Page 1 of 1
128a1 clock caps
Posted: 21 March 2011, 8:57 AM
by FFMan
just about to test the new 128a1 and i see don't have any 27pf caps for the crystal connections as shown in the hardware reference.
I have some 22pf caps - will these be ok or is the value more critical than this ?
thanks
Re: 128a1 clock caps
Posted: 21 March 2011, 11:40 AM
by dkinzer
FFMan wrote:I have some 22pf caps - will these be ok or is the value more critical than this ?
22pF is the value most people use. The actual value to use is given by a formula that includes the parasitic capacitance of the wiring/printed circuit board which, of course, depends on the specific layout and construction.
In any event, I would guess that you won't see any difference using 22pF capacitors versus the 27pF that were suggested. The effect of the load capacitance being too small or too large is that the resulting frequency will be slightly fast or slow, respectively. I suspect that measuring the difference in frequency would be a difficult undertaking.
Posted: 21 March 2011, 15:32 PM
by FFMan
I put the 22pf caps in with the 14.7 crystal, but i can't get the IDE to talk to the processor.
I have applied 3.3v to the BOB and the red led is on.
I have a trusty MAX232 board that I have tested on a 328l and its ok so i transferred the connections to the 128a1 with the reset on pin 90 and the serial connections on pins 27 & 28. I am running the max232 on 5v so the gnds of the processor and max232 and joined.
However i can't get the ide to identify the device. the baud rate is set at 115200 - could this be an issue ?
thanks
Posted: 21 March 2011, 16:33 PM
by dkinzer
FFMan wrote:I am running the max232 on 5v so the gnds of the processor and max232 and joined.
That means that you're applying a 5-volt RxD signal to the ZX-128a1, something which should not be done. You must use a level converter between 3.3 volt devices and 5 volt devices if the 3.3 volt devices are not 5 volt tolerant (the xmega devices are *not* 5 volt tolerant).
FFMan wrote:However i can't get the ide to identify the device. the baud rate is set at 115200 - could this be an issue ?
Possibly, it depends on whether the IDE is set to use DTR for the ATN signal or to use an ATN character.
The ZX-128a1 was shipped with a "Hello, world" program installed. Unless you've re-programmed it already (probably not, given your comments) you should be able to see data being transmitted at 19.2K baud every half-second. The xmega devices use UART D0 for Com1, so the ZX-128a1 will be sending on pin 28 (D.3) and receiving on pin 27 (D.2).
Posted: 22 March 2011, 14:30 PM
by FFMan
looks like my 5v efforts have blown com1 on the processsor, pretty much as you would expect really. even with a 3.3v converter attached i see no 'hello world' output. if i loop tx and rx on the 3.3v side of converter i get echo back so i'm confident that is working ok.
would you expect the rx or tx of the processor to be damaged and is no output the logical conclusion (blown tx), or an inability to program (blown rx)
Posted: 22 March 2011, 15:04 PM
by dkinzer
FFMan wrote:would you expect the rx or tx of the processor to be damaged and is no output the logical conclusion (blown tx), or an inability to program (blown rx)
I wouldn't expect the transmit output to be damaged although the corresponding input to the converter may be since it was being driven with +5 instead of 3.3V. The ZX receive pin
might have been damaged, though.
In any event, you can't download to a ZX without both RxD and TxD working correctly.