It would be nice if there were a sheet with two columns- the first column shows "a thing that might fry an arduino", and the second would show "in what way the arduino would likely be fried".
For example: (THIS INFORMATION IS NOT CORRECT, JUST SHOWING THE FORMAT)
Try to drive > 2amps through the arduino 5v pin : ATMega chip fried : Replace ATMega Chip
Apply >5v into the 5v pin : ftdi chip fried : Replace ftdi chip; ATMega salvagable
etc...
Not only would this be a good reference for troubleshooting/repair, but it would be an excellent resource for beginners to see a consolidated list of things that you should take care to not-do.
(I know this is an open source community, where the answer to this inquiry might be "why don't you make one yourself and share it!"- but I know very little about electronics, don't know how to verify outcomes, and don't have the money to experiment)
Would be nice. Bear on mind that the Arduino is very simple prototyping board. There is no SW except the bootloader, very small program which is able only to upload programs via USB.
Such features would require more advanced HW and SW.
Oh I'm not expecting any onboard diagnostics- I'm talking a literal excel spreadsheet (maybe saved to .pdf) that can be passed around on the internet as a reference.
It would communicate "if you think you accidentally did X, the outcome is likely Y". I would expect the correlation of Xs to Ys to not be too noisy? Someone who really knows the schematics of the board (+ has experience w/ electronics) I'd expect would have a somewhat intuitive ability to draft that up, even purely theoretically.
I'd be willing to chip a couple bucks in for some rigorous testing too if that's what's needed (apply too great a voltage to an input pin on 3 arduinos, then go through the components 1 by 1 and figure out which still work, write it down, then move on to the next 3 arduinos where we instead test too great a current, etc...). If someone is willing/able to put together a gofundme on the subject, I'm sure it would be met well!
(Then again, maybe I'm way out of my league in understanding here, and the reality of the situation is that regardless of what goes wrong, it could have one of many different results and its impossible to predict which...)
Here's a symptom : here's a potential problem : here's a solution
for example:
My upload isn't working : you might have the wrong serial port selected : select the correct serial port in the Tools tab
What I'm looking for instead is:
here's something that could have fried your arduino : here's what components that would likely be fried : here's how to fix it
for example:
I drove 3amps through the USB port : the usb->serial converter is fried : salvage the ATMega chip and use a new board
(^ like I said previously, that isn't actually a coherent diagnostic/solution; my problem is that I don't know what the correct event/diagnostic/solution pair is, which is why I want this sheet!)
The current troubleshooting page (unless I'm missing something) only offers solutions assuming the problem is misconfiguration of an ultimately functional board. My problem is I blew my board, and want to know my situation/options. The troubleshooting page essentially says "do none of these troubleshooting tips work? then all bets are off- start with a new board". <- insufficient!
Edit:
One additional benefit to this sheet would be to rule out that you've blown the board- that is, you could say "hmm, I might have accidentally shot 12v through a data pin, but am unsure if it ultimately fried my board, and now it's also not working! apparently this sheet says that if that were what happened, it would result in the ATMega chip frying. But I've replaced the ATMega chip, and the board is still failing to work! Maybe I didn't fry the chip, and I should look more closely into configuration issues...". etc...
Your Arduino is fried or otherwise damaged ==> Replace it.
Hardware failures are not so predictable. Electrically abusing a chip will NOT cause a predictable failure that will tell you what caused it. It is up to you to ensure you do not abuse it in the first place. Once you do, ALL bets are off.
Not valid for e.g. UNO with damaged chip. It is very easy to repair.
Not true for SMD versions unless the user has re-work facilities.
Even less true if the user is not able to chase down other possible damaged components that may or may not be fried or partially damaged.
In a lot of cases the user appears to be NEW to the hobby.
In as many of those cases they may not even own a DMM or soldering iron yet.
So really only true in a few sporadic cases.
Whilst a search decision tree troubleshooter might be useful in some cases, so many times you will not be able to move down the tree because you may want the user to perform a function/test that they do not yet understand or they think the problem is somethings else.
A simple example we all come across is "did you read the posting guidelines"...We know it was probably NO but thats not what they want to do. They want help regardless of NO.
And in our good natured way we aid and abet that break in the decision tree.
RayLivingston:
Hardware failures are not so predictable.
Gah I was afraid of that. Alright- that's fair (unfortunate, but understandable).
A couple intermediate resources that would be nice:
A consolidated list of common accidents that fry a board
Some best-practices to prevent those common accidents (am I supposed to just litter my breadboard w/ diodes+fuses?)
A section in the troubleshooting guide that outlines symptoms/behavior of a fried board
A resource outlining ways to diagnose what parts of a board are fried (even if imperfect)
A resource outlining ways to salvage partially fried board (even if imperfectly reliable)
example: ATMega fried on an Uno? You can buy replacements (or salvage from other board) + easily swap!
example: USB->Serial connection fried? If that's all that's wrong, you can use another arduino to upload to it (for use in production w/o need for serial connections) (<- I don't actually know if this is true/how it could be done?)
I know "buy a replacement" is the fool-proof solution, but I live in an area where there is nowhere available to run out and pick up a new arduino. So if something goes wrong, I'm stuck twiddling my thumbs until amazon arrives. (Not to mention- I'm low on cash...)
A consolidated list of common accidents that fry a board
The forum is littered with those. Would not take you long to come with abut 20 common reasons.
Some best-practices to prevent those common accidents (am I supposed to just litter my breadboard w/ diodes+fuses?)
Common sense would seem to dictate the items from the first list.
A section in the troubleshooting guide that outlines symptoms/behavior of a fried board
Again can be garnered from list one.
A resource outlining ways to diagnose what parts of a board are fried (even if imperfect)
Presumes the user has electrical / electronic knowledge and equipment which is often not the case or that knowledge is limited.
A resource outlining ways to salvage partially fried board (even if imperfectly reliable)
Again would only really cover those boards that have socketed items or easily workable components which now only a small fraction of the boards available.
example: USB->Serial connection fried? If that's all that's wrong, you can use another arduino to upload to it (for use in production w/o need for serial connections) (<- I don't actually know if this is true/how it could be done?)
Yes quite often it can be done either with another Arduino or a dedicated ISCP device (available cheaply)
That is often the last thing before a board hits the garbage can and only then if it is suspected that the USB side only is fried.
I applaud your effort and would love to see your finished tree.
However you should be prepared to take some serious critique from some members more knowledgeable than lowly me.
ballscrewbob:
Would not take you long to come with abut 20 common reasons.
...
you should be prepared to take some serious critique from some members more knowledgeable than lowly me.
Yep- which is why I'mreally not qualified to do the consolidation.
ballscrewbob:
I applaud your effort and would love to see your finished tree.
lol I sense a hint of annoyance that I'm asking for work to be done on my behalf. (If I'm misreading that- sorry! but-) honestly, that's fair!
I'm owed nothing by this community- not even all of your already-helpful responses (an earnest thanks for the info!). At the same time, I know the structure of these technical hobby forums attracts newbies barging in, demanding help, and then leaving (I promise I'm self-aware). Resources like the "troubleshooting" page probably cut that down by 90%. If the forum is already littered w/ repeat questions- here's an opportunity to cut it down by another 90%!
(If nobody wants to step up to do it, then nobody wants to step up to do it- nothing I can get mad about there. I'm just offering a data point that hey- in case anyone is wondering, I as a newbie would love to see the resource! Heck, if someone sets up a gofundme to produce a free publicly available resource (maybe just an addition to the troubleshooting page!) I'd chip in $10.)
Regarding these critiques:
ballscrewbob:
...presumes the user has electrical / electronic knowledge and equipment...
...would only really cover those boards that have socketed items...
If the answer to a troubleshooting question is "no, you can't fix it", that certainly does nothing to invalidate the question. Having a consolidated resource that answers it w/ "no" is the difference between a wasted additional forum thread and someone who understands their only available route forward (buy a new board).
ballscrewbob:
Yes quite often it can be done either with another Arduino or a dedicated ISCP device (available cheaply)
More than 95% of the people here are simply giving thier time and expertise at zero price.
Some have written fantastic libraries etc. We are all just giving back one way or another.
In all honesty I did think you were offering to build such a tree.
Slightly miffed thats not the case :-[
At the risk of giving a flippant answer, they are pretty hardy in my experience and I've never damaged one through for example reversing TX and RX or MISO and MOSI or SDA and SCL which are common errors.
"Frying" them pretty much is down to stuffing PSU into the regulated pins, or stuffing 5v into something expecting 3.3v. You can also kill a pin by trying to get it to sink too much current, though even a small resistor prevents that and I've not personally blown one up using say a LED without a resistor (not that I've done that since my early days).
They will even survive momentary shorts to earth though it isn't to be advised (most common reason for this in my experience is when prototyping without using a proto board and having lots of peripheral devices hanging off and then inadvertently touching each other)
The most common failure issue I have is the micro USB connectors on cheap boards physically breaking off when plugging in and unplugging the USB connector, and then trying to glue it back on whilst keeping sufficient electrical connection.
