I'm working on a sensor which purpose is to measure the water level of a sinkhole without getting in touch with the wastewater. The Sensor is mounted beneath the manhole cover --> The underside is formed like a rhomb and the sensor fits perfectly in the center bulge --> something like this --> similar Manhole Cover. The whole system is attached to the Manhole cover with strong magnets. The heart of the sensor is an ESP8266-12F. The whole system runs on 3,6 Volt.

An ESP8266-12F measures the distance between manhole cover and waterline via ultrasound (HC-SR04) uploads the measurement to Thingspeak if its able to locate a known WiFi nearby and additionally transmits the measurement via 433 MHz to a Receiving Unit. The receiving Unit uses and Arduino micro, a RTC modul and a 433 MHz receiving unit.

433 MHz transmitting and receiving unit are equipped with those antennas AntennaP Picture This are the 433 MHz Receiving and transmitting Units: 433 MHz Superheteodyn I want to achieve approximately 10 Meters transmitting distance from the manhole cover to the receiving unit.

I tested the system in my flat and on numerous distances without any problems and was really confident that an obstacle like the Manhole Cover + 10 Meters Air wouldn't be any problems. I knew that getting the signal out of the sink would be the tricky part because of all concrete and steel and I've got pretty much no experience with radio frequencies and the way they behave.

Today I was able to test the whole system --> The upload to ThingsPeak works like a charm. The WiFi has to be within 10 Meters from the manhole cover so that the ESP8266-12F is able to establish the connection. The 433MHz connection however fails. I need to be within 2 Meters of the manhole cover to get it to work.

Question: What could be the reason for this bad 433 MHz result. Why is the 2,4GHz Signal so much better than the 433 MHz signal ---> I thought when it comes to obstacles a lower frequency is better? Are the antennas bad? Can I improve them? Would it be better to use a long Cable with 17 cm? (Lambda/4) Would it work better with higher VCC than 3,6 V. As far as I know the transmitting unit isn't supposed to go over 3,8 V. The receiving unit runs on 5 Volt from the arduino micro?

  • 8
    \$\begingroup\$ The 433 MHz radio you link is a much less robust design using a crude modulation scheme, operating at lower transmit power (perhaps limited by regulation) and with a less sophisticated receiver. No conventional radio will transmit "through" an iron manhole cover, so any path you are getting is either around, or through holes in it. \$\endgroup\$ – Chris Stratton Jun 17 '17 at 23:30
  • \$\begingroup\$ Get composite material manhole covers. \$\endgroup\$ – Jeroen3 Jun 18 '17 at 13:08
  • 1
    \$\begingroup\$ We get so many questions of the nature "I found this 433MHz module in a packet of corn flakes but it turns out it has really poor range". The reason why might just be that the module itself is complete crap - you get what you pay for. \$\endgroup\$ – Lundin Jun 19 '17 at 6:20
  • \$\begingroup\$ The ESP8266 Module comes at nearly the same price as the 433 MHz modules ... 2€... so my questions was why is the 2€ 2,4Ghz signal maker better than the 2 € 433MHz signal maker..... what high quality 433 MHz Equipment would you recommend for arduino/raspberry projects? \$\endgroup\$ – Peter S Jun 28 '17 at 17:02

The antennas that you are using for the 433 MHz link are very marginal. While there may be regulatory issues at play, if you can increase the gain of both antennas, you would see far better performance on the 433 MHz link.

At the very least, build a 15 dBi gain yagi antenna for the receiver and point it at the transmitter.

  • \$\begingroup\$ I'm going to try a simple 173mm antenna to see if that changes anything. Any recommendations on good professional 433 MHz Modules for those kind of endeavours? I can only find the classic Chinese stuff. \$\endgroup\$ – Peter S Jun 18 '17 at 15:09
  • 1
    \$\begingroup\$ If you use that type of antenna, make certain it has an equal radius ground plane. It should be oriented in the same axial direction as the transmitter antenna. \$\endgroup\$ – Glenn W9IQ Jun 19 '17 at 1:38
  • \$\begingroup\$ I added a simple 173mm antenna to the transmitting and the receiving unit and now I reach the same distance as with the 2,4GHz signal. \$\endgroup\$ – Peter S Jun 28 '17 at 16:58
  • 1
    \$\begingroup\$ Good work. Enjoy the project. \$\endgroup\$ – Glenn W9IQ Jun 28 '17 at 17:01

There will be gaps around the manhole cover, either air due to sloppy fit, or in the concrete/bricks. The 2.4GHz exploits those gaps with its 100cm/(11 * 4) = 100cm/44 = 2.2cm quarter-wavelength to leak out.

The 433MHz, 6 times longer in wavelength, has exponentially tougher problem in "leaking out".

EDIT Here is link to prior answer on shielding, explaining the exponential behavior. Why are many IR receivers in metal cages?

  • \$\begingroup\$ Thank you for that explanation. I did not realise that 433 isn't as good in this particular scenario than 2,4 GHz. \$\endgroup\$ – Peter S Jun 18 '17 at 15:07

You might be able to make this work by using the manhole cover as the antenna. A 434 MHz, the wavelength is 690 mm or 27 inches. If the manhole is about half that in diameter, you might be able to resonate the cover in one direction.

This won't be a very efficient antenna, and the impedance will likely be much less than the transmitter wants to see, but you might get enough signal to the top side to be useful. 10 m reception range is probably too optimistic if you want to stay to legal transmitter power levels. However, holding a receiver while standing next to the manhole, or with a receiver on the underside of a truck, might just work.

  • \$\begingroup\$ That sounds interesting. How would I do that the easiest way? I can't just solder a wire from the antenna port of the Module onto the manhole cover can I? \$\endgroup\$ – Peter S Jun 18 '17 at 15:52
  • 1
    \$\begingroup\$ You'd probably screw-clamp the two antenna leads to opposite points near the edge of the circle. \$\endgroup\$ – Olin Lathrop Jun 18 '17 at 23:50

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.