Okay, I want to make this clear - I am from computer science background, and would require more explanation than an average electrical guy.

The scenario, I have piezoelectric and LDR sensors, which give analog output of course. I plan to transmit the output from this using a serial 434Mhz RF Module, which will be received by a Raspberry Pi on the receiver side and processed further. Now, I am using HT12E encoder to encode 4 lines of parallel data, which according to my knowledge is 4 digital inputs, each of which I plan to use for 4 different sensors. I have setup the transmission part, and got the RF working.

The question, can I convert the output from the sensors to digital output, without using a micro controller, and send it to HT12E to encode, if so, how can I achieve this? If I have to use a micro controller, what would you recommend, Arduino or RPi, portability is important for me, and as far as I know, RPi doesn't have inbuilt ADC and is much bigger? Ideally, I would like to use the first method without micro controller mainly for portability and power consumption reasons. Please ask me about anything if I haven't been clear.

  • \$\begingroup\$ There are stand-alone ADCs.. \$\endgroup\$ – Eugene Sh. Jan 30 '17 at 17:48
  • \$\begingroup\$ ... and the easiest way to deal with standalone ADC's is with a microcontroller. \$\endgroup\$ – Scott Seidman Jan 30 '17 at 17:50
  • 4
    \$\begingroup\$ What you want (do it without a computer of any kind) is certainly possible, but I doubt that you can do it (which is, strictly speaking, waht you are asking). And I think no-one who can do this would do this, because they would take the easier route: using a micro-controller. \$\endgroup\$ – Wouter van Ooijen Jan 30 '17 at 17:55
  • 1
    \$\begingroup\$ There are so many different microcontrollers, and just about all of them would work for you. \$\endgroup\$ – Scott Seidman Jan 30 '17 at 18:01
  • 1
    \$\begingroup\$ Re: "without micro controller mainly for portability and power consumption reasons", you are throwing out microcontrollers for the wrong reasons. You can get very tiny MCUs at very low power ideal for a battery powered device. You can put the MCU to sleep, wake up, convert analog to digital, send a radio burst, then go back to sleep. \$\endgroup\$ – Vince Patron Jan 30 '17 at 18:22

You want to use a microcontroller. Microcontrollers can be tiny integrated circuits. This is the right approach.

Arduino and Raspberry Pi's are not microcontrollers. The Raspberry Pi is a single board computer, and the Arduino is a microcontroller development board, associated with a "wiring" development environment.


Before we had lots of small microcontrollers, we had voltage to frequency converters. Perhaps that would work for you.

  • \$\begingroup\$ In simpler cases, indeed it would (choose the right frequency range and it can be debugged by ear or recorded on audio tape, too). But it's going to be a challenging fit for the desired channel multiplexing. Could potential do a series of pulse positions, like an analog hobby-grade radio control set does... but while traditionally done with small scale integration, it's now been a while since anyone sensibly implemented it other than with a microcontroller. \$\endgroup\$ – Chris Stratton Jan 30 '17 at 18:40

If you are connecting four sensors to four digital inputs, it seems fairly obvious that you only expect to get binary data (on/off, 1/0, active/inactive) from each one.

In that case, all you need is a quad comparator chip, with the threshold on each comparator set to an appropriate level for that sensor.

However, there are additional issues, such as how often you need to sample the data in order to capture all of the information you're interested in. You need to provide additional details about what that information might be.

Especially when using a low-bandwidth radio link, it usually makes a lot of sense to use some local intelligence — i.e., a microcontroller — to convert the raw data into the information you're really interested in.

  • \$\begingroup\$ I am planning to use the sensors along with an accelerometer to detect crashes on an helmet. So, I need to sample the data almost continuously. \$\endgroup\$ – Ketan Shenoy Jan 30 '17 at 18:25
  • \$\begingroup\$ Are you merely detecting that a crash has occurred, or are you also trying to measure parameters of the crash, such as direction and acceleration? How much information do you need to transfer, and how much latency can you tolerate? What kind of accuracy are you looking for? \$\endgroup\$ – Dave Tweed Jan 30 '17 at 18:35
  • \$\begingroup\$ @KetanShenoy you've probably chosen the wrong components. A small micro controller and one of the higher data rate, already packetized 2.4 GHz modules will likely give better results, potentially with less software complexity. You may end up wanting an MCU on the raspberry pi end too, as it may be easier than correctly manipulating the radio with the necessary timing under Linux, and can also let you do development test with your desktop machine playing the role of the pi. \$\endgroup\$ – Chris Stratton Jan 30 '17 at 18:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.