| Author |
Message |
|
|
 Posted: Jul 21, 2008 - 04:05 PM |
|
Joined: Dec 26, 2007
Posts: 3
|
|
Hi All,
I am newbie. I have simple question.
For XMEGA, Is I/O pin 5V tolerant or need level shift ?
dbird |
|
|
| |
|
|
|
|
|
|
Posted: Jul 21, 2008 - 04:28 PM |
|
Joined: Feb 19, 2001
Posts: 25922
Location: Wisconsin USA
|
|
Not a bad question. AFAIK, the answer is "not 5V tolerant".
There is no mention of protection diodes in the datasheet that I can seel, but the Absolute Maximum Ratings table imply that they are there. A 5V level on an I/O pin with Vcc of 3.3V violates the Absolute Maximum level.
Lee |
|
|
| |
|
|
|
|
|
|
Posted: Jul 22, 2008 - 07:47 AM |
|
Joined: Dec 26, 2007
Posts: 3
|
|
|
|
|
|
|
|
Posted: Oct 24, 2011 - 11:14 AM |
|
Joined: Dec 06, 2010
Posts: 12
|
|
| I am sorry that I am asking the same question again, and I still couldn't answer and thank in the other subject to people who helped me with their suggestions. I want to verify again, are ATxmega series 5V tolerant?, especially ATxmega16D4. I also saw that in the datasheet, absolute maximum ratings say that it is not 5V tolerant. Is there anything changed since 2008, do I miss something on the datasheet? |
|
|
| |
|
|
|
|
|
|
Posted: Oct 24, 2011 - 11:51 AM |
|
Joined: Sep 05, 2001
Posts: 2499
|
|
|
aerkol wrote:
Is there anything changed since 2008
No 5V tolerant and no CAN planned in future.
Peter |
|
|
| |
|
|
|
|
|
|
Posted: Oct 24, 2011 - 06:02 PM |
|
Joined: Apr 01, 2003
Posts: 186
|
|
|
|
|
|
|
|
Posted: Oct 24, 2011 - 07:26 PM |
|
Joined: Aug 23, 2008
Posts: 31
|
|
| It is not necessary for a plain 3.6V controller system to be 5V tolerant. This is a feature of many peripheral chips with only some interface pins to connect to a host. It is expensive to make all portpins of a controller accept a higher voltage than the core. And you don´t want to pay for it. As long as you can run your peripherals at max. 3.6V too, its a good advice to do so. |
|
|
| |
|
|
|
|
|
|
Posted: Oct 25, 2011 - 12:09 PM |
|
Joined: May 27, 2010
Posts: 396
Location: The land of Cyrus the Great
|
|
| I have connected a 5V supply T6963c-based 240x128 GLCD to XMEGA with 1Kohm series resistors on DB0-DB7 and it works fine. |
_________________
Ozhan K.
|
| |
|
|
|
|
|
|
Posted: Oct 25, 2011 - 02:33 PM |
|
Joined: Aug 23, 2008
Posts: 31
|
|
|
electronic.designer wrote:
I have connected a 5V supply T6963c-based 240x128 GLCD to XMEGA with 1Kohm series resistors on DB0-DB7 and it works fine.
We´ll see, how long... If the display is set for outputting data, it will broil the XMEGA some time. |
|
|
| |
|
|
|
|
|
|
Posted: Oct 25, 2011 - 03:06 PM |
|
Joined: Jul 27, 2001
Posts: 7429
Location: St. Leonards-on-Sea (UK)
|
|
| Why? That technique is often used without any problems. |
_________________
Leon Heller
G1HSM
|
| |
|
|
|
|
|
|
Posted: Oct 25, 2011 - 06:52 PM |
|
Joined: Aug 23, 2008
Posts: 31
|
|
| As long as the display is only in write mode (it gets data from the controller) then yes. You have to hardwire the R/W-Pin to the corresponding level so that the display never can transmit data. |
|
|
| |
|
|
|
|
|
|
Posted: Oct 25, 2011 - 09:12 PM |
|
Joined: May 27, 2010
Posts: 396
Location: The land of Cyrus the Great
|
|
| There is no problem for reading from lcd, bacause of series resistors and internal protection diodes. |
_________________
Ozhan K.
|
| |
|
|
|
|
|
|
Posted: Oct 26, 2011 - 09:20 AM |
|
Joined: Jul 12, 2002
Posts: 1
|
|
Hi,
Waiting for having the XMEGA-B1 Xplained schematics on Atmel website, you will find them attached.
Thierry |
|
|
| |
|
|
|
|
|
|
Posted: Oct 26, 2011 - 12:42 PM |
|
Joined: Aug 23, 2008
Posts: 31
|
|
|
electronic.designer wrote:
There is no problem for reading from lcd, bacause of series resistors and internal protection diodes.
1k series resistors are too low in resistance. The "protection" diodes are able to take no more than .5mA of current. Another thing: If all outputs of the display are logic high, the XMEGAs Vcc will be shifted over the body diodes; maybe higher than the specified maximum. This is a not reliable technique. |
|
|
| |
|
|
|
|
|
|
Posted: Oct 26, 2011 - 12:51 PM |
|
Joined: Jul 27, 2001
Posts: 7429
Location: St. Leonards-on-Sea (UK)
|
|
| It depends on the application, it can be reliable. |
_________________
Leon Heller
G1HSM
|
| |
|
|
|
|
|
|
Posted: Oct 26, 2011 - 04:50 PM |
|
Joined: Nov 22, 2002
Posts: 12054
Location: Tangent, OR, USA
|
|
I think that TravelRec's point has to do with the behavior of most regulators.
IF the external device is powered from a separate supply
AND IF the current load on the regulator supplying the controller is small
AND IF you are supplying a lot of current though transient protection diodes
THEN you could supply enough current INTO the regulator output that would require the regulator to sink instead of source. This would cause the regulator output to rise above its designed output voltage level, and possibly exceeding the voltage ratings of the micro.
With some care, this situation can be avoided. But, you have to recognize the possibility in order to know what to do. Its a sneaky one because it operates outside of "normal" circuit behavior (current through normally inactive diodes).
This is basically the same issue that happens when someone tries to shut off the power to the micro while keeping active voltages connected to port pins.
Of course, if the current source is powered from the same regulator as the mcu, then the net regulator load current remains positive and that little corner of the world will remain happy.
Jim |
_________________
Jim Wagner
Oregon Research Electronics, Consulting Div.
Tangent, OR, USA
"The only thing standing between us and victory is defeat" P.G.Wodhouse in Wooster & Jeeves series
|
| |
|
|
|
|
|
|
Posted: Oct 26, 2011 - 06:21 PM |
|
Joined: Aug 23, 2008
Posts: 31
|
|
Jim,
you are absolutely right. I´m a bit careful due to my job to develop things that nearly live forever 
It happens to me in the past that I freezed up AVR port pins because of feeding too much current into the body diodes. This pins were constantly high, constantly low, open or something between that after the treatment, so I will not recommend to conduct the diodes with more than the specified current, which is .5mA for XMegas and 1mA for normal Megas and Tinys. |
|
|
| |
|
|
|
|
|
FAQ
Search
Register
Log in to check your private messages
Log in
Maximize
RSS
