Wireless Thermometer Wireless Data

For a school project I am trying to take a wireless thermometer and make it transmit the data I want.

I have figured out how to transmit, but I don’t know how to decode the data.
It transmits on 433.92 MHz.

Here is an audio sample of the transmission.

I have already compared it to the ones here with no luck.

I have had success asking here before.
Any input is appreciated.

What is the spec or the thermometer part number?

The spec sheet for the thermometer would probably have all the info needed to figure out how to use it, but those can be confusing sometimes. If we had some more information (make, model, part number, link to a spec sheet, etc), we could probably help better.

Without that… i would suggest doing a few tests. Start with the thing at room temperature and get a recording of what it sends like you did above. Then stick it in the fridge, let it get acclimated, then get another recording. Transition to the freezer, and get a third recording. For all of these recordings, make sure you also record the temperature at the time of the recording, using another thermometer with a simple digital readout if needed.

You should be able to compare those three recordings and figure out the differences - how does the transmission pattern change as the device is in colder environments? You should be able to plot those changes against the temperature, and see some sort of trend. If the trend isn’t linear, you’d probably want more data points at other temperatures to be sure you understand it correctly. All in all, it sounds like a pretty fun little experiment :slight_smile:

This is the thermometer.
Last I looked I could not find a part number on the casing of the transmitter or receiver.
When I opened up the transmitter there were only two labeled things. The chip was covered in a conformal coating, so no luck getting a part number from that.
One side of the PCB has “TX_(30)” on it. Googled it with no luck. It also has “SAW RF433” which lead me to the correct frequency of 433.92 MHz.
(Not sure if the labels I wrote are accurate I’ll correct them when I get home)

If pictures of the circuit boards would be helpful, let me know and I’ll add a few as soon as possible.

I think you’re saying that you want to use this system to send data that isn’t the temperature information it was designed for. Is that correct?

If so, then you can decide to just use the Radio parts, or the whole system, to send data. If you use the radio parts only, then it is up to you to decide how you want to modulate that data. Look up “AX.25 Packet” for one example I’m familiar with. If you want to use the whole system, then you need to figure out what modulation/encoding scheme the manufacturer has used. It could be something standard, or something they developed themselves.

The best way to find out is to ask the folks who designed it. Identify the manufacturer from either the packaging or the FCC ID Number (all transmitters must have one, and you can link that back to the manufacturer via online search).

Next is to do what eagle suggests: have it send some known data and decode the bits manually using a storage oscilloscope or data analyzer.

The audio doesn’t help much. I can tell that it’s sending some bits, likely without protocol since there doesn’t seem to be a preamble. (“Protocol” means some method of encoding the data do it doesn’t get lost easily. Preamble is a brief burst of non-data bits to fill the time until the transmitter gets up to power and stabilizes)

Just a word of caution here: Just because you have something that sends data from here to there doesn’t mean you can use it to send some other data. If you give us more information on what it is exactly you want to do, and what you think this device will do for you, along with the timeframe for getting it done, we can get much deeper into this.

When I say I know how to make it transmit information I want, that pretty much means I’m going to replace the thermistor with a different resistor or analogue voltage source like a photo resistor or an analogue out of a micro controller.
As for the people who designed it, I cant find them anywhere. However, there was a sticker on the bottom which turned out to be the FCC ID. A search gave me lots of cool datasheets. Even telling me that the modulation is ASK.
I did notice two things though.

  1. There are 6 contact points not soldered to anything that lead directly to the chip on board. GND, p5.6, p5.5, p5.4, RESET, VCC (these look like fun to mess with. I think that they’re a way to reprogram the IC)

  2. there are some slight differences in the circuit of mine and the one in the picture of the FCC. Mine has a capacitor the FCC’s doesn’t and mine is missing a diode that the FCC’s has. My theory is that mine transmits intermittently and the FCC’s transmits constantly. (Their pictures, my pic 1,my pic 2)

The schematic shows that the transmit is simply turned on and off by the processor in the transmitter. Effectively this is just a digital CW signal and the data scheme is coded in the transmit/receive processors. It you varied the temperature or voltage input at the transmit side you might be able to see some correlation to what is actually transmitted. It could be something as simple as a BCD encoding of the transmit carrier. Since the thermometer is used over a relatively narrow range and the resolution is likely 100 degrees or less, the encoded data is not critical. It may even be only necessary to count the on/off transitions of the carrier to determine the data sent. It is also possible to only send absolute temperature when the unit is first turned on and then only send delta temperature since the last transmission. i.e. the first data sent is a long encoded signal of the absolute temperature. The remainder are short bursts indicating the difference in temperature since the last data was sent. This would save battery life as well.

Cool, BCD seems likely.
However, I am having trouble differentiating the 1s from the 0s because all of my audio spectrum software can’t give me a high enough detail picture. The closest i came to success was using a program called CWGet.
Also, I don’t think that it transmits the delta temperature because i can turn the receiver on after a minute or so after the transmitter and it will display “correct” data. (I say “correct” because it wasn’t a good thermometer in the first place. That’s why I took it apart.)

Your audio equipment should do just fine, the data can’t be much more than 2400 or 4800 baud (transitions per second), probably much slower than that. If you are finding high-frequency carrier waves on the signal, you can demodulate it (ASK) with a simple diode (use germanium fo low loss) - google “AM demodulator”. A small (several picofarads) capacitor from signal to ground will shunt any stray RF as well.

This looks like a simple modulation. Carrier on for ones and off for zeros. A morse code reader will not decode this. As Don pointed out, a simple AM detector/diode should give you back some form of digital signal. Then you must store the data somehow to allow it to be read. If the thermometer only reads from 70 to 180 degrees then the data is likely only representing the scaled temperature. 70 is transmitted as a zero and 180 as 100 and then the scaler value is added in the receiver.