Small-footprint Due

ok RMII is in. Same board size (just reduced number of ground pins in the corners. Sending to PCB house today

sample boards ordered. 4 sets of:

  1. 2x MicroDue V1.1
  2. 2x WiFly shield (same 1.4x1.4" footprint)
  3. 2x u144 LCD shield (same footprint)
  4. 1x MicroDue 'motherboard' in Arduino UNO R3 layout - with USB jack, power input, SPI breakout, second CAN tranceiver (in addition to the first tranceiver that is on MicroDue already)

They arrive next Fri. I plan to populate using our T962A over that weekend and start testing.

Who wants some test boards?

Val.

Hi Val. A very good initiative. I would like to get involve in the testing of your boards, specially with the EMAC and CAN peripherals.
Regards!

p

My list:

USB Host on native USB port while externally powered (12V Vin).

Persistent RTC with breakout for backup battery.

3mm+ mounting holes in corners.

vanduino:
My list:

USB Host on native USB port while externally powered (12V Vin).

Persistent RTC with breakout for backup battery.

3mm+ mounting holes in corners.

thx. the only thing we don't yet have from your list are mounting holes. These would increase the size of the board by 30% (or result in the loss of ~10 pins) which is pretty significant. We are designing this board to be a pluggable MCU board for small production runs - similar to how Arduino Pro Mini or Spark.io boards are used...

But pls let me know if you have some creative ideas.

thanks,
Valery

In a kickstarter project the same discussion ended with that solution:

https://www.kickstarter.com/projects/214379695/micro-python-python-for-microcontrollers/posts/727834

gogol:
A fraction of the size of the original, with an estimated price, where you have not to think about letting the board in a project.

As someone who makes ARM-based Arduino compatible boards (but not with Atmel SAM3), I'm curious what price point you'd consider low enough to just leave the board in a project?

gogol:
In a kickstarter project the same discussion ended with that solution:

https://www.kickstarter.com/projects/214379695/micro-python-python-for-microcontrollers/posts/727834

nice! thx for linking!

mounting holes. These would increase the size of the board by 30%

How about having mounting tabs like the board on Gogol's Kickstarter link, they can snap off if not wanted.


Rob

re: Mounting Holes

I can appreciate keeping the size to a minimum, so I understand your hole aversion.

I work with mobile projects: vibration and temperature swings. The board would have to be removable, therefore pin headers. I haven't seen many pin headers come loose, but I worry about them over time. So I like the idea of a removable module that can be locked down. How about one bolt hole near the center, can you squeeze that in?

vanduino:
re: Mounting Holes
How about one bolt hole near the center, can you squeeze that in?

center hole will be tough as the entire center of the board is taken up by a huge SAM3X8E chip :wink: We tried to work BGA-144 in and gave up :wink: LQFP is a 1x1" package (with leads)

In our designs using Pro Mini today, we actually solder in the Pro Mini board to the main board. No worry about pins loosening and also saves spend on mating connectors.

Cannot afford to solder such a module to a motherboard that will go through option/revisions/updates in its life cycle. Has to be swappable. Mounting hole does not need to be large or dead center, just helps to keep pin headers from creeping loose.

You get the picture, I wont belabor it. Good luck with the project. Will be a welcome addition to the family.

quick update - first sketch!

playing pick-and-place with 0402 parts is a b*tch :wink:

will confirm USB upload next week and will be ready for small production run I think...

:fearful: Wow. Awesome video! One small foot-step for eMotorWerknstein, one giant leap for the Arduino Due based-kind!

p

Palliser:
:fearful: Wow. Awesome video! One small foot-step for eMotorWerknstein, one giant leap for the Arduino Due based-kind!

p

;-))

thanks P. the idea is to do a bunch of tests now for every periphery and pin. Ideally I would get ~100 of these made and shipped to people like you to test. The only snag is that we don't (yet) have an SMT lab here and contracting a 100-unit run will cost us ~$2,500 in fixed labor costs. Trying to figure this out now

guys - couple of questions as we are finalizing design:

  1. is CAN termination resistor (120R) needed on this board? If it's connected to fully configured CAN bus, one would have to desolder 0402 part from the board...

  2. As you saw in the video (Electric Motor Werks - VM MicroDue! - YouTube), the idea is to have a separate landing board in the UNO format. That board would provide micro-USB connection, ability to power from 7-12V source, etc. The question is what else we should provide on that board. For now, we have added a second CAN tranceiver - after all, most of our own uses will be automotive-related. What else?

  3. If you were really paying attention to the video, you could spot a few MicroShields on the PCB array. At the moment, we have a WiFi MicroShield and a LCD MicroShield. What else could be interesting?

Any other thoughts?

V.

valerun:

  1. is CAN termination resistor (120R) needed on this board? If it's connected to fully configured CAN bus, one would have to desolder 0402 part from the board...

  2. As you saw in the video (Electric Motor Werks - VM MicroDue! - YouTube), the idea is to have a separate landing board in the UNO format. That board would provide micro-USB connection, ability to power from 7-12V source, etc. The question is what else we should provide on that board. For now, we have added a second CAN tranceiver - after all, most of our own uses will be automotive-related. What else?

  3. If you were really paying attention to the video, you could spot a few MicroShields on the PCB array. At the moment, we have a WiFi MicroShield and a LCD MicroShield. What else could be interesting?

Hi Val.

My answers:

  1. Yes. A termination resistor 120R is necessary. Normally, a select jumper is used in series with it to select it or not.

2 & 3. Think about adding an Ethernet PHY as a separate module (similar to the one from elechouse) or soldered in your UNO landing board.

http://www.elechouse.com/elechouse/index.php?main_page=product_info&products_id=2229

See also replies 66 and 92 for schematic and example.

http://forum.arduino.cc/index.php?topic=142908.0

I am also adding an 2.8 LCD TFT to Due but need ebi.

Regards!

p

thanks P. will keep 120R on. jumpers will be tough - no space - so will have a trace one would have to cut (restoration would be possible by wire-jumping two small vias)

ETH PHY as shield is a good idea. Don't want to burden the landing board too much as it's likely going to be shipped with every MicroDue board and will be really sensitive to cost.

Unrelated question but hopefully some of you guys have perspective on this. We are exploring possibility to lease some in-house SMT assembly equipment for small / proto runs. Any recommendations? Has to have automatic pick and place...

P - I have read your thread on ETH in its entirety a few days ago. You are doing truly amazing work for the community (ETH, CAN, etc...).

Sadly I can't figure out sourcing for DM9161 chip...

quick question: how bad would it be to deviate from 0.1" pitch on external pins for MicroDue board? Right now those pins take up HALF of the board space by surface!

For example, we could use same spacing as XBee (2mm) and save some overall space and use space in the corners for mounting holes...