Suggestions welcome for Antenna for tranceiving 433 MHZ over 1000 meters

[speedEC]

Hello friends,

I am using Si4432 RF Module (bought thru' Ebay.in) in my project. I can able to send and receive on 433 MHz up to 100 meters approximately using the spiral antenna (30mm (h), 5mm (dia), 25 no turns = Total length = 250 mm approx) came along with Module. But I want to transmit and receive up to 1000 m. Transmitter power is +18.5 dbm at 80mA current (tested and Confirmed). Receiver sensitivity is -117 dbm (as per datasheet at 1200 bps). Should I use high gain antenna on receiver side in order to receive up to 1000 m? I came to know from the "google"

that Antenna Length should be proportional to the distance travel required.
For example, to transmit 433 MHZ to 1000 m distance,
The Antenna Length = 1000m/433 MHz = 2.30m = (2.30/4) (i.e lambda/4) = 0.57 m = 57 CM.

Any amplifier available for increasing transmit power to 3W or so as to achieve 1000 m using small antenna at receiver side? Antenna at receiver side will be located at LOS (Line of sight). Can any one recommend the suitable Antenna to satisfy my requirement?

thank you all,
pmk

[FvM]

Your lambda calculation is flawed. It's the ratio between speed of light and frequency, 0.59 m for 433 MHz. λ/4 is 173 mm.

A λ/4 antenna rod with sufficient ground plane or a λ/2 dipole for transmitter and receiver is the usual solution. Antennas with higher gain are not permitted for license free 433 MHz band, neither power boosters. Not sure if 18.5dBm is legal in your area.

[speedEC]

FvM, you are right. it is 173 mm. ok. My PCB is single sided of 60mm x 70mm with GND Filled. The Si4432 Module connected to my PCB through berg connectors. Antenna pin located on Si4432 not on my PCB. As I said earlier, my spiral spring lamda/4 antenna length is 25 CM approx.

RF SETTINGS: CFREQ= 433 MHz,
+/- 30KHz deviation,
datarate 1.171875 Kbps (1200 bps),
BW = 64.1 KHz,
IF = 937.5 KHz.

But, I can able to realize only up to 100 meters.

Any suggestions?
pmk

[c_mitra]

You will need a line of sight with a dish.

If you get at 100 m (max) the signal will be 100 times less (20dB) at 1000 m.

Else you need to amplify the transmission power to more than 1W.

[speedEC]

Both TX and RX is in Line of Sight (LOS).

pmk

[betwixt]

If they are line of sight, presumably they are in fixed locations. Why not make up the extra 20dB or so by using a directional antenna at both ends. A small Yagi should be adequate and inexpensive and would also help by reducing interference at the receiver end. You might be able to use a Yagi at only the receiving end if carefully designed.

Brian.

[volker@muehlhaus]

If you get at 100 m (max) the signal will be 100 times less (20dB) at 1000 m.

So we just need a small yagi antenna with 20dB gain on one end

http://www.m2inc.com/FG43213WLA

SCNR

[FvM]

Yagi Antenna will surely work but doesn't comply with EIRP limits for 433 MHz. I would try with a real lambda/4 or lambda/2 antenna first.

[speedEC]

If they are line of sight, presumably they are in fixed locations.

No. It is not in a fixed location. Slave (Receiver) is a hand held device. It can move around. If the Master (TX) find the Slaves (RX) within its region (Line Of Sight - 1 Km), it will send a message. But it does not need to wait for acknowledgement from slaves (so no need for higher powered antenna at RX end). It is one-way transmission only.

Why not make up the extra 20dB or so by using a directional antenna at both ends

It is not possible to use bigger antennas in RX side. Because it is a hand held device.

I would try with a real lambda/4 or lambda/2 antenna first.

lambda/4 tried. The result was only 100 m. I am going to try dipole (λ/2) antenna. I will post the result tomorrow.

thank you all,
pmk

[FvM]

If you apply Friis free space propagation formula, you get -71 dBm received power for 10 dBm transmitter power at 433 Mhz, 1000 m distance and dipole gain for both antennas. That's more than 40 dB above specified receiver sensitivity.

[volker@muehlhaus]

It is not possible to use bigger antennas in RX side. Because it is a hand held device.

This was only to show you what very large antenna (20dBi gain) you need for 100m -> 1000m, if you go for legal operation and antenna at the receive side. A dipole will maybe give 100m -> 150m increase.

[FvM]

A dipole will maybe give 100m -> 150m increase.

As calculation shows, there's a huge difference between the theoretical range and your tests. The question is, what are the shares of bad spiral antenna efficiency and matching losses, obstacles blocking line of sight, RF interferences reducing the sensitivity.

[speedEC]

As calculation shows, there's a huge difference between the theoretical range and your tests. The question is, what are the shares of bad spiral antenna efficiency and matching losses, obstacles blocking line of sight, RF interferences reducing the sensitivity.

Yes. I also looked in to FRIIS calculator. Free Space propagation loss at 75.17 dbm. So, if TX is at +18.5 dbm, then at 1000m the Rf signal strength should be at -56.67 dbm (+18.5 dbm - (+75.17)). The receiver sensitivity at -177 dbm as per datasheet. It well above the RX sensitivity. but, I can't able to realize more than 100 m. So, as you said, I have to look in to Spiral Antenna, trying with Dipole Antenna, etc.

And also, In Si4432 datasheet, they recommend to set Sgin bit of non-default register 0x69 in order to work AGC (Automatic Gain Control in RX mode) correctly. I will set and test. Let us see whether this will make any difference.

thank you,
pmk

[speedEC]

Edit:

....The receiver sensitivity at -117 dbm as per datasheet....

thank you,
pmk

- - - Updated - - -

[AndreyG]

Assume your link is line-of-sight. If so you are 40 - 50 dB short of link budget (following your own calculations). Something is wrong in your system. First you figure out what is wrong.

Is your link really line-of-sight? How high radios are above the ground? What RSSI you get reported from your receiver at the fringe of the link? What happens when you elevate radios?

Perform conducted sensitivity test: Connect Tx and Rx with long cable (use shielded chamber or put Tx and Rx in different rooms to avoid leakage through the air); insert attenuators. Increase variable attenuator until you stop receiving. Figure out what is power level on Rx terminals. Register RSSI reading at or before the cut-off. This should give you initial idea what is happening.

[speedEC]

Is your link really line-of-sight?

Yes

How high radios are above the ground?

case 1: 10 foot
case 2: 3 foot

What RSSI you get reported from your receiver at the fringe of the link?

RSSI = 0x54 approx when Both TX and RX placed between 1 foot. reading RSSI when RX receives signal from TX.

Note: RSSI Chart from datasheet attached. As per this chart, RSSI Value should be above 200 ( > 0xC8). Am I right?

Thank you,
pmk

[AndreyG]

I do not know. You have a graph - figure it out.
0x54 corresponds to 84 decimal, according to your graph it is ~-80 dB. At 1ft?
Free space path loss calculator:
https://www.pasternack.com/t-calculator-fspl.aspx
gives us for 400MHz at 1m 24.5dB path loss.
start by taking few points at few close distances (1ft is probably too close - within 1 wavelengh. Start with 1m and step back) . Do RSSI readings make sense?
Switch to cable.

[speedEC]

Dear friends,

I have tested with Various Antenna and stored results in xl sheet. it is attached FYRef.
The RF settings are:
CFREQ = 433 MHz, DEVIATION = +/- 20 kHz, Data Rate = 1.171875 kbps (1200 bps), BW = 42.7 kHz

I read the RSSI value at the time of Sync Word Interrupt. Now I can able to correct RSSI value at Receiver. The distance I can able to communicate is far more than 150 meters and I am getting signal still at -80 dBm. So, I have to increase the test distance and post in a day or two.
I came to know that, as FvM said, Lambda/4 Antenna gives a good result. But copper wire with 1mm dia and 17.5 CM antenna gives much more good result than other types including dipole antenna. I have to check one more time with dipole antenna, however.
Can we able to reduce the data rate to extend the distance further. In datasheet, it says The Modulation Index should be between 1 and 32. Modulation Index = (2 x Freq deviation)/datarate (in kbps). I read some other document which recommends, faster data rate with retransmission option is good than slow data rate.

Your suggestions are welcome.

thank you all,
pmk

[biff44]

If you reliably want to get out to 1000 meters, you need a different transceiver. Try a Semtech heavily coded modulation scheme

www.semtech.com/IOT