When I move my hand in the vicinity of an arduino that isn't running a program the pin 13 LED comes on. This is only by waving my hand over it. Has anyone else experienced this? Can anyone explain? It's quite disconcerting. :-/
If the Arduino is powered up, it IS running a program.
Even an out of the box one. How so? :-?
Have you ever experienced what I'm referring to?
A new Arduino typically is delivered with the Blink sketch uploaded. It's uploaded to verify that the Arduino is indeed working properly before it is shipped.
Which Arduino do you have, and where did you get it?
I have a Duemilanove 328 from oomlout. The blink program definitely isn't running at the moment. Yet when I put my hand over the board the LED lights up. I think I need to visit the doctor. ;D
The blink program definitely isn't running at the moment.
Then some program is running. To stop a program hold down the reset button and keep it down. I bet it doesn't happen then.
This action is perfectly normal if you are using the state of an unconnected pin to drive the LED. It's called pick up and beginners are constantly amazed by it.
Im STILL amazed at it 
Haha, nice one geekaway. I tested it with other people on different days and the "pick up" from me was far more effective than anyone else. I think i need grounding.
Ok, then you are grounded for a week, young man! ;D
Just don't take away my arduino!
Whoever it is who turns certain street lights off when I walk by has obviously got bored and moved in to your arduino.
So you know all those "extreme" precautions you're supposed to pay attention to when working with electronics (non-synthetic clothing, wrist straps, anti-static mats, etc)?
This phenomenon helps demonstrate WHY...
So you know all those "extreme" precautions you're supposed to pay attention to when working with electronics (non-synthetic clothing, wrist straps, anti-static mats, etc)?
I have the clothing and wrist straps, but I don't have a mat yet; probably should find one...
Yesterday I was researching MOSFET-based h-bridge, and one site mentioned something about how MOSFETs are very prone to spontaneous self-destruction due to something or another (gate voltage?), unless you install certain resistors first before installing the MOSFET.
What's the "truth" on this?
I am not arguing against it, but it would be useful to know how/what/why - so if I do some experimentation on my breadboard, I will know this so if I need to install my resistors (well, those particular ones) first before the MOSFET - I will know to do so (and can plan the layout accordingly).
What's the "truth" on this?
I Dunno. MOSFETs are generally the sensitive things, even within chips (essentially all modern chips are build out of MOSFETs), because they have a thin oxide layer designed to hold off about 20V, and it's not good to discharge a static charge (1000+ V) though it; it leaves holes behind that interfere with further correct operations. But I don't know how a typical high-power MOSFET (high power ought to mean pretty sturdy, right? But on the other hand, they don't have any other circuitry around to protect them) compares to a typical IC (which I generally regard as pretty robust in spite of very tiny MOSFETs, but then there's lots of other circuitry there...)
I'm usually inclined to install passive components first in a PCB anyway, just because of the heat issues...
The gate on a MOSFET apparently can become charged/discharged via the small internal leakages if it isn't positivly held in the desired rest state using a resistor. If the MOSFET is in a partial "on" state, it has a fairly high internal resistance, which can result in it smoking if tied to a high current capable source. I've seen N MOSFETs start to conduct if the gate is allowed to float.
zoomkat, your answer makes sense, and seems closest to what I was reading about - but that seems to only apply if you are building a circuit, and apply power before you have connected the gate up (you are talking about a pull up/down resistor, in effect, right?). This makes sense, but it felt like what I was reading was saying not to install the MOSFET until you had the resistors in place, nothing about powering it up. Maybe I misread, or the author wasn't clear (it didn't seem right to me that something would spontaneously fail before you had everything hooked up).
Though - with what westfw is saying - maybe you need to have everything set up (resistor on gate) before connecting because of stray static charges causing the failure? It sounds like ultimately I need to keep myself well-grounded when working with these devices, while taking into account the gate resistor thing.
Thanks for the explanations.
A 5kv static charge generated while walking across a carpet has little current capability, but at least a couple of electrons can blow thru the very thin material surface in a MOSFET, possibly causing a permanent fault in the surface material. Floating inputs on semiconductors can cause unexpected behavior. I made a video switcher using a 4017B chip, and when the clock wire was left to float, it would start switching on its own when my cat got within ~1" of the wire.