Arduino DUE Schematic Questions

The inductors are there (I believe) to prevent back-EMF. They're not strictly necessary, but it's good practice to add them in anyway. Capacitors C9 and C10 are bypass/decoupling capacitors, and again, whilst your circuit probably still will function without them, they prevent electrical noise and ensure that sensitive ICs that don't like voltage ripples function properly :P. In order for them to work properly, they need to be placed as close to the IC they're for as possible. I haven't thoroughly read the schematic, but I think C2 and C3 are for the power supply, to prevent voltage spikes and sudden drops. Note that they are polarized.

I'm not sure about the rest though. D1 confuses me. If it is a protection diode, shouldn't it be the other way round? Anyway, good luck on your board! If you need a place to get them made, Seeed studio were pretty good in my opinion (cheap, fast, decent quality). Check out my video review if you're interested:

nanohex:
... D1 confuses me. If it is a protection diode, shouldn't it be the other way round? ...

Depends on what you're protecting against. In this case I would assume reverse voltage (i.e. if you accidentally cross the wires from your poweer supply). D1 limits reverse voltage to -0.7V, so the circuitry will not be damaged. The fuse in the power supply will blow, but a fuse is easier to replace (mostly) than a whole DUE-board.

Greetings

Tom

Thanks for helping me! As i understood, those millions of Inductors and Capacitors are used for filtering. The reason why they put inductors/capacitors filtering on the VCC of PLL and VCC of AD conversor is to "decrease" noise on those readings, and noise on the internal PLL circuit, to prevend shifting too much the clock.

But the others questions remains unanswared for now...

I have not tried SeedStudios PCB services, but i always use the iteadstudios.com one, they are REALLY good also, and they provide FREE lead free and 100% E-Test (i don't know if that is "really" true, but it's what they said...), i will test with them some day! maybe with this board.. but i belive they use the same chinese guys to produce PCBs :slight_smile:

About D1, in case of "power suplying" a reverse voltage, current would flow direcly through the Diode, short circuiting the power supply, but, if you have a "good" power supply, the diode would blow and stop "rescuing" the circuit... is this correct?

About D1, in case of "power suplying" a reverse voltage, current would flow direcly through the Diode, short circuiting the power supply, but, if you have a "good" power supply, the diode would blow and stop "rescuing" the circuit... is this correct?

This is correct, and seems to me that D1 wired as it is is a mistake.

Those ckts with L1,L2 are clearly for filtering the particular lines. It's always a good idea to
have some noise filtering when VDDANA is derived from a digital buss voltage, ie +3.3V here.
The more so, the higher the resolution of the ADC. I don't know the function of VDDOUTM1,
however.

To understand the function of T2, you will need to ascertain what the purpose of the UVCC
pin on IC6 is. Possibly it's a sense pin, but who knows?

If you want to remove Q2, you need to replace it with an n-channel MOSFET, not a p-channel.

This is a very complicated design, with a lot of hidden meanings. The designers certainly
had their own reasons for including the ckts as they did.

Thanks for helping me! i took into consideration everything you said, and this is the rev. 1.0 of the main "Core" of the board. It's a simplified version of the Arduino DUE. I considered circuits present on Mbed, Arduino Mega, and also IOIO to build this revision.

As switching regulators are REALLY expensive where i live, i decided to change to a 7805 regulator. I have more separated circuits to build, and integrate a switching regulator would make the project really expensive...

Auto select power supply circuit was changed (I took into consideration the Ruggedduino schematic).
The 2x2 header was removed from Atmega uController

Here is the first revision of the schematic, if you can take a look and see if something is wrong, i apreciate that =]

First, if U2 is an adjustable v.reg, then you certainly have the schematic wrong - the
arrow means nothing to me, you need 2 resistors. I would rather use a part with fixed
output 3.3V voltage. Also, do you mean to have a DOT above C16?

Secondly, you might try to get the people from the Arduino DUE development team to
review your design, as they would know the specifics better.

Thirdly, don't buy 1000 boards initially. Buy a few prototypes and test it like crazy. I know
a guy with 5000 useless pcbs in his closet, with a small but nasty layout error.

I wrongly inserted the component... the one i really wanted was the 3.3V fixed one, that's why i grounded the ADJ, in fact I just didn't notice :blush:

How can I get in touch with Arduino DUE Team?

I'm going to order 10 pcbs (the minimum at Itead Studios), as there is no other place with quality that is cheap enough...

Thanks a lot for helping me with this!

The new version is already exported to pdf: Dropbox - Error - Simplify your life
(Just changed the LM1117 to the 3.3v fixed one...)

What about the DOT above C16? Should there be one?

You mean both caps connected to ground?
If yes, in the datasheet says to connect a capacitor like this to ground, to filter small noises, and is also in the Ruggeduino circuit: http://ruggedcircuits.com/AM010/am010.pdf

Is it right?

No, is the +5V v.reg supposed to be powering U2, or does U2 get power only via Q1?

It's weird! They are not connected, because altium doesen't accept (or I don't know how to doit) crossing connections! i make one, then the other one just converts to a passing by connections... so, i changed to different guide lines.
Thanks god you noticed it!

Searching for "tips" on how to place Bypass caps, i found an interesting article: http://www.akm.com/AppsNotes/layout2.pdf
And all the Arduino DUE caps, are placed with the same "How not to do" :~
It can be dificult to route everitying, but placing them as they are, is not causing any effect...
I will try to put them in the righ way

Yeah, bypass caps are usually placed as close to the chip Vcc pins as possible. Normally,
you want to do the power and ground trace layouts first, along with making space for
the bypass caps, and then add the other traces.

With so many pins, it's always a problem doing the layout. Your layout has an enormous
number of vias, and fewer are better. Auto routing always ends up producing zillions
of those darn things. With manual routing, you can play with things, move and rotate
components, and reduce the via count.

Also, with so many vias right next to the pads for the chip pins, you may get some solder
bridging when you solder the chip in place.

Also, you want to try an eliminate quick changes in direction of traces. There are a lot of
90-degree and even lesser angles, especially where you have vias. Doing layouts isn't always
the simplest job.

Hi,
This is not my routing yet =]
It's the Arduino DUE routing... I think that this is a really mess... just of looking at it, i would simplify things 30%.
But, it's not me who do it, so i can just learn with teir mistakes ^^

In that case, you can see the sort of problems you run into. So many pins, so many
traces stepping on each others toes. The one good thing is, most of the traces have
angled rather than right-angle corners - except for a lot passed through vias.

All my boards, i try to route with a minimum not-right-angle-edge, and also with the minimum of vias (when i used to do my PCB's at home, i hated to put jumpers, so i trained a lot basic routing...)

What I read in that article, was that there is no effect, putting the capacitor after the power supply... so, removing them wouldn't cause any effect, it's just like "not" having bypass capacitors on those pins... is it right?

Naw, I think you need to keep the bypass caps, whether or not they're wired in the best
possible way, per that other document. I'm sure they'll have some effect, as long as
located close to the pins.

What I read in that article, was that there is no effect, putting the capacitor after the power supply

No, that's not what it said. It said another way was better. And I'm not sure I believe it, either, depending on the length of traces involved. For the capacitor to be valueless, the trace length between the pin and the capacitors would have to be longer than the length to the actual power source (not the vias.) And the SAM3X is not an AKM converter, anyway.

I'm sorry, you are right... It just says the "right" way, it doesen't says that doesen't work...
Anyway, I'm doing with caps, when possible, i will do in the "right" way... =]