Got into a techno-junk fug :p

JensAyton · Post by **JensAyton** » Mon Dec 10, 2007 1:24 pm

The aforesoldered phone has given up the ghost; it stopped ringing again, entirely coincidentally after hitting the floor for the umpteenth time, and it wasn’t a simple torn-off wire this time. So I’ve got a new one.

I also got a new phobile to replace the one which went in the washing machine. However, the iPod that went in at the same time is working fine. Pity, I thought I’d have an excuse to get a better one along with my new computer. :-)

I would like to state for the record that this is the first time I’ve put an electronic device in a washing machine unintentionally.

Hoopy · Post by **Hoopy** » Mon Dec 10, 2007 1:39 pm

as long as the battery isn't connected then most electroinc things should survive getting dunked. They use aqueous solvents to clean the circuits during manufacture so you should just have to wait for it dry out. Of course if the battery is connected and has some juice in it then all bets are off...

Rxke · Post by **Rxke** » Mon Dec 10, 2007 1:45 pm

Ahruman wrote:
I would like to state for the record that this is the first time I’ve put an electronic device in a washing machine unintentionally.

JensAyton · Post by **JensAyton** » Mon Dec 10, 2007 1:45 pm

The old phone starts up and mostly functions, but can’t connect to a network (and the 4 key doesn’t work). I’m guessing physical damage or corrosion on the radio. Can’t be bothered to work out how to open it up, though.

And yeah, the batteries were in both. I rarely take the batteries out of stuff I’m carrying around in my pockets. :-p

Arexack_Heretic · Post by **Arexack_Heretic** » Mon Dec 10, 2007 1:48 pm

@Winston
Hmm.

I got a unetched 15x10 PCB lying around somewhere...from a wifi antenna project for my sister a while back...which got shorted out due to her sharing ADSL with her neighbour instead.

One could get creative with that technique.

Don't have any acid lying around though, and currently don't have access to a lab.
What type and molarity of acid you use?

Most components inside MP3 players etc just look like shiny metallic squares (<1mm^2) usually R and C comps...actually I guess those are empty sockets used in other designs using the same PCB layout.

@Ahruman: The circumstances surrounding that 'aqueous event' sound mildly suspicious...

especially since you conveniently provide a motive...

JensAyton · Post by **JensAyton** » Mon Dec 10, 2007 1:59 pm

Arexack_Heretic wrote:
@Ahruman: The circumstances surrounding that 'aqueous event' sound mildly suspicious...;) especially since you conveniently provide a motive...

Nah. The motivation came after the realization of what I’d done, and also shortly after discovering that I had more liquid funds (hah) than I thought. Anyway, these went to a bigger screen (24" iMac, woo), so all’s well, etc.

Arexack_Heretic · Post by **Arexack_Heretic** » Mon Dec 10, 2007 2:12 pm

I'm really a noob at anything technical, I can find and solder a loose wire and understand a simple circuit diagram after pouring over it for hours,
but homebrewing PCBs from scratch is really beyond my capabilities right now.
Just got my first multimeter, but still need to get hold of a small breadboard.

For my first project "frankenstein a razor, printer, MS mouse and loose components into a small moving techno artwork" I almost got all parts I think to need...save the relays to controll the motors.

...
with SM components: I see brown indicates resistor and beige capacitor...is size the only way to deduce their properties?

Post by **winston** » Mon Dec 10, 2007 2:44 pm

Arexack_Heretic wrote:
Don't have any acid lying around though, and currently don't have access to a lab.

I use ferric chloride to etch. It's the least nasty of a nasty lot, and any electronics place sells it (I get it from Bowood Electronics - http://www.bowood-electronics.co.uk - since they stock the easier-to-mix pellets, as well as selling copper clad board at a fraction of the price that most places do).

As for SMD part identification, you'll need a component tester to figure out the properties of the capacitors, the small ones don't have markings and size isn't always an indicator (for example, you can get 100nF capacitors in 0603 size, and 10nF in 1206, which is twice the physical size but a 10th of the capacitance). You could get a good idea of capacitance by rigging up a simple oscillator with a 555 timer, and using the unknown capacitor as the timing capacitor, and deduce the capacitance from the frequency you get out of the 555. Larger capacitors (like SMD electrolytics) usually do have some kind of markings.

Resistors usually are marked though, and with a magnifying glass you can normally read 'em. They usually have a 3 digit code like 102 (which means 10 with two following zeros, so 1000 ohms). There may also be a tolerance marking on them. An intarwebs lookup will tell you how to decode these.

Inductors are generally marked as well.

Arexack_Heretic · Post by **Arexack_Heretic** » Mon Dec 10, 2007 3:04 pm

yeah, i figured one would need an oscillator to figure out capacitor capacity.

(incidently this is where I typically got into trouble with electronics or instrumentallity classes, my math was not up to the task at that time.)

don't have one. Probably never get one, too expensive.

nor a 555timer, but this at least seems to be a common component widely available for a small amount of $$. Probably a usefull component to have also for other uses.

this may be a viable option for me ( a cash strapped beginner):
(sorry it's in Dutch)
http://www.pi4zlb.nl/Zelfbouw/LC-meter.html

Post by **winston** » Mon Dec 10, 2007 4:28 pm

There are calculators on the web now, so you don't have to do the sums

Just so you understand the principles of why you're making the calculation, of course!

There's a web calculator for 555 timers which will accept frequency, R1 and R2 values, and return the capacitance. Just round the capacitance to the nearest standard value. The 555 is at least very cheap!

Arexack_Heretic · Post by **Arexack_Heretic** » Mon Dec 10, 2007 4:34 pm

I'll study your 555timer proposal in some more detail,
the LC-meter mentioned above requires me to first build a PIC-chip programmer... so I can load the software needed to calculate the value Cx as compared to a fixed paralell C (aswell as controll the LCD display).
A nice project, but beyond my current level of operations.

Hmm...interesting.
The simplest diagrams based on 555 I found used a LED to indicate the pulse duration (1/3 chargetime capacity of Cx IIRC). A graph would then be used to correlate the chargetime with the unknown capacity.

Now trying to find a circuit that would make accurate timing of this pulse more automated (as opposed to using a pocketwatch).
But short of buying a cheap stopwatch and slaving the controlls to the timer switches I don't think I can get around Programmable ICs.
...ore use the quartz clock i just accidentally exploded as a 1sec timer.
couple this to a 4-bit counter and voila....maybe
...8-bit would be better for a count up to 15seconds, otherwise even more offset resistors will be needed....If I can find an IC with two counters on....

huh? you saying some 555s take and output frequency? I was thinking of using pulse length of a single pulse, but frequency automatically samples many pulses, hence should be better.

...
messing up that clock made me want to repair it in brass, steamfunk-style!
I always wanted to use a lathe ....better stick to retro-hightech though, less I next wan't to reinvent the wheel.
Gears are so cool though.

Arexack_Heretic · Post by **Arexack_Heretic** » Tue Dec 11, 2007 4:24 pm

Hmm.

Ok.

I wonder whether using a set of resistors ranging from 10K to 10M will result in low reliability? I read that for good timing you be better of with big capacitors in lieu of big resistors.

(adding two in parrallel each time, 10+100K to get 9,09K etc)
t(s) = 1.1 * R(ohm) * C (*10^-6 F)
1G : unavailable
100M : 10-100nF (1.1-11s) [25Euro!]
10M: 100nF-1uF (1/R=1/100M+1/10M: 1-10s) [3ct]
1M: 1uF-10uF [3ct]
100K: 10-100uF [3ct]
10K: 100uF-1mF [3ct]
1K: 1.1mF-11mF [3ct]
highest capacitance I found is 4700micro. But I wonder whether C's of this size will lack markings...
So 100nF will be the lowest capacitor measurable with any reliability using this method. unless a stopwatch is added that can count milliseconds.

Post by **winston** » Thu Dec 13, 2007 1:26 pm

With a 555, you'd wire it as an astable multivibrator (i.e oscillator), and feed its output to a frequency counter. For measuring normal components (i.e. you don't need stunning accuracy!) a frequency counter can be made with a couple of BCD counters and 7 segment decoders, and a 7 segment display.

However, it's probably easier to use an 8 bit microcontroller like an Atmel AVR running a program as your frequency counter. Or perhaps buy a second hand frequency counter off ebay

Arexack_Heretic · Post by **Arexack_Heretic** » Fri Dec 14, 2007 1:42 am

Remember: I have currently a total of 5Euros cash
and several hundreds in overdue bills.

I'd like to buy some copper, but at current prices I can't even afford a strip of 2x20x1000mm.

I've gotten, with some of my last funds, two 555s a sequence of resistors and various other parts.

Post by **winston** » Fri Dec 14, 2007 11:35 am

In which case, you can use your computer's sound card input and write a software frequency counter without spending any more money

Connect the 555's output to the sound card's input (making sure you're running it on a voltage that won't kill your sound card), and it should be pretty easy in software to count the frequency if you make sure R1 and R2 are sized such that you'll get audio frequency out for the capacitances of interest.