Can ADC output be directly fed to a 433mHz transceiver without encoding/decoding?

Thread Starter

Orly____

Joined Oct 6, 2020
1
Hi,
I'm new to the world of electroncis and I'm trying to get my head around transmitting wireless signals.
I was pretty sad when I discovered most of those modules (nRF24, LoRa, 433mHz etc...) work only with microcontrollers such as arduinos, and if you don't have one you should use serial encoders/decoders.
All I have is a hall effect sensor with an analog output, so my problem is to transfer this analog reading wirelessly.
Converting the signal to a digital value is pretty straightforward, but can I directly feed the data to a "serial input" pin of one of those modules?
 

ronsimpson

Joined Oct 7, 2019
1,025
Part of the problem is how do you receive the signal. (decode, process)
Some of the little transmitters do have a computer.
The analog voltage from the sensor needs to do something to the 433mhz transmitter that can be picked up by the receiver.
One option is have the analog voltage drive a "voltage to frequency" circuit. So 0V=100hz tone to 5V=1khz. Then that tone drives the transmitter. On the receiver you count the frequency and know what the sensor is doing.

I can not remember but I think the law says that 433mhz transmitters can not send all the time. They need to be on for a very short amount of time. If this is a personal thing then maybe no one cares.

The 433 I played with does not transmit very far.
 

Ian0

Joined Aug 7, 2020
768
If you take a single channel A/D with an SPI output, and the option of internal or external clock such as the LTC2482, and set the clock to "internal" it will send a data stream every time it is triggered by the CS pin. (A 555 could do the triggering)
Then, at the receiver, you have to identify the start of the data, which may be tricky if the data starts with zeroes.
So you could possibly exclusive-or the clock and data to give a Manchester-coded datastream, from which you can extract the clock and data signals.
Solves the "continuous transmit" problem, better resolution than the V-to-F, but leaves you having to decode the received signal.
 

DickCappels

Joined Aug 21, 2008
6,659
...you need to encode the data as it is transmitted because, if nothing else. the data channel is not direct coupled. If you provide a more detailed description of what you want to do then certainly somebody here can help you with a more detailed solution.

As per @ronsimpson in many countries 433 MHz is reserved for brief transmissions signaling that an event has taken place. This is so that many can use the channel and tolerate some interference from others.
 

BobTPH

Joined Jun 5, 2013
2,602
If you are talking about the cheap 433MHz transmitters you can get on EBay and amazon, you need a robust protocol. Simply sending the output dirctly from a serial ADC would not cut it. You would be getting errors all the time with no way to detect them. This is why you need a micro.

To get reliable operation of a remote I built using these I use an extra long start bit (10 periods) to let the receiver AGC adjust, then send the data 3 times + a checksum. I still miss maybe 1 out if 10 commands, which is okay since there is visual feedback.

If you get one and watch the receiver on the scope you will see constant noise pulses at a rate of over 1000 per second, which was the speed I was going for. The long start bit caused suppression of the noise, presumably by lowering the gain.

Bob
 
Top