shematic development board

Go To Last Post
9 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hello,
i want to make a usb development board with serial port to use usart,too.I like to program in PDI mode. somethnig like stk600 but only for xmega128a1. Is it possible to use codevision?do you know any shematic/circuit?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

That device doesn't support USB!

Leon Heller G1HSM

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Leon_heller I want USB only to program the avr

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
I like to program in PDI mode.

You have three options for programming the Xmega:
PDI Mode
JTAG
Bootloader

If you use the PDI mode you need three wires: Clock, Data, and Gnd. You would typically use a 6-pin Header that matches the Atmel "standard" ( AVR042: AVR Hardware Design Considerations ), and then use an STK600, AVRISPmkII ($34 USD), Dragon (Maybe...), etc., to program the Xmega.

I've not used IEEE 1149.1 JTAG, but it exists.

If you put a Bootloader into the uC, then it can load the program into the uC using whatever the Bootloader is designed to accept, (Serial Port, USB, FTDI-USB, SD Card, XBee, etc.) One would still need an Xmega programmer (PDI or JTAG), to get the Bootloader into the uC the first time.

There is an ever increasing number of Xmega development boards available. You might want to look at Gabotronics . Gabriel regularly posts here, and has sold a number of the Xprotolabs as well as his other systems. (Nice module!)

Boston Android also has a bunch of Xmega boards, but I've not personally ordered from them.

Atmel has its own XPlain board for Xmega eXploration.

Spark Fun sells an Xmega breakout board. You would put it on your own development board. It saves you the task of soldering the Xmega chip, which can be challenging to do by hand.

I tend to use my own Xmega development board, with USB and RS-232, shown below:

JC

Attachment(s): 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

I want USB only to program the avr

Consider doing what Atmel did - put both a AT90USB and an Xmega on the board and let the AT90USB handle the interface to the PC. 'course an FT232R on the board and a bootloader in the AVR may be a cheaper solution (but will require one time access to PDI to get the bootloader into the Xmega). Cheaper still (esp. if you have a USB-RS232 cable already) may be to simply put a MAX3232 on the board and use that with a bootloader (with the same caveat).

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

DocJC, how did you solder the xmega ( QFN ? ) chip( good work )?! Toaster oven ? :shock:

1) Studio 4.18 build 716 (SP3)
2) WinAvr 20100110
3) PN, all on Doze XP... For Now
A) Avr Dragon ver. 1
B) Avr MKII ISP, 2009 model
C) MKII JTAGICE ver. 1

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
how did you solder the xmega ( QFN ? ) chip

Very painstakingly...

I've been soldering since my first radio kits, which had tubes..., yet I found hand soldering the Xmega to be very challenging.

I had not actually smoked a chip in years, yet I fried two of my four Xmega prototype boards. What a learning curve. To be fair, I think I botched the 3.3 V linear regulator design. I had a couple large Cout caps and did not pay attention to their ESR. I think the regulator oscillated, was not in regulation, and over-voltage fried the first board. The other one, I believe, had a solder bridge under the chip. Fried it. No way to repair it on my bench. The other two have held up well ever since. Frustrating and expensive learning curve, but helps to keep one humble...

I have a headband magnifier and a jewelers loupe to see the pins and pads. I had never used liquid flux before soldering the Xmega, but it would have been impossible without it. Lots of threads talk about drag soldering, dragging the iron tip along the row of pins. Didn't work for me. Just bent the pins and made it even harder to line things up. I used the ATXmega64A1-AU. I believe it is a TQFP package. 100 pins, with 0.5 mm lead spacing. I tinned the pads. I then tacked down a couple pins on opposite corners of the chip. I then soldered each pin, individually. I then touched up a few bridges, because doing each pin "individually" was the goal, not the actual outcome.

My original plan was always to do a version 2 of the board. This first one was quickly designed to take on vacation when the first Xmegas hit the public market. Actually I had the board laid out before the chip was available. Ver. 2 was to be all SMD, (Mega on Ver. 1 board is through hole), add an SD card interface, use a FTDI chip instead of the FTDI module, change the on board op-amps, (for the ADC and DAC modules) a bit. The challenge in soldering the chip is one of the factors that has stalled my ever doing Ver 2, at least as of yet.

I am sure that it will get easier with experience, but gaining that experience can be a time consuming, expensive, and frustrating process.

That said, it is a great chip to work with, once you have a board in hand!

JC

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

It helps if you use the correct tip for drag-soldering - I use a mini-hoof cartridge with my Metcal system which holds a small amount of solder. Plenty of jelly flux is the secret. The hardest part for me is getting the chip positioned correctly, I use a cheap stereo dissecting microscope to tack down the leads on two opposite corners.

Leon Heller G1HSM

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Has anyone done this one?

http://xmega.mattair.net/docs.html

is it good?