FTDI Chip on chipKIT Max32

Created Fri, 22 Jul 2011 13:36:07 +0000 by shitake

shitake

Fri, 22 Jul 2011 13:36:07 +0000

Hi all,

as far as i know the Pic32 has USB,Ethernet & CAN , why would there be a need to break them out to a seperate board when all the power is already inside the mcu ?? What i also dont get is, why is there an extra ftdi chip used when the pic32 has the capabilities to serve as a USB host?

Can someone please clear this up for me. Because i cant understand it. Thanks for your help. Cheers.

rtestardi

Fri, 22 Jul 2011 14:16:52 +0000

The main advantage of the FTDI chip is that when the MCU is reset, the USB device exposed by the board does not disappear and reappear, from the host perspective. It allows the normal Arduino mechanism of the host opening the USB COM port to force an MCU reset and cause immediate entry into the bootloader in preparation for a new download.

With all that said, we've managed to get boards without the FTDI chip (CUI32, UBW32, etc.) working quite nicely, though these require (for now) a manual reset and an explicit pushbutton to cause entry into the bootloader.

I think the bottom line here is that the FTDI chip was seen as a quicker route to a "more compatible" board, but it is not necessarily a required piece long term.

KM6VV

Fri, 22 Jul 2011 17:26:51 +0000

The FTDI chip appears to be well worth it, just to keep it more compatible, and eliminate the manual resets. Good decision!

I'd also think that it would allow the use of that second USB port if desired.

Alan KM6VV

GeneApperson

Fri, 22 Jul 2011 19:07:28 +0000

Independent of whether including the FTDI chip was a good idea or not, there needed to be an expansion board for the Max32 to keep the cost of the basic board down and to preserve the Arduino Mega form factor.

The PIC32MX795 provides an ethernet MAC. It doesn't contain the PHY. There needs to be the PHY chip plus magnetics and connector to complete the ethernet interface. The situation is the same for the CAN interfaces. The PIC32 part contains the CAN controllers, but they need an external transceiver to connect to the network wiring. For USB device operation, there doesn't need to be anything more than the connector. For USB host operation a bit more circuitry is needed.

All of these interfaces need connectors. Within the form factor of the Max32 (i.e. Mega board) there is no place where those connectors can go. A second board is needed just to get the room on the board edge to place the connectors (plus the board area for the additional electronics). And, of course, if the shield maintains the same form factor as the Max32, the only place those connectors can go is on the left edge where the power and USB connectors are on the Max32.

Gene Apperson Digilent

shitake

Mon, 25 Jul 2011 00:59:14 +0000

@GeneApperson: Thanks for your really great explanation. I found out about the "problem" with the phy and ordered the one from national which is recommended in the Datasheet.

Cheers.