[dorkbotpdx-blabber] ATmega48-TQFP32

[dave madden]

Is there anything wonky about the ATmega48 in the TQFP package?
Specifically, with regard to SPI programming? I have a (homebuilt)
programmer that happily talks to ATmega8, ATmega168 and ATmega328s in
the PDIP-28 package, but it doesn't get any response from the ATmega48
in TQFP. It's really disappointing, because the PDIP chips were
stupidly easy to get running.

Thanks,
--
David H. Madden - Mersenne Law LLC - www.mersenne.com
1500 SW First Ave. / Suite 1170 / Portland, Oregon 97201

_______________________________________________
dorkbotpdx-blabber mailing list
dorkbotpd...@dorkbot.org
http://music.columbia.edu/mailman/listinfo/dorkbotpdx-blabber

[Erik Walthinsen]

dave madden wrote:
> Is there anything wonky about the ATmega48 in the TQFP package?
> Specifically, with regard to SPI programming? I have a (homebuilt)
> programmer that happily talks to ATmega8, ATmega168 and ATmega328s in
> the PDIP-28 package, but it doesn't get any response from the ATmega48
> in TQFP. It's really disappointing, because the PDIP chips were
> stupidly easy to get running.

I've used the 48 TQFP in a prototype [1] and 168 in production, and
didn't run into anything different from any other AVR. The breadboard
version of the circuit used the DIP, and the programmer didn't change
when I moved to the TQFP prototype board. Both were programmed by
another ATmega32 using custom code, originally designed for programming
ATtiny15's and later 45's. I don't think the core code changed much at
all between the different iterations.

I'd do a detailed cross-check of the pinout, that's what usually bites me.

[1] http://pulsarprofx.com/PCBfx/main_site/pages/start_here/index.html
- 5 pics of round PCB