next topic »

[on:]

Hello, I've dismantled a indoor/outdoor thermometer and I'd like to hack it and take control of it via Arduino.

Will it be possible to "take" control of the screen and read the sensors? Won't be a problem to remove the sensors to easily fit them on the arduino, but I'm puzzled about how to drive the liquid crystal display.

So here is the "thing" :



1. Just a switch that change the unit from celsius to fahrenheit
2. Switch to show Min temperature
3. Switch to show Max temperature
4. Unknown and unlabeled, what is this guy?
5. Resistor (S8050)
6. Switch to turn on a led for backlight illumination
7. Temperature sensor
8. I believe a Microprocessor?

Now, most of the things in the list can be ignored I believe, but how can I take control of this display?
All this run on a single AAA 1,5V battery, will the Arduino with his 5V just burn this out?

Thanks for help!

[Reply #1 on:]

4 is a crystal oscillator.  5 looks like a transistor, not a resistor.  :-)  8 is most decidedly a microprocessor.

You're not going to be able to cooperate with the existing micro, so you may as well plan on removing that and replacing it with the Arduino.  Before you do that, you need to analyze the signals going around and determine what voltage level is "safe".  If the temp sensor and transistor both have readable part numbers, look up datasheets and find out how to power them.  Trace the circuit and find out what connects to what.

This isn't trivial, and you'll probably ruin at least one while you figure it out.  If you're OK with that, dig in and next time "don't do that", whatever it is you did to break the first one.  ;-)

[Reply #2 on:]

The whole thing is powered by 1 AAA battery, so I believe 1,5 volts?

The only thing with a part number is the transistor, how can I "analyze" the flow of information within the board?

[Reply #3 on:]

Hi,

When I have been using simmular things in the past, to trace them out, I use a battery such as a 9V one and a resistor, but in this instance I would not reccomend using this approch, I would in this instance use a AAA battery so that you can be fine with the (logical) voltage levels.

Now in order to test the displays (I do this with 16 segment / 7 segment displays) I pic one of the "pins" as a ground and run the positive across the other pins and make note of the result.

If you get a segment flash, this is a good sign it means you have the ground right (most of the time).

If you get nothing, it could be one of two things, 1st is that you havent got the ground / power right (most common), or the didplay is fried, not often the case, deffinately if you do it quick.

Once you have these you can drive the display as you want.

[Reply #4 on:]

Once you have these you can drive the display as you want.

Not really - bare LCDs use an AC signal, not DC. If you apply DC (for longer than the brief instance you mention), you end up destroying the LCD (well, that segment). PWM'ing the signal won't be enough, either (as that only fluctuates between 0-5 volts DC); what you would need to do (minimum) would be to bias the signal in the negative direction by 2.5 volts to get it to fluctuate between -2.5 and 2.5 volts; depending on the display, this may or may not be enough to drive the segments (and thus an amplification stage would be needed).

[Reply #5 on:]

Once you have these you can drive the display as you want.

Not really - bare LCDs use an AC signal, not DC. If you apply DC (for longer than the brief instance you mention), you end up destroying the LCD (well, that segment). PWM'ing the signal won't be enough, either (as that only fluctuates between 0-5 volts DC); what you would need to do (minimum) would be to bias the signal in the negative direction by 2.5 volts to get it to fluctuate between -2.5 and 2.5 volts; depending on the display, this may or may not be enough to drive the segments (and thus an amplification stage would be needed).

Really, that is news to me, I has assured they were DC in the past. Everyday is a school day.