Pajero Mod 1.0

Monday evening, when got in my car going back home, I suddenly got the itch to modify the car a lil.

It is annoying that there is no USB port at least for charging:

So I decided to add USB port for charging, and while I'm at it, but some blue LEDs :)

I first removed the passenger drawer

and got access to all the wire-mess inside, yay!

Air Bag!!

anyway, I started with the LED for door,

the LED is super bright and is rated for 80mA, the supply is 12v

R = V/I = 12/0.08 = 150 ohm

I soldered 150 ohm directly to the LED and tried to make the assembly in similar form-factor as the original bulb, the result:

amazing, now to my USB charger, the circuit is like:

and I added hot glue to make sure no short circuit happens

Then fixed the LED to place:

awesome!

and the USB:

now I can charge my phone :)

Battery in Salt Solution

For some reason or another, I got this silly idea of dropping a battery in a water saturated with salt :)

I thought, the salt will act as a conductor with resistance ~ 0 and Imagined all the sparking (really?! spark inside water! just imagine! :P)

I started with the Test Aperture :D

1- Resistance test leads



I cute two ~ 3 inch of steel wire and stripped out 5mm of insulation from both ends and soldered one end to binding post



which I fixed to a glass bottle's cover









And here I got some interesting results

the probes were about 7cm apart from each other (the bottle diameter is about 9cm)


resistance with tap water ~ 70K ohm
resistance with salt water ~ 14K ohm

what a finding!!!!!

This means with the battery inside the water, it won't effectively be a short-circuit, no sparking!!!!


it was kinda depressing, but I continued, I added temperature sensor (LM335) to MCU (ATmega168) to serial port on PC and then I logged the data


Once in water, I saw small bubbles forming on the negative pin and then going out, this continued and continued and continued without spark!!!, after sometime I got bored and stopped the data recording.


you can find the recorded raw data here : Temperature Readings

I graphed it on excel and I found this:



the x-axis numbers are just the index numbers, they dont mean anything, but it is actually for about 16 mins of continues data recording, I assume the ADC clock is maximum which is half the MCU clock which was 16MHz so the ADC is 8MHz

the y-axis is the decimal value of the ADC, so you can go:

V = 5*(y/1023)

V is the actual voltage read by the ADC
5 is the ADC reference voltage
y is the number from the file (the decimal value)
1023 is the ADC counts, for 10-bit ADC (2^10)

then to get the actual temperature, you use the formula from the data sheet of LM335 which is

V = 10mV / K

so

T (K) = V * 0.01

and you get the temperature in Kelvin, however, you might need to do initial calibration with a know temperature and then you subtract/add the calibration value to all other measured values.

Since I don't have this equipment, I used the raw decimal value to see the trend


here, I put the bottle outside and went to sleep, the next day:

what a mess!

of course, everything including the poor LM335 inside will all go to be disposed  off the universe!

Shall I count it as a fail?!
why the resistance was large?!!
How can I decrease it?!!