5v on input pins, and not damaging the DUE

Greetings folks, I understand that putting 5v on an input pin of the Arduino DUE will damage the board; I want to avoid that.

I have 'one' input pin that is held HI with a pull-up resistor to a 5v supply; 10k pullup. I want to briefly pull the line low with my DUE; specifically, I want to know if this situation will cause harm to the DUE.

In other words, is the harm a voltage thing, or a current thing? Does the high value pullup resistor protect the DUE by limiting the current, or is it still a problem for the DUE that the voltage on the pin is over 3.3 volts?

Seems like interfacing some projects to work with both the DUE and the Mega is going to be a problem, because the DUE is going to need a special interface because of this voltage consideration.

TIA

marcus

A 19¢ transistor with a 1¢ resistor solves the problem. Are you really trying to save 20¢?

No. Its not the cost of parts, nor even knowing what to do to correct the problem that is bothering me. I want to use the DUE (Arm, speed, memory) in place of the Mega in several projects, but was not expecting to have to modify those projects (only expecting to modify the sketch, probably). So, this voltage thing has thrown a kink in my works, so to speak.

I would still like to know under what conditions the DUE can be damaged, is it current or voltage related, and what the tolerances are for mistakes (how forgiving is the circuit?).

If the DUE gets damaged, is it just that one I|O pin, or does it damage the board more thoroughly?

TIA

PS I don't have any problem throwing a transistor switch in there, to solve the problem. I just have additional questions. Is there a resource I could read?

Well, the SAM datasheet lacks a description of the pin driver so I will have to describe what happens when the voltage on an AVR pin is too high...

There are a pair of internal protection (clamping) diodes connected to each I/O pin. When the voltage is above VCC+0.5 the "upper" diode conducts (clamping the pin voltage near VCC+0.5). When the voltage is below -0.5 the "lower" diode conducts (clamping the pin voltage near -0.5).

The protection diodes can tolerate about 1mA of current. Beyond that they will be damaged / destroyed leaving the rest of the chip vulnerable to damage / destruction.

I would still like to know under what conditions the DUE can be damaged, is it current or voltage related...

Assuming the SAM processor also has protection diodes, over-voltage causes them to conduct. Current flow destroys them.

...what the tolerances are for mistakes (how forgiving is the circuit?).

I can't find anything in the datasheet to indicate. I would take that to mean DO NOT EXCEED VCC+0.3.

If the DUE gets damaged, is it just that one I|O pin, or does it damage the board more thoroughly?

Assuming the SAM processor also has protection diodes, they will be damaged / destroyed first followed by whatever Mother Nature feels is the most valuable part of the chip.

Is there a resource I could read?

The datasheet.

In the situation you described it is both the over voltage and the subisiquent over current that causes the damage.
Who knows what will be damaged when this happens, you can't say it will only be one pin, basically don't do it.

Personally I wouldn't exceed the absolute maximum ratings mentioned in the datasheet.

Even here: http://arduino.cc/en/Main/ArduinoBoardDue

Warning: Unlike other Arduino boards, the Arduino Due board runs at 3.3V. The maximum voltage that the I/O pins can tolerate is 3.3V. Providing higher voltages, like 5V to an I/O pin could damage the board.

I want to know if this situation will cause harm to the DUE.

Consider the question answered.

I want to know if this situation will cause harm to the DUE.

It's not an acronym. The spelling is Due.

If you don't need to use shields, or all the i/o pins,
you could consider a Teensy 3.1 which is 5v tolerant.

Alternatively you could switch to one of the chipkit boards
which uses pic32 processors which does have the shield support and is also 5v tolerant.

I always thought it was pretty silly to create the DUE without 5v tolerant i/o pins,
particularly given the Arduino target audience and especially
since the "Arduino way" as described in Massimo's Arduino book
encourages just plugging things in and trying it vs spending time learning
the technical details.

--- bill

bperrybap:
I always thought it was pretty silly to create the DUE without 5v tolerant i/o pins,
particularly given the Arduino target audience and especially
since the "Arduino way" as described in Massimo's Arduino book
encourages just plugging things in and trying it vs spending time learning
the technical details.

I agree totally. I am building a generic 'calculator shield' which will run over the Mega2560, or the Due. What a complete pain. I ended up placing an interfacing comparator in-line on every I|O pin (1/4 LM339) with pullup resistors. The 5v LCD LMB162ABC runs nicely over the 3.3v Due with the LM339 isolation buffers|shifters. I really want to use the Due mainly for its speed, memory, and DMA. But, again, the board defeats the Arduino purpose of more-or-less plug and play. Of course all of this is really electrical engineering (and software engineering) but with the Due, the electrical engineering aspect is definitely back. The good news is that I did not damage my board... some very thorough testing is revealing all is good! :slight_smile:

Thanks

I haven't seen much use for the Due myself. The Uno, even, running at 16 MHz is much faster than the old IBM PCs (4.77 MHz).

The Mega2560 with 256 kB of program memory and 8 kB of RAM, must surely have more memory than most early calculators. Or indeed, early computers, like the first Apple II with 4 kB of RAM.

The good news is that I did not damage my board... some very thorough testing is revealing all is good!

You can not conclude that your board is not damaged, you may conclude that it is still functioning which is not the same thing. It might fail earlier than it otherwise would have.

Grumpy_Mike:
You can not conclude that your board is not damaged, you may conclude that it is still functioning which is not the same thing. It might fail earlier than it otherwise would have.

True... but I am mostly certain that I did not exceed the 1 mA current on the clamping diodes, and the board has been in 'burn-in' mode now for a couple of days just clicking along within specs and without failure.

I honestly believe the board (Due) is more robust than the forum and the chatter is speculating about. I am not an advocate of exceeding specs by any means, I'm just saying that the board seems to be more forgiving than I was at first led to believe.

Thanks

hi Nick, I took the spelling right off the board itself. The photo is the generic calculator shield I am prototyping for my homies on MoHPC. Have a great day dude!

Cheers,
marcus
:slight_smile:

I'm just saying that the board seems to be more forgiving than I was at first led to believe.

There has been lots of reports of the D/A pin blowing up with too much load.

Grumpy_Mike:
There has been lots of reports of the D/A pin blowing up with too much load.

This is really sad. In my opinion the Due should have overload protection, and should have all I|O pins rated for 5v on-board. I am hoping for another board from Arduino that will take advantage of the Arm processor and yet be 5v compliant.

Your point is well-said; "stay within the specs"

In my opinion the Due should have overload protection, and should have all I|O pins rated for 5v on-board.

The problem here is that the processor itself is lacking this feature. If you try and put buffers on them then you run into problems with defining them to be inputs or outputs. If you make predefined assumptions then you limit the functionality of the board. It also greatly increases the cost of the board.

In short the Due is a board for grown ups, it is not a beginners board, if you can't cope with the 3V3 input / output then you are not ready for the board.
This is exactly the same situation as the Raspberry Pi.

Grumpy_Mike:
There has been lots of reports of the D/A pin blowing up with too much load.

Sorry to respond to you twice, but what do you mean by "load"? According to the specs the 5v pin can handle 800 mA (as well the 3v pin) and the combined I|O line current can fall within 130 mA. Each pin can source (3 mA or 15 mA) and each pin can sink (6 mA or 9 mA) depending on pin.

This is 'different' than voltage 'overage'. Supposedly none of the pins can stand any more than 3.3 volts without damage. The spec sheet is not clear how or why the damage occurs; but that's not the main point--- the Due should NOT be this fussy. The direction of this open hardware project has clearly been opposed to that from the beginning. Just saying.

Cheers,
marcus
:slight_smile:

Mark_H_Harris:
hi Nick, I took the spelling right off the board itself. ... Have a great day dude!

Thanks, dude!

However I think the all caps there is a style thing. Look at their page:

Sure the photo shows DUE on the board, but all through the page it is "Due".

Look at this picture of the Uno:

It is a Uno, not a UNO, and the company name is Arduino not ARDUINO despite the photograph.

None of those are acronyms and in English they are not properly capitalized.

Nor is Italy spelt capitalized ITALY in ordinary text.

I could go on, but will resist.

he Due should NOT be this fussy.

I think you are misunderstanding the entire Arduino project.
This is not some sort of bullet proof training bike with stabilizers, if you think it should be then that is not how electronics works.