want to build a power supply; don't want to die

Started by MmmPedals, July 06, 2010, 11:38:39 PM

Previous topic - Next topic

MmmPedals

I plan to build one of these http://tonepad.com/getFile.asp?id=111
i am going to wire the 12v ac pads to a power jack and plug a wall wart into it.
My questions are 1) do i have to worry about the 1000uf cap storing enough charge to hurt me? (although i hope to never open it after i plug it the first time i want to be prepared for a debug.)
                         2) Is the metal enclosure necessary?

stringsthings

Quote from: MmmPedals on July 06, 2010, 11:38:39 PM
I plan to build one of these http://tonepad.com/getFile.asp?id=111
i am going to wire the 12v ac pads to a power jack and plug a wall wart into it.
My questions are 1) do i have to worry about the 1000uf cap storing enough charge to hurt me? (although i hope to never open it after i plug it the first time i want to be prepared for a debug.)
                         2) Is the metal enclosure necessary?

the answer to both questions:

it depends upon how much you value your personal safety   :icon_biggrin: .... and your threshold of pain ...  :icon_biggrin:

seriously:  never take AC voltages for granted .... if it's plugged into the wall, it can hurt you ....

PRR

> do i have to worry

No.

First off: "wall warts" (with few exceptions) are "safe enough for idiot users".

Oh, with US wall-plugs you can always get your finger on the wall-prongs while half-in and get a shock; but that's harder to do with a bulky wart than a slim lamp-plug, and you haven't been killed by a lamp yet.

Second: 12V shocks are "generally safe". If your skin is dry you can't even feel it. When sweaty, you can feel it, but 12V on human skin appears to be quite benign.

Yes, you -can- find trouble. You know that licking a 9V battery a while tingles then numbs your tongue. And cats (smaller mouths) who chew wart-cords can be very seriously mouth-burned. If you open your chest and get good contact to the heart, 12V could make real trouble. So don't do these things. Most folks won't.

And in this specific case: the stored charge on the 1000uFd is (if correctly wired) drained by the chip and set resistors (and this plan seems to want you to trim "live"). Once the wall-power is removed, 63% of the stored voltage drains each 2 seconds. So starting from 15V: 5.5V after 2 seconds, 2V after 2 more seconds, 0.7V after 6 seconds total. Killing yourself with 0.7V is not gonna happen.

This project is LESS dangerous than working on your car. The car battery is a similar 12V, and shock is nearly unknown, but if you get your wedding ring across the car battery it will glow white hot and burn your finger to the bone. Even the cigar lighter outlet (where you plug your iPod, cellfone, etc to charge) can melt metal to skin (I know). A large wall-wart or 12V 1000uFd would hardly pit the plating.

It does not HAVE to be a metal case. Unless you plan to step on it, crack it, and leave it for the cat to lick. I have no cat, but I have a low opinion of most plastic cases in high-abuse homes, so I'd be thinking metal. Also: you do have 12V AC, a low audio frequency, and nearby you have very precious 0.02V audio signals. If you set a plastic stompbox on a (unshielded) plastic power supply, it may hum/buzz bad.

One thing about a metal box: in this plan the "12VAC" jack must _NOT_ ground to chassis. Mount it, then use your ohm-meter to be sure neither terminal conducts to chassis.
  • SUPPORTER

igerup

If you open your chest...   ;D  :icon_biggrin: :icon_mrgreen: :icon_lol:   

Otherwise very good advice.

amptramp

It is a standard for military work that anything below 30 volts is benign.  Most aircraft have the equivalent of two 12-volt lead-acid cells in them in series in the electrical system and at full charge, they are 14 volts each, so the system is called 28 volts.  It is possible during power switching to get transients above 30 volts, but this is not a safety issue as these voltages come down by themselves.

Note that the time constants posted by PRR are in the form of resistance x capacitance = time to decrease to 37% of operating voltage.  If you have a 1000 uF cap and 2000 ohms of resistance, you get the 2-second time constant he is talking about.  Less resistance means shorter time but more 120 Hz ripple if the cap remains the same.

stringsthings