Go Down

Topic: Your latest purchase (Read 115 times) previous topic - next topic

retrolefty

That is dirt cheap. Tayda has them for .10 each which I thought was awful cheap.

http://www.taydaelectronics.com/crystals-resonators-oscilliators/20-000-mhz-20-mhz-crystal-hc-49-s-low-profile.html

Should be a pretty simple hack in arduino, recompile bootloader fro 20Mhz and make new modified entry into core's boards.txt file. Anything else I forgot?

Lefty

Nick Gammon

Just for fun I programmed via the ICSP interface with the ASCII table sketch, set to output at 9600 baud.

At 20 MHz it output (seemingly correctly) at 12000 baud which sounds right (9600 * 20 / 16 = 1200).

Then I put in one of the 24 MHz crystals that I also ordered (because dhenry claimed you could clock it at 24 MHz) and it output at 14400 baud (9600 * 24 / 16 = 14400). So on the face of it, you can run at higher speeds. Now I can't say for sure that it is working perfectly, but after running for 10 minutes or so, the ASCII table is still coming out.

Trap for young players: I initially saw no output, which I spent 15 minutes mucking around trying to fix, until I realized the sketch only outputs the table once. At 24 MHz that doesn't take long!

Nick Gammon


Coding Badly

Anything else I forgot?


millis and its ilk will be off.

This... http://arduino.cc/forum/index.php?topic=70475.0 ...cures millis (for 20 MHz).

I can probably rustle up a fix for micros (which fixes delay) if you need it.  For delayMicroseconds I suggest using _delay_us from AVR Libc.

Coding Badly

Now I can't say for sure that it is working perfectly...


I vaguely recall from the folks at avrfreaks that EEPROM writing is the first to "go" as the clock speed increases.

Go Up