STM32F429I-DISCO LCD and ETHERNET CONFLICT

[magore]

Posted on December 30, 2013 at 05:41

I think I found a serious pin conflict with the CPU being used on the STM32F429I-DISCO boards if you want use the TFT LCD and networking at the same time.

The TFT LCD interface function conflicts with the only two possible alternate function options for ETH_RM11_TX_EN pin (needed if you want a RMII network interface - its worse if you want an MII interface) Specifically PG11 conflicts with DB1/B3 on the LCD_TFT interface and PB11 conflicts with DB9/G5 of the LCD_TFT interface. There are no other alternates for the ETH_RM11_TX_EN pin! The remaining conflicts with the USB interface: OTG_FS_PSO, OTG_FS_OC, the L3GD20 device: INT1,INT2,CS - can all be worked around - but the ETH_RMII_TX_EN issue seems to be a show stopper. If you want a TFT/RGB display - you can not have a network. So much for an internet connected system with a display - this appears to be a CPU limitation by design ? Am I wrong ? My question is why did the designers do this when there were other free pins ? Is there any other option for using the DISPLAY and the RMII network transceiver ?

[magore]

Posted on February 11, 2014 at 05:01

I am answering my own post. I just found that EmCraft solved by making changes to the board - there can be 1 lsb worth of noise in the LCD display - however the changes work

See my recient post called: [DEAD LINK /public/STe2ecommunities/mcu/Lists/STM32Discovery/Flat.aspx?RootFolder=/public/STe2ecommunities/mcu/Lists/STM32Discovery/EmCraft%20supports%20the%20smt32f429i-disco%20with%20networking&FolderCTID=0x01200200770978C69A1141439FE559EB459D75800084C20D8867EAD444A5987D47BE638E0F]EmCraft supports the smt32f429i-disco with networking

[Tesla DeLorean]

Posted on December 30, 2013 at 07:27

The STM32's muxing fabric as long been an area of much frustration.

The STM32F429I-DISCO being a particular example of wide external buses for the external memory and screen. The design being incapable of DCMI and Ethernet, and seriously hobbled SDIO. I don't expect to see many break-out boards. Would suggest considering 176/208-pin devices if attempting to break out a fuller set of peripherals, and using the STM324x9I-EVAL as a starting point.