AVR-Chat

hi.
I have some problem with my printer port(I think so).
becouse when I programed my AVR chip for the second time i couldent 
program it again.
(in the condition that i wase made no chenges in my softwar & hardwar.)
the error seys that byte0=FFh byte1=FFh byte2=FFh  and it must be ... .
afte I installed the new windows I could program my chip just for 1 
more time & the error came back.
& now i dont know what to do.

At 03:50 PM 31/10/2007, you wrote:
>I have some problem with my printer port(I think so)

It usually kills your chip, parallel port programmers
are bad news even though some people seem to get them working
without too much trouble.

You have probable switched off the internal clock of the chip
and will need to feed a clock into xtal 1.


Regards

John Samperi

********************************************************
Ampertronics Pty. Ltd.
11 Brokenwood Place Baulkham Hills, NSW 2153 AUSTRALIA
Tel. (02) 9674-6495       Fax (02) 9674-8745
Email: john@ampertronics.com.au
Website  http://www.ampertronics.com.au
*Electronic Design * Custom Products * Contract Assembly
********************************************************

I have fried a couple AVRs using a LPT port SPI programmer.
	The problem went away when I first connect a ground wire between the 
PC and the hardware being programmed.
	The SPI cable will be inserted only after the ground is connected.

	Usually the fried chips identify as 00-00-00 or FF-FF-FF.
	They can't be recovered even using a parallel (HV) programmer.

	Mark Jordan

On 2 Nov 2007 at 6:36, John Samperi wrote:

> At 03:50 PM 31/10/2007, you wrote:
> >I have some problem with my printer port(I think so)
> 
> It usually kills your chip, parallel port programmers
> are bad news even though some people seem to get them working
> without too much trouble.
> 
> You have probable switched off the internal clock of the chip
> and will need to feed a clock into xtal 1.
> 
> 
> Regards
> 
> John Samperi
>

Someone already mentioned what might be wrong.  You may have
inadvertently switched oscillator sources by programming the fuse
bits.  I know ponyprog (last time I used it) would invert the fuse bits.

The other person also already mentioned the solution to the wrong
oscillator issue...but I figured I'd get a little more specific (I had
the exact same problem).  

If connecting an xtal (as would be normal for an externally clocked
MCU) to the clock pins doesn't fix it, you can setup another
externally clocked MCU and feed the output clock pin from the working
MCU to he input clock pin of the non-working one.  Yes, you'll need to
do that while the non-functional chip is also connected to the programmer.

Failing that, ummm, I suppose you could wire up the output of a 555
timer to the clock input pin.  It just has to be a higher frequency
than the I/O on the programmer.


--- In AVR-Chat@yahoogroups.com, "mooodij" <mooodij@...> wrote:

>
> hi.
> I have some problem with my printer port(I think so).
> becouse when I programed my AVR chip for the second time i couldent 
> program it again.
> (in the condition that i wase made no chenges in my softwar & hardwar.)
> the error seys that byte0=FFh byte1=FFh byte2=FFh  and it must be ... .
> afte I installed the new windows I could program my chip just for 1 
> more time & the error came back.
> & now i dont know what to do.
>

On Nov 1, 2007 3:19 PM,  <enkitec@gmail.com> wrote:
>
>         I have fried a couple AVRs using a LPT port SPI programmer.
>         The problem went away when I first connect a ground wire between the
> PC and the hardware being programmed.
>         The SPI cable will be inserted only after the ground is connected.

It should be obvious, but the ground is a required connection to the
target system.

On 8 Nov 2007 at 7:51, David VanHorn wrote:

> On Nov 1, 2007 3:19 PM,  <enkitec@gmail.com> wrote:
> >
> >         I have fried a couple AVRs using a LPT port SPI programmer.
> >         The problem went away when I first connect a ground wire between the
> > PC and the hardware being programmed.
> >         The SPI cable will be inserted only after the ground is connected.
> 
> It should be obvious, but the ground is a required connection to the
> target system.
> 

	Sure, the SPI cable has a GND wire, but I guess it wasn't making 
contact first and some static was damaging the IC under programming.
	I avoid that by using another GND wire, separated from the SPI 
cable.

	Mark Jordan