAG header.
  23. 8-pin IO header, connects to PC0-7. Accessed as D22-D29. PC0 and PC1 are also SCL & SDA.
  24. Jumpers to select function connected to shield header pins A4/A5. One side connects A4/D18 and A5/D19, the other side connects SDA and SCL. Nothing connected if no jumpers are installed.
  25. Female Analog header for shield. Extended 2 pins for A6/D20 and A7/D21.

(picture of board with arrows to features)
(picture of 10 boards, boards with ATMega installed have been bootloaded and have been checked out, the others have started testing and will ship less the 2 jumper fields noted unless requested otherwise).

PM me for details where to send Paypal payment and address to send card to. All cards will have all IO lines tested, RS232 serial interface tested, RTC tested, and SD card tested. I expect to be bootloading and downloading sketches/testing all this week (and install a component or two more, think we were short a 16 MHz crystal and 4 header shields when we packed up).

(wow, that took a while to type up!)

Thanks!
Robert

..can you show us Skyjumper's reflow oven, pls?

Delicious... GREAT WORK!!!

excellent looking board
bet you're well pleased with self!

@pito,
I have some pictures, one might be the boards cooling in the oven. Will check when I get home.

@buulindo, mmcp42,

We sacrificed a bare board to make a jig to hold the thru hole connecters straight for soldering, makes the board look very sharp having everything all lined up nice & neat.
I am very pleased with the way the boards turned out. The FTDI chip has turned out to be a total pain in the ass, requiring touch up on every board. I think next go round will look into maybe putting paste down and no part for the oven reflow phase, then use hot air rework tool to clean up any shorts between pads, and then put the chip on. That's definitely not a hobby friendly package.

@pito,
Turns out I Don't have pictures - only a shortcut to a desktop folder! Blew that one. Will get them soon.

CrossRoads:
@pito,
Turns out I Don't have pictures - only a shortcut to a desktop folder! Blew that one. Will get them soon.

Apparently that's my fault, sorry :frowning:

I'll shoot a pic of my oven and post it.

Hello Robert!
What a beautiful card!!! Awesome :astonished:
Very very nice work. :wink:

Thanks Leo. I translated another post, looking for "core files".
I have posted the changes I made to boards.txt and pins_arduino.c for the D0-D31 assignment to pins that I used, and to a SdFat library file also.
If folks there want to use the Sanguino files for the 644 that might be more straightforward.

hi, i'm erik and been following the great work in this topic for a while because this board is exactly what i need, please tell me how to buy a board.
erik

Eric, read the last paragraph of CrossRoads's post.

CrossRoads:
PM me for details where to send Paypal payment and address to send card to. All cards will have all IO lines tested, RS232 serial interface tested, RTC tested, and SD card tested. I expect to be bootloading and downloading sketches/testing all this week (and install a component or two more, think we were short a 16 MHz crystal and 4 header shields when we packed up).

Moderator edit: Trimmed the unnecessarily large quote.

Thanks Erik.
Let me investigate shipping costs to Europe.
Do I need to include a piece of paper in the shipment saying it is Not Leadfree?

Robert