AVRtiny input port sinking current

Go To Last Post
39 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi

I have an AVRtiny84A in a really space sensitive application where it needs to digitally detect a signal that is either 0V (to detect as 0) or anything between 10V and 20V (to detect as 1). The generic way would be to feed the signal through a resistor to a port pin and add a 4V7 zener (for 5V power) to GND. Since I do not have the space for the Zener (the PCB is 10.1mm x 23.2mm x 1.8mm right now and it cannot be much larger), I would like to do without the zener. This should be possible as I believe pretty much all modern controllers have a zener on port pins anyway, but it requires a careful choice of resistor value. The $100000 question is: "How much current can it sink?" It seems to me Atmel (Microchip) does not list the information in the datasheet?

So if it could sink 1mA, for instance, I'd choose half for safety, and use a (20V-5V)/0.5mA=30Ohms. But Atmel seems to only specify maximum overvoltage, which really isn't enough to settle this question. An input port is really high Z, so if for instance there are no zeners, then even 1MegOhm would drive the input voltage beyond anything the chip can tolerate. For signal integrity I would rather not use such high resistor values unless they are necessary.

Anybody in the know?

This topic has a solution.

Cheers  Peter

Last Edited: Sat. Aug 18, 2018 - 01:26 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

AVR pins are symmetric, they can generally sink or source the same amount. It's usually about 20mA but will be listed in the datasheet.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

OP is not talking about the drive capability - he's asking about how much the protection diodes can stand when an over-voltage is applied to a pin ...

 

Some manufacturers do specify this as "injected current" or some similar term - but the Microchip datasheet does not.

 

But this is basically an abuse of the part - the datasheet says, "don't do that!"

 

EDIT

 

typo - "stand"

Top Tips:

  1. How to properly post source code - see: https://www.avrfreaks.net/comment... - also how to properly include images/pictures
  2. "Garbage" characters on a serial terminal are (almost?) invariably due to wrong baud rate - see: https://learn.sparkfun.com/tutorials/serial-communication
  3. Wrong baud rate is usually due to not running at the speed you thought; check by blinking a LED to see if you get the speed you expected
  4. Difference between a crystal, and a crystal oscillatorhttps://www.avrfreaks.net/comment...
  5. When your question is resolved, mark the solution: https://www.avrfreaks.net/comment...
  6. Beginner's "Getting Started" tips: https://www.avrfreaks.net/comment...
Last Edited: Fri. Aug 17, 2018 - 08:12 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

perot wrote:
I believe pretty much all modern controllers have a zener on port pins

Not a Zener; just a diode - see the diagram in the datasheet:

http://ww1.microchip.com/downloads/en/DeviceDoc/doc8183.pdf

 

Top Tips:

  1. How to properly post source code - see: https://www.avrfreaks.net/comment... - also how to properly include images/pictures
  2. "Garbage" characters on a serial terminal are (almost?) invariably due to wrong baud rate - see: https://learn.sparkfun.com/tutorials/serial-communication
  3. Wrong baud rate is usually due to not running at the speed you thought; check by blinking a LED to see if you get the speed you expected
  4. Difference between a crystal, and a crystal oscillatorhttps://www.avrfreaks.net/comment...
  5. When your question is resolved, mark the solution: https://www.avrfreaks.net/comment...
  6. Beginner's "Getting Started" tips: https://www.avrfreaks.net/comment...
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

awneil wrote:

perot wrote:
I believe pretty much all modern controllers have a zener on port pins

Not a Zener; just a diode - see the diagram in the datasheet:

 

My apology, I used incorrect wording. The issue at hand though stays the same.

Cheers  Peter

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

1mA has been mentioned in earlier datasheets for other  models, but I don't know definitively for the tiny84A. I cannot imagine that it would be higher.

 

Ross McKenzie ValuSoft Melbourne Australia

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

For a definitive answer, you'd have to ask Microchip themselves.

 

But the fact that it's not documented in the datasheet may well mean that it's just something that they won't specify.

 

Top Tips:

  1. How to properly post source code - see: https://www.avrfreaks.net/comment... - also how to properly include images/pictures
  2. "Garbage" characters on a serial terminal are (almost?) invariably due to wrong baud rate - see: https://learn.sparkfun.com/tutorials/serial-communication
  3. Wrong baud rate is usually due to not running at the speed you thought; check by blinking a LED to see if you get the speed you expected
  4. Difference between a crystal, and a crystal oscillatorhttps://www.avrfreaks.net/comment...
  5. When your question is resolved, mark the solution: https://www.avrfreaks.net/comment...
  6. Beginner's "Getting Started" tips: https://www.avrfreaks.net/comment...
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

perot wrote:
The issue at hand though stays the same.
perot wrote:
anything between 10V and 20V (to detect as 1)

And it >>is<< in the datasheet -- ever hear of Absolute Maximum Ratings?

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

Last Edited: Fri. Aug 17, 2018 - 03:34 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I seem to recall an app note with a zero crossing detector using a similar arrangement, IIRC a 1 Meg resistor was used, though an external diode (zener) was recommended....

A site search for ZCD should find it.

 

Jim

 

 

(Possum Lodge oath) Quando omni flunkus, moritati.

"I thought growing old would take longer"

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

ki0bk wrote:
A site search for ZCD should find it.

Or just Product Page --> Documents --> http://ww1.microchip.com/downloads/en/AppNotes/Atmel-2508-Zero-Cross-Detector_ApplicationNote_AVR182.pdf

 

EDIT

 

perot wrote:
AVRtiny84A
 

Using the correct part number helps: ATTiny84A

Top Tips:

  1. How to properly post source code - see: https://www.avrfreaks.net/comment... - also how to properly include images/pictures
  2. "Garbage" characters on a serial terminal are (almost?) invariably due to wrong baud rate - see: https://learn.sparkfun.com/tutorials/serial-communication
  3. Wrong baud rate is usually due to not running at the speed you thought; check by blinking a LED to see if you get the speed you expected
  4. Difference between a crystal, and a crystal oscillatorhttps://www.avrfreaks.net/comment...
  5. When your question is resolved, mark the solution: https://www.avrfreaks.net/comment...
  6. Beginner's "Getting Started" tips: https://www.avrfreaks.net/comment...
Last Edited: Fri. Aug 17, 2018 - 01:15 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

theusch wrote:

perot wrote:
The issue at hand though stays the same.
perot wrote:
anything between 10V and 20V (to detect as 1)

And it >>is<< in the datasheet -- ever hear of Absolute Maximum Ratings?

 

<datasheet quote in huge font deleted>

 

 

I do not know why you shout at me.

 

If you read my original posting carefully, you'll discover that I wrote

 

Quote:
But Atmel seems to only specify maximum overvoltage, which really ...

 

so I'm very well aware of maximum ratings. And that is exactly what I'm asking: How much current flowing through the protection diode raises it's voltage drop beyond the 0.5V specified? (I assume I do not need to teach you about how the voltage drop of a diode depends on the current flowing through it, do I?)

Cheers  Peter

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

awneil wrote:

ki0bk wrote:
A site search for ZCD should find it.

Or just Product Page --> Documents --> http://ww1.microchip.com/downloads/en/AppNotes/Atmel-2508-Zero-Cross-Detector_ApplicationNote_AVR182.pdf

Thank you ki0bk and awneil, I'll take a look at this.

 

Quote:

perot wrote:
AVRtiny84A
 

Using the correct part number helps: ATTiny84A

Sorry, an old unfortunate habit of mine. I guess it is because to me ATTiny looks like ATT and iny concatenated, which does not appear to make sense. So somehow I figured ATtiny looks more right to me. Maybe I should make an effort to get rid of that habit.

 

Edit: Typo

Cheers  Peter

Last Edited: Fri. Aug 17, 2018 - 02:48 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

perot wrote:
I do not know why you shout at me.

There is no shouting.  Simply put, when you paste an image into a post, this forum software expands it to fit the width of >>your<< window.  There is NO font involved; it is a "snapshot" from Firefox.

 

If you make your viewing window narrower, then the "font" is smaller.

 

Is all the above my fault?

 

perot wrote:
(I assume I do not need to teach you about how the voltage drop of a diode depends on the current flowing through it, do I?)

Sure, why don't you make this a teaching opportunity.  But don't shout.

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

This reply has been marked as the solution. 
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

perot wrote:

awneil wrote:

ki0bk wrote:
A site search for ZCD should find it.

Or just Product Page --> Documents --> http://ww1.microchip.com/downloads/en/AppNotes/Atmel-2508-Zero-Cross-Detector_ApplicationNote_AVR182.pdf

Thank you ki0bk and awneil, I'll take a look at this.

 

Beyond a datasheet specification, I guess an application note is the next best information I can get. In this sense, this note solves my problem.

 

The note uses two resistors of 1MOhm to isolate the 230V mains from the pin. That means the diodes accept 160µA of current. In my application, this translates to using a (20V-5V)/0.16mA=100kOhm.

 

Thanks again for the help.

Cheers  Peter

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

theusch wrote:
when you paste an image into a post, this forum software expands it to fit the width of >>your<< window.  

But >>you<< can double-click the image, and set its size; eg, taking a copy of your image:

 

Top Tips:

  1. How to properly post source code - see: https://www.avrfreaks.net/comment... - also how to properly include images/pictures
  2. "Garbage" characters on a serial terminal are (almost?) invariably due to wrong baud rate - see: https://learn.sparkfun.com/tutorials/serial-communication
  3. Wrong baud rate is usually due to not running at the speed you thought; check by blinking a LED to see if you get the speed you expected
  4. Difference between a crystal, and a crystal oscillatorhttps://www.avrfreaks.net/comment...
  5. When your question is resolved, mark the solution: https://www.avrfreaks.net/comment...
  6. Beginner's "Getting Started" tips: https://www.avrfreaks.net/comment...
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

perot wrote:
Beyond a datasheet specification, I guess an application note is the next best information I can get

Yeah - not great having Atmel Microchip themselves publishing documents which rely upon unspecified features!

 

You could still raise a support case requesting clarification ...

 

In this sense, this note solves my problem.

 

Might as well mark the solution, then?

(see Tip #5)

 

Top Tips:

  1. How to properly post source code - see: https://www.avrfreaks.net/comment... - also how to properly include images/pictures
  2. "Garbage" characters on a serial terminal are (almost?) invariably due to wrong baud rate - see: https://learn.sparkfun.com/tutorials/serial-communication
  3. Wrong baud rate is usually due to not running at the speed you thought; check by blinking a LED to see if you get the speed you expected
  4. Difference between a crystal, and a crystal oscillatorhttps://www.avrfreaks.net/comment...
  5. When your question is resolved, mark the solution: https://www.avrfreaks.net/comment...
  6. Beginner's "Getting Started" tips: https://www.avrfreaks.net/comment...
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

awneil wrote:
But >>you<< can double-click the image, and set its size; eg, taking a copy of your image:

See -- a teaching moment.

 

I asked for another about the diode drop, but apparently is not forthcoming.

 

Off to edit my post above...

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I would update the Wiki page to note this - but that's been broken for years.

 

angry

 

(it's now in the link for Tip #1)

 

Also note that editing a post will lose all your image sizes.

That's also been reported for years.

 

angry

 

 

Top Tips:

  1. How to properly post source code - see: https://www.avrfreaks.net/comment... - also how to properly include images/pictures
  2. "Garbage" characters on a serial terminal are (almost?) invariably due to wrong baud rate - see: https://learn.sparkfun.com/tutorials/serial-communication
  3. Wrong baud rate is usually due to not running at the speed you thought; check by blinking a LED to see if you get the speed you expected
  4. Difference between a crystal, and a crystal oscillatorhttps://www.avrfreaks.net/comment...
  5. When your question is resolved, mark the solution: https://www.avrfreaks.net/comment...
  6. Beginner's "Getting Started" tips: https://www.avrfreaks.net/comment...
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

 (20V-5V)/0.5mA=30Ohms

huh?   50000 ...I weep for the future 

https://www.youtube.com/watch?v=SMGz3sXRhvE

 

The diode is usually a parasitic diode due to the making of the chip. 

 

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

avrcandies wrote:

 (20V-5V)/0.5mA=30Ohms

huh?   50000 ...I weep for the future 

 

Ahem.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 1

Guess I need a new calculator...

 

I had written a lengthy reply, and then the neighborhood lost power this morning.

 

If you can fit one resistor, can you not use a smaller (physical) resistor and mount two of them for a proper voltage divider.

If you are making only a few boards then one can spend a lot of extra time and effort to mount super small parts, under a microscope.

 

You could, in fact, mount a Zener physically on top of the lower resistor, (put them in parallel), to further clamp the divider.

 

JC

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

DocJC wrote:

If you can fit one resistor, can you not use a smaller (physical) resistor and mount two of them for a proper voltage divider.

If you are making only a few boards then one can spend a lot of extra time and effort to mount super small parts, under a microscope.

 

You could, in fact, mount a Zener physically on top of the lower resistor, (put them in parallel), to further clamp the divider.

Another option is a BJT with built-in base resistor, and use the internal pull-up as the collector resistor.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

DocJC wrote:
If you can fit one resistor, can you not use a smaller (physical) resistor and mount two of them for a proper voltage divider.

Now, I may need another lesson -- in OP's quest for miniature, are high-voltage-rated resistors as small as low-voltage-rated?

 

That said, OP's stated signal range isn't mains but a more modest level.

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Atmel has "said"  1ma, though take this with a grain of salty whole wheat snacks (and it may not be the same for all chips & fab processes)...but 1ma seems reasonable. In fact, it could easily be 10x that.  Many logic gate chips I used  years ago actually spec'd the internal diodes at 20ma for clamp limits.

https://www.avrfreaks.net/forum/clamping-diodes-0  (mentions Atmel "spec" of 1ma)

 

Be sure that your power supply will resist the intrusion & prevent the Vcc from simply trying to rise too high!  In a big board, all the other stuff (such as pull downs, power leds) can load things down to prevent this...but if your board has few parts to do so, could be an issue....most regulators won't do anything to prevent the voltage from rising.  Then BANG your chip is destined for a bowl of solder soup.

 

Note high ohms will knock down your max bit rate (RC filter effect).  All of the tiny84 I/O have Schmitt trigger inputs, a benefit for slowed up/down inputs.

 

Do you have parts on both (dual) sides of the PCB??...that's 2x the room (nicely ignoring the need for traces).   You could perhaps put a series resistor off board, in the cabling (not really a great idea) or at the source (not much better).

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

Last Edited: Sat. Aug 18, 2018 - 02:21 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

perot wrote:
and use a (20V-5V)/0.5mA=30Ohms

You meant 30 kOhms.

 

I have been using 12V / 39 kOhm on pin RX (M88) for a long time with no harm.

Only the maximal usable frequency is lower due to combination of the resistor and input pin capacity.

Last Edited: Sat. Aug 18, 2018 - 06:16 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Visovian wrote:
with no harm.

How do you know there is no "harm" ?

 

I'd guess this is the kind of "over stress" which doesn't instantly kill, but can gradually degrade the part ...

Top Tips:

  1. How to properly post source code - see: https://www.avrfreaks.net/comment... - also how to properly include images/pictures
  2. "Garbage" characters on a serial terminal are (almost?) invariably due to wrong baud rate - see: https://learn.sparkfun.com/tutorials/serial-communication
  3. Wrong baud rate is usually due to not running at the speed you thought; check by blinking a LED to see if you get the speed you expected
  4. Difference between a crystal, and a crystal oscillatorhttps://www.avrfreaks.net/comment...
  5. When your question is resolved, mark the solution: https://www.avrfreaks.net/comment...
  6. Beginner's "Getting Started" tips: https://www.avrfreaks.net/comment...
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

avrcandies wrote:
may not be the same for all chips & fab processes

I think that would be the main concern.

 

Possibly, the reason they don't specify it is precisely because it is dependent on fabs and/or processes ...

 

avrcandies wrote:
Atmel has "said"  1ma

When, exactly, did they "say" it? And which part(s) were they talking about at that time?
What were the processes & geometries at the time? 

Does it still hold for current  processes & geometries ... ?

 

avrcandies wrote:
Many logic gate chips I used  years ago actually spec'd the internal diodes at 20ma for clamp limits.

but geometries back then were very different ...

 

EDIT

 

The original publication date is not stated in the revision history, but the first update was 2004. 

There is no mention of any update to cover new parts or processes.

http://ww1.microchip.com/downloa...

 

Figure 1 shows an ATmega163 

The ATmega163 no longer seems to appear on the Microchip website, but their search finds the Rev-E datasheet, dated 2003 - and even then it was NRND ...

It does not specify protection diode current.

http://ww1.microchip.com/downloads/en/DeviceDoc/doc1142.pdf

Top Tips:

  1. How to properly post source code - see: https://www.avrfreaks.net/comment... - also how to properly include images/pictures
  2. "Garbage" characters on a serial terminal are (almost?) invariably due to wrong baud rate - see: https://learn.sparkfun.com/tutorials/serial-communication
  3. Wrong baud rate is usually due to not running at the speed you thought; check by blinking a LED to see if you get the speed you expected
  4. Difference between a crystal, and a crystal oscillatorhttps://www.avrfreaks.net/comment...
  5. When your question is resolved, mark the solution: https://www.avrfreaks.net/comment...
  6. Beginner's "Getting Started" tips: https://www.avrfreaks.net/comment...
Last Edited: Sat. Aug 18, 2018 - 09:01 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

awneil wrote:
When, exactly, did they "say" it?

LOL -- now you are going to strain my memory and/or search skills.  IIRC there are references/quotes buried in the annals of past discussions.

awneil wrote:
And which part(s) were they talking about at that time? What were the processes & geometries at the time? Does it still hold for current processes & geometries ... ?

...which will bring us to these questions.  Surely the new overlords will be straining to help us clear this up.

 

 

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

awneil wrote:
When, exactly, did they "say" it?

The zero-crossing app note.  (Didn't we just examine that above?)

https://www.avrfreaks.net/forum/...

...where the OP there raised many of the same follow-on questions a year or so ago.

 

Many similar discussions:

https://www.avrfreaks.net/forum/...

https://www.avrfreaks.net/forum/...

https://www.avrfreaks.net/forum/...

...

 

Now I have to dig through them to see what "Atmel said"...

 

[edit]  lol -- indeed, much seems to be tribal knowledge/urban legend/zero-crossing app note.

 

-- one of the links asked about Xmega, and apparently there is no mention in the datasheets of protection diodes?

-- https://www.avrfreaks.net/commen...

has a datasheet quote about "injection current".  Is that the same thing?  [and that is a more recent/current part]

-- same here https://www.avrfreaks.net/commen... "Injection current" given in all Automotive datasheets?  Do Automotive parts use the same process? devil

-- in 2003 apparently a 'Freak got an Atmel response with the 1mA figure

https://www.avrfreaks.net/commen... [what are the wagering odds that the figure came from the app note?]

-- and here https://www.avrfreaks.net/commen...

-- hmmm, a Trondheim address -- does that imply "Atmel"? https://www.avrfreaks.net/commen...

eiriksle wrote:
Just FYI, The protection diodes on an AVR can tolerate up to 1mA and even more (10mA) over short periodes of time (in the us range)... Sletten

-- and here https://www.avrfreaks.net/commen...

 

 

 

 

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

Last Edited: Sat. Aug 18, 2018 - 01:10 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

did you see my edit to #27?

 

(not sure if the times shown by the forum are the viewer's local time, or the poster's local time)

Top Tips:

  1. How to properly post source code - see: https://www.avrfreaks.net/comment... - also how to properly include images/pictures
  2. "Garbage" characters on a serial terminal are (almost?) invariably due to wrong baud rate - see: https://learn.sparkfun.com/tutorials/serial-communication
  3. Wrong baud rate is usually due to not running at the speed you thought; check by blinking a LED to see if you get the speed you expected
  4. Difference between a crystal, and a crystal oscillatorhttps://www.avrfreaks.net/comment...
  5. When your question is resolved, mark the solution: https://www.avrfreaks.net/comment...
  6. Beginner's "Getting Started" tips: https://www.avrfreaks.net/comment...
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I'm sort of lost. I cannot see who wrote each individual message, the whole left column with the author's name and avatar is not displayed. Is that because the stupid forum software "thinks" it's got to save screen space on my 2" Android display, which in reality is a 12.2" tablet with 2560x1600, screen space galore ... :-( I cannot find any means to explicitly select a desktop like view.

Whatever.

I marked the solution now.

Guys suggested to use smaller resistors and two of them for a divider. However, the "on" voltage range is to large. I could probably find a working divider ratio, but whichever it is, it is either very close to not reaching the switching threshold at the range's lower end, or very close to exceeding the specs if the input voltage is at the upper end. Far to close for any reasonable engineering standard.

It was also suggested to stack components, or put some on the second side. I often populate the second side too, but here it is impossible. I gave you the space I have: 1.8mm in height, and that includes the PCB! The thinnest affordable PCBs are 0.6mm, that leaves maximum 1.2mm for components and a thin solder layer. I'm happy the Tiny84A is available in QFN20 which is only 0.8mm high, the other semiconductors also are 1mm height maximum. MLCCs were a bit of a challenge, resistors no problem. Typical zener packages are higher than 1.5mm and so are BJTs and other discrete components.

Cheers  Peter

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

It seems that the consensus is that relying on the internal diodes MIGHT be reasonable, but there are no guarantees.  I suppose it depends on how much risk you can take of parts beginning to fail in the field.  If this is a small project (in terms of numbers and consequences-of-failure) that's one thing.  If a large project then let the higher-ups make the final decision.  I know I would do everything in my power to add external signal conditioning (I'm also the kind of person who never overclocks anything laugh ), but if, as you say, you really and truly don't have the room, well then good luck.

Last Edited: Sat. Aug 18, 2018 - 03:02 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

oops, accidentally posted this in the old quoting thread ...but this is what Atmel "said"

Not much to go by & may not apply rto your chip!

 

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

awneil wrote:
How do you know there is no "harm" ?

 

Because the uart still works after years of using this.

 

Atmel Application Note "AVR182: (2016)

 

To protect the device from voltages above VCC and below GND, the AVR has internal clamping diodes on the I/O pins (see Figure -1).

The diodes are connected from the pins to VCC and GND and keep all input signals within the AVR’s operating voltage (see the figure below).

Any voltage higher than VCC + 0.5V will be forced down to VCC + 0.5V (0.5V is the voltage drop over the diode) and any voltage below GND - 0.5V will be forced up to GND - 0.5V.

By adding a large resistor in series, these diodes can be used to convert a high voltage sinus signal down to a low voltage square wave signal, with amplitude within the AVR’s operating voltage ±0.5V.

The diodes will thus clamp the high voltage signal down to the AVR’s operating voltage. 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

awneil wrote:
did you see my edit to #27?

Well, yeah, I have now.  But can you answer about whether "injection current" is pertinent?  And if it is, then we find it in later datasheets from current models.  [the secondary is whether Automotive is the same process used in the models we know and love; I'd guess that it is the same fab/process with different testing]

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

avrcandies wrote:
... (and it may not be the same for all chips & fab processes)
... and designs of the I/O pad (circuit and layout as in sizing of semiconductors)

Some I/O have SCR protection where a brief over-voltage can occur due to the time duration for the SCR to switch.

SIGNAL CHAIN BASICS #66: How to interface a 5V transceiver to a 3V controller

Planet Analog, 6/25/2012

https://www.planetanalog.com/document.asp?doc_id=528335

SCR for protection is very low leakage; most XMEGA AVR are low leakage except for XMEGA E.

There are a few posts here about the Littelfuse SP72X series (SCR rail clamp) but these won't work for perot because SP72X are in SOT-23 or SOIC.

IIRC, an SP72X app note has a small-value resistor between the SP72X and the input to be protected (approx 10 ohms for current division)

General ESD Protection Devices - Littelfuse

(page 2 of 2, near page bottom for 'SCR & Diode Rail Clamp Array')

https://www.littelfuse.com/products/tvs-diode-arrays/general-purpose-esd-protection.aspx?curPage=2

 

perot might evaluate an R-TVS-R in 01005 for all parts.

A highlight at Digi-Key :

SP1053/SP1054 Series - Littelfuse | DigiKey

ESD Protection for I/O and Power Ports - SP1053/SP1054 Series

https://www.digikey.com/en/product-highlight/l/littelfuse/esd-protection-for-io-and-power-ports-sp1053-sp1054-series

 

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 1

To protect the device from voltages above VCC and below GND, the AVR has internal clamping diodes on the I/O pins

I wonder if they Atmel purposely added extra diodes...often parasitic diodes are there regardless ...maybe they just relabeled them as a "feature", making it seem like they care about more about protection.

If they went to this extra trouble, seems like they'd give details.

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

awneil wrote:

did you see my edit to #27?

theusch wrote:
Well, yeah, I have now.

I wasn't sure if my editing had just overlapped with your posting

 

  But can you answer about whether "injection current" is pertinent?

I think "injection current" is the term used to describe the current under discussion here.

 

then we find it in later datasheets from current models. 

I thought I had checked the DS for the OP's model - and didn't see it.

 

[the secondary is whether Automotive is the same process used in the models we know and love; I'd guess that it is the same fab/process with different testing]

Yes - that would be my guess.

Top Tips:

  1. How to properly post source code - see: https://www.avrfreaks.net/comment... - also how to properly include images/pictures
  2. "Garbage" characters on a serial terminal are (almost?) invariably due to wrong baud rate - see: https://learn.sparkfun.com/tutorials/serial-communication
  3. Wrong baud rate is usually due to not running at the speed you thought; check by blinking a LED to see if you get the speed you expected
  4. Difference between a crystal, and a crystal oscillatorhttps://www.avrfreaks.net/comment...
  5. When your question is resolved, mark the solution: https://www.avrfreaks.net/comment...
  6. Beginner's "Getting Started" tips: https://www.avrfreaks.net/comment...
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

awneil wrote:
Yes - that would be my guess.

Follow the links where I mentioned the injection current, see the quotes and links where they came from, and note the models involved.  I'd speculate that the same phrasing would be found for all applicable models and generations.  [does it matter to go any further?]

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.