Fullscreen HTML5Med Res (??MB)HTML4

Tim Blythman’s 433MHz Digital Receiver Module We recently published our version of the ubiquitous 433MHz (LIPD band) transmitter module, which performs better than many prebuilt versions. Having found a suitable receiver IC, we then created our own version of a matching receiver module, and it is better too! W e wrote about our 433MHz Transmitter Module in the April issue (siliconchip.au/Article/17950). That article discussed the LIPD (Low Interference Potential Devices) RF band, which covers 433.05MHz to 434.79MHz and can be used without a paid licence. There are some simple provisions, including that the EIRP (equivalent isotropically radiated power) must not exceed 25mW. This is of concern for a transmitter, and we explained how our Transmitter Module was compliant. Of course, this should not be a problem with a receiver, so we don’t need to worry about that aspect for this project. These sorts of receivers and transmitters are typically used to send digital data at a low bit-rate (up to around 10kbit/s) to provide a wireless link over distances up to 100m, such as around a home. Typical applications include remote control of devices like garage doors and gates, or for sending data from remote sensors back to a base unit, as might be found in a wireless weather station. Fig.1 shows a block diagram of such a system. As we mentioned in the earlier article, multiple layers of encoding are often used to make the best use of the medium and to allow systems to coexist with others nearby by providing identity and data validation (checksum) features. We noted how the receivers use AGC (automatic gain control) to receive data at differing signal strengths. Simple OOK (on-off keying) means that it is quite straightforward to extract a digital signal from the ambient background RF noise. 62 Silicon Chip Effectively, the receiver keeps track of the average RF signal strength (over millisecond time scales). If the instantaneous RF signal is stronger than the average, a logical high is sent to the data output, while a low output is when the RF signal is weaker than average. (That explains why you get noise from the output when there’s no RF signal; atmospheric noise on that frequency will constantly vary above and below its average.) We’ve used these sorts of transmitters and receivers in numerous projects. The most recent was the Battery Powered Model Train from January (siliconchip.au/Article/17607). John Clarke even designed a 433MHz Wireless Data Range Extender, which was published in May 2019 (siliconchip. au/Article/11615). 433MHz receiver Our earlier 433MHz Transmitter Module is a drop-in replacement for the likes of the Jaycar Cat ZW3100 and Altronics Cat Z6900. It has the same pinout, general size and shape as those modules. The Transmitter Module article has comparative performance tests between our unit and the Jaycar ZW3100. This Receiver Module is intended to be a substitute for the corresponding receivers, Jaycar’s Cat ZW3102 and Altronics’ Cat Z6905A. If you compare the photos above and below, you can see that we have aimed for the same pinout and size, but you’ll note that our unit has an extra pin (for RSSI, a useful extra feature). The Transmitter Module uses a Microchip MICRF113 IC and the Receiver Module uses a MICRF220 IC. These are both intended for use in these sorts of applications on LIPD bands, which makes our design task easier. The MICRF220 IC One of the quirks of these receiver modules is that when there is no nearby transmission, the DATA pin will produce a stream of random data. This is analogous to an AM radio (for those that remember radios before they became digital!) playing static when tuned between stations. This can be a source of frustration for those using these modules for the first time, since the decoder (typically the right-hand microcontroller in Fig.1) must be able to separate this background noise from valid data. Fig.1: a matching transmitter and receiver pair form a one-way wireless link to transmit small quantities of digital data. A previous article covered the construction of a transmitter module. Australia's electronics magazine siliconchip.com.au Features & Specifications » Drop-in replacement for Jaycar ZW3102 and similar 433MHz receiver modules » Operates from 3.3-5V » Optional RSSI output (analog voltage, 0.5-2V) » 6mA nominal operating current » Optional squelch feature, configurable by resistors » Faster rise and fall times, less latency than some prebuilt receiver modules » Sensitivity with onboard antenna is superior to other modules with an external wire antenna » Short AGC settling time Fig.2: our circuit is based on the MICRF220 RF chip for receiving digital data on the 433MHz band. It requires a 3.3V supply, provided by REG1. As well as receiving data, it provides an RSSI signal so the controller can determine whether an RF transmission is occurring and how strong the signal is. One simple strategy is to look for signal transitions occurring more frequently than anticipated for the expected data stream; the presence of high frequency components is typical of the white noise that occurs with no signal. So, when frequent transitions are seen, the data can be ignored. The data sheet for the MICRF220 describes it as a “300MHz to 450MHz 3.3V ASK/OOK Receiver with RSSI and Squelch”. Squelch is a handy feature on radio receivers that can suppress the output unless a strong enough signal is received. To maintain compatibility with the older modules, our Receiver Module can be built with or without the squelch feature. It is enabled by simply fitting a single resistor. We made this optional, since some designs may depend on the noise to detect a valid signal. RSSI stands for ‘received signal strength indicator’, and it is exactly what it sounds like. There is an RSSI pin on the MICRF220 that produces an analog voltage related to the received signal strength. The data sheet gives figures of 0.5V for a -110dBm RF input level and 2.0V for -50dBm. This is the extra pin on our design. Note that the theoretical frequency range of the MICRF220 extends well beyond the 433/434MHz band. The components we have selected are intended to optimise the operation for this band; different values are needed for the likes of the 315MHz band, which sees similar use in the USA. The MICRF220 data sheet discusses this in more detail. Circuit details Fig.2 shows the circuit diagram of our Receiver Module. Power comes in through the various GND and Vcc pins on CON1 and CON2. These are chosen to match the pinouts of other receiver modules, so a few are duplicated. The MICRF220 is a 3.3V device, so we have provided a 3.3V regulator Our Receiver Module (shown in the lead photos) is the same size as boards like the Jaycar ZW3102 shown here, but has a couple of extra features. The extra RSSI pin produces a voltage related to the received signal strength. It also has an optional onboard PCB trace antenna. These photographs are shown at 125% scale for clarity. siliconchip.com.au Australia's electronics magazine to allow operation with a 5V supply. REG1, an MCP1700, can tolerate up to 6V on its input. The two 1μF capacitors are recommended input and output bypassing capacitors. The remainder of the circuit is centred on IC1, the MICRF220 receiver IC. Pin 8 (SHDN) is tied low with a 100kW resistor to enable the chip whenever it is powered. Capacitor C10 is an optional part noted in the data sheet. When fitted, it will assert the shutdown state momentarily while the chip is powered on. We didn’t find it was necessary to fit it. Power from the regulator comes into pins 5 (power) and 9 (ground) of IC1, with a 100nF capacitor providing further bypassing. The circuit around the two inductors at lower left is the recommended matching network for the RF signal going into pin 3 of IC1 from the external antenna (‘ANT’) connection. Adjacent pins 2 and 4 are RF ground. We were able to comfortably fit all the required parts in the necessary PCB area, with room to spare, so we added a PCB trace antenna. It can be connected by closing jumper JP1 with a solder blob or 0W resistor. The antenna is about 16cm long, suitable for use at 433/434MHz. Adding the length of the other connected traces, it is very close to the nominal 173mm needed for a quarter-wave antenna at 433MHz. Otherwise, an external antenna can be connected via the module’s ANT pin. June 2025  63 Scope 1: the current consumption of our module, measured with a low-side 100W shunt resistor, is very close to the 6mA noted on the data sheet. It rises slightly when the data output is high. Scope 2: the ZW3102 that we tested only drew 3mA during operation, although its data sheet indicates a maximum of 10mA. Scope 3: the blue trace is a signal applied to a transmitter module, while the green trace is the DATA output from our Receiver Module. The red trace is the output from a prebuilt ZW3102 module. Our module is clearly quicker to respond, with sharper edges. 64 Silicon Chip Australia's electronics magazine Pins 1 and 16 of IC1 connect to a 13.52313MHz crystal and its loading capacitors. Like the Transmitter Module, this circuit uses a ×32 PLL (phaselocked loop) to generate a reference frequency. You might notice that the crystal for the Receiver Module is a different frequency to that on the Transmitter Module (13.56MHz). That is because the MICRF220 uses an IF (intermediate frequency) demodulator. The PLL frequency is mixed with the incoming RF signal to produce a signal with a frequency about 1MHz lower. This lower-frequency signal is easier for the IC to extract the data from. Pins 7 and 11 (SEL0 and SEL1) select the demodulator bandwidth. We have chosen the 13kHz low-pass filter setting by leaving both of these pins to be pulled high by their internal current sources. Fitting a 10kW resistor for either or both of R1 and R2 will change this setting. Pin 13 of IC1 (CAGC) is connected to a 470nF capacitor; this value is also dictated by the data sheet and the bandwidth setting described above. The level on this pin sets the gain of the internal amplifier; it is part of the AGC control loop. This capacitor value ensures that the AGC responds at the correct rate to allow the data of interest to be received. The RSSI signal from pin 14 is internally derived from the CAGC signal by being inverted and buffered. It is fed to the extra pin on CON1 via a 1kW resistor. This protects the chip from potential short circuits. The capacitor on the pin 12 (CTH) provides bypassing of an internal reference voltage that is used by a comparator to generate the output on pin 10 (DO). Like RSSI, the DO output is protected by a 1kW series resistor between it and the external DATA pin. Pin 6 (SQ) enables the squelch feature. When left open, an internal pullup current disables squelch. Fitting R5 will pull the pin low and enable squelch. For any of pins 6, 7 or 11, the pullup is around 5μA, so a resistor of 10kW or lower will be more than sufficient to overcome the pullup. Operation The MICRF220 uses around 6mA when configured for 433MHz operation. At this level, the dropout voltage siliconchip.com.au of the regulator is less than 100mV, so it will not have much effect on the output voltage, even if a 3.3V supply is used. Most 5V microcontrollers we have seen will happily accept 3.3V logic levels, and the MICRF220 works down to a 3.0V supply voltage, so this Receiver Module will be suitable for 5V and 3.3V systems. If you are considering changing the SEL0 and SEL1 settings by adding resistors R1 and/or R2, you should check the MICRF220 data sheet closely as some other parts may need to change values. You shouldn’t need to do this, as the default bandwidth settings should work fine with lower data rates. Comparisons We thought it was important to describe the operation of our Receiver Module and the MICRF220 because it has quite an impact on the performance of the Module compared with other receiver modules. We compared our Receiver Module (using its onboard antenna) to the Jaycar ZW3102 fitted with a simple wire antenna. We used our previously described 433MHz Transmitter Module as the RF source for the tests. Our first test was to confirm the operating current of the modules. The MICRF220 data sheet notes a typical current of 6mA; the MCP1700 has a quiescent current of 1.6μA, so it does not contribute significantly to the Receiver Module’s consumption. We rigged up a 5V supply and a 100W resistor as a low-side current measuring shunt on a breadboard. The breadboard allowed us to change between the two modules without otherwise altering the circuit. Our Receiver came in right on 6mA, as seen in Scope 1. You can see that the current does come up slightly when the output pin is high; it reaches 6.1mA. The ZW3102 measured just under 3mA, whether its output was high or not (Scope 2). Interestingly, its data sheet notes a 10mA maximum, so there may be more variability amongst these modules. This shows us the latency, or delay, between the input and output. The output of our Receiver Module is not only faster (24μs vs 28μs on rising edges and 24μs vs 34μs on falling edges), but more symmetrical and it also has sharper edges. We monitored the output of both receivers when a 1kHz square wave was applied to the transmitter’s DATA input. Scope 3 shows the falling edge of a pulse on the DATA input, with the two receivers responses following. We also performed some tests to see how the receivers would respond to different OOK modulation frequencies. As we changed the frequency at the DATA input of the transmitter, we watched the receiver outputs to see how well they followed the input. Above 10kHz, the output is 90° or more behind the input for both receivers, as seen in Scope 4. You’ll see that our Receiver is still delivering a signal that is closer in time to the original signal than the ZW3102. Both receivers are receiving a solid signal at this frequency. Sensitivity You might recall from our article on the Transmitter Module that its output power can be set by altering a single resistor value. This allows us to easily produce weak signals to compare the sensitivity of the two receivers. We performed some tests to compare the relative sensitivity of the receivers. With the two receivers side-by-side on the same breadboard, we monitored how they responded to a transmitter on the other side of our laboratory; this was an Arduino connected to one of our Transmitters to output a typical encoded waveform. The first test was with the Transmitter at full power, and Scope 5 shows an interesting result. Here, we see how quickly the receivers ‘lock on’ to the signal. Our Receiver Module settles its AGC at the correct level a full Scope 4: at 10kHz, a higher frequency than used for Scope 3, you can see the difference in latency between the two modules. This is quite a bit higher in frequency than most 433MHz transmissions we have seen, with 1kHz being more typical. Latency and bandwidth For the next few tests, we rigged up the two receivers side-by-side on a breadboard, allowing them to be seen responding to the same transmissions. siliconchip.com.au Scope 5: the blue trace here is our Receiver’s RSSI pin, while the green trace is its DATA output; the red trace is from a ZW3102. You can see how much more quickly our Receiver locks on to the incoming signal and starts producing valid data. Australia's electronics magazine June 2025  65 Parts List – 433MHz Reciever Module 1 double-sided PCB coded 15103252, 11.5 × 43mm 1 5-way right-angle pin header (CON1) 1 4-way right-angle pin header (CON2) 1 13.52313MHz two-pin SMD crystal, 5.0 × 3.2mm (X1) [Abracon ABM3-13.52313MHZ-10-B4Y-T] 1 39nH inductor, M1608/0603 size (L1) [Murata LQG18HN39NJ00] 1 33nH inductor M1608/0603 size (L2) [Murata LQG18HN33NJ00] Semiconductors 1 MICRF220AY 300-450MHz ASK receiver IC, QSOP-16 (IC1) 1 MCP1700-3302 3.3V LDO linear voltage regulator, SOT-23 (REG1) Capacitors (all M2012/0805 size, X7R 50V ceramic unless noted) 2 1μF 1 470nF SC7447 Kit ($20 + postage): 2 100nF includes all the parts listed here 2 10pF NP0/C0G 2 1.5pF NP0/C0G Resistors (all M2012/0805 size, ⅛W 1%) 1 100kW 2 1kW Extra resistors for option selections 3 10kW M2012/0805 ⅛W 1 0W M1608/0603 OR bridge JP1 with solder Scope 6: using the same trace colours as Scope 5, we see the two modules responding to a weaker signal. The RSSI is lower, and the ZW3102 is producing glitches that are not seen in our Receiver’s output. 15ms before the ZW3102; you can still see glitches in the latter’s output for this time. The blue trace that we have used as a trigger is the RSSI output of our Receiver. Then we used a 1kW resistor to set the output power to 12dB below nominal. Scope 6 is the result of this. The RSSI trace sits at around 1.4V or -74dBm, and our Receiver has picked out clean data, while the ZW3102 is seeing some data but is delivering glitches too. At lower levels than this, we could not see any data on either receiver. This is useful information in that we now know that a level of around 1.4V indicates sufficient RSSI to receive a valid signal. Remember that these tests were done with the Receiver’s onboard antenna; an external antenna should give even better results. While running these tests, we also used a software-defined radio receiver to monitor the relative RSSI. It indicated that these active transmissions were only about 10dB above the background RF level. Squelch We also ran some tests to try out the squelch feature. For these, we simply shorted out the R5 pads on the PCB to pull IC1’s pin 6 low. The data sheet notes the chip will “monitor incoming pulse width before allowing activity on DO pin.” So it doesn’t appear that RSSI is used to control the squelch. Scope 7 shows a typical waveform with squelch active. You can see that there is still activity on the DATA line even when the RSSI is low. It appears that this is where the 13kHz filter is used, as signals at a higher frequency are cut off and do not appear on the output. So the squelch is helpful, but does not completely remove the need to filter out unwanted activity on the DATA pin. Construction Scope 7: even with Squelch enabled, our Receiver still produces the occasional spurious pulse on the DATA line when the RSSI is low. So you shouldn’t expect the Squelch to completely eliminate the need to reject noise on the DATA line, but it helps to reduce it quite a bit. 66 Silicon Chip Australia's electronics magazine The 433MHz Receiver Module uses some small SMD parts, although nothing that can’t be hand-soldered with a little patience. IC1 comes in a 16-pin QSOP (quarter-size small-­ outline package) with a 0.635mm pin pitch, and the regulator is a SOT-23 part. Most of the passives are M2012 (0805) size at 2.0 × 1.0mm, although the two inductors are M1608 (0603) size or 1.6 × 0.8mm. siliconchip.com.au We’ve used M2012-sized pads for the passives throughout to ease construction. Where possible, we have lengthened the pads on the PCB to make it easier to apply solder. This also gives a bit more room between the components. So you’ll need the standard surface-­ mounting gear; a fine-tipped soldering iron and some flux paste are the bare minimum. You should also have tweezers, a magnifier, solder-wicking braid and some good illumination. Your flux will probably also require a solvent for cleanup, although we find that isopropyl alcohol is a good generic option. The Receiver Module is built on a double-sided PCB that’s coded 15103252 and measures 11.5 × 43mm. Figs.3 & 4 are the overlay diagrams that show where the parts are placed. You can also refer to the adjacent photos during construction. All the mandatory components are on one side of the PCB, as shown in Fig.3. Apply flux paste to the pads for all the components on that side. Start by placing the IC over its pads on the PCB, noting the orientation of the pin 1 marker; our chip had a moulded divot, which was easy to find. Clean the iron’s tip and apply a little fresh solder. Tack one lead and check that the other pins are lined up on both sides. If so, carefully solder the remaining pins, cleaning the tip and adding extra solder as needed. Otherwise, use the iron to melt the solder and tweak the chip with tweezers until it is located correctly. If you end up with a solder bridge joining two or more pins, add extra flux paste and press the braid against the bridge with the iron, then gently draw both away once the excess solder is drawn up into the braid. Next, fit REG1, the SOT-23 regulator. It should only fit one way, with its leads down flat on the PCB, so place it, tack one lead and check the position. If all is well, solder the other two pins. If any joins don’t look great, add some extra flux and touch the iron to the pad and pin to refresh the solder. The two inductors at bottom left should be fitted next as they are the smallest remaining parts. We’ve seen some SMD inductors that only have pads on the underside, which makes them a little more tricky to solder. Don’t forget that most SMD parts are siliconchip.com.au Figs.3 & 4: to use the onboard PCB trace antenna, close JP1 with a blob of solder or 0W resistor. The rear of the PCB shows the functions of the external pins. If you wish to enable the squelch function, you can fit a 10kW resistor for R5. These diagrams are at 200% scale. usually designed to be soldered by a machine! You might be able to make out a black mark on one end of the inductors. We’ve fitted our prototypes with the band to the left and on the top, which you can see in the photos. We don’t think it will make any difference, but we recommend you do the same. RF can be strange and we don’t want to tempt the fates! Use the same technique of soldering one lead then the other once the location has been correctly fixed. You can check that the inductors are connected to their pads by doing a continuity test; they should read well under 10W. For the 33nH part, you can probe between the ANT and GND pads of CON2. For the 39nH inductor, probe between ground and the right-hand pad of the 1.5pF capacitor directly above the inductor. If either inductor reads high resistance, add more flux and try soldering each lead again. Solder the crystal (X1) next. It probably will have leads only on its underside, but the PCB pads are generously sized, so they will be easy to press the soldering iron against. As long as there’s flux paste on the pads when you place the crystal, solder should flow between the pads and crystal. Unfortunately, you can’t check a crystal for continuity as you can with an inductor. The remaining mandatory small parts are all M2012 (0805) passives, and they are marked on the PCB silkscreen. Check their values closely against the overlay diagram, since the markings are quite small. There are nine capacitors and three resistors that must be fitted. Be careful not to get the Australia's electronics magazine capacitors mixed up once you remove them from their packages; they will not be marked with values. If you wish to enable the onboard trace antenna, you need to close JP1. An M1608/0603 0W resistor will work, but the easiest way to do this is to generously apply solder with an iron to both of JP1’s pads. The solder mask will cause the solder to bead, but if you add enough solder, you should be able to bridge the pads (you can see we did it in our photos). You can now enable the squelch feature by adding a resistor (we suggest 10kW) to the R5 pads if you want. We have also labelled this with Squelch text. At this stage, we recommend cleaning the board thoroughly with your recommended flux cleaner or another solvent. Allow the board to dry and scrutinise it for bridges and pins not soldered to the pads below. If you see any problems, touch up the board, then clean it and allow it to dry again. Fitting the headers We recommended using right-­ angle headers since they will match the headers found on other common modules. However, you could choose straight headers if you need to mount the Receiver Module parallel to another PCB. It will depend on your planned application. If you are connecting the Receiver Module to an existing design, use two four-way headers. Older designs will not expect a connection for the RSSI pin, so you should leave that pad unconnected (you could run a flying lead from that pad if you want to monitor the RSSI output). June 2025  67 We recommend slotting the two groups of headers into a longer header socket to keep them aligned to the correct 0.1in pitch before soldering (see the photo below). Solder the pins in your preferred orientation, then remove them from the header socket. If you have jumpered the onboard antenna, the external ANT pin does not need to be connected. In this case, all the connections that are usually needed (GND, DATA and Vcc) are at one end of the board, and can be made using a single four-way header. An NPN transistor with a 1kW resistor between its base & emitter could also be used as a threshold detector, as shown in Fig.5. The RSSI signal is fed into the base and the 1kW resistor on the Receiver PCB forms a divider with the external resistor to set the threshold. We tried this out on a breadboard and it worked quite well. You could also bypass the LED and use the voltage at the collector as an active-low digital RSSI threshold signal to a microcontroller or other circuitry. Using it Our 433MHz Receiver Module has some handy features that make it a better choice for new designs. It generally responds more quickly to an incoming RF signal. At the same time, it is backward-­ compatible with older modules for use in legacy circuits that require a 433MHz receiver. The Receiver Module works with 5V and 3.3V systems, which we think will cover most cases. The squelch feature does not appear to eliminate noisy data output during the gaps between RF transmissions, but it does reduce it. We think that the RSSI output will be more useful in testing the validity of a signal on the DATA pin. Our Receiver consumes a bit more current than the ZW3102, but it is still low enough that it could, for example, be powered from a microcontroller GPIO pin, allowing it to be completely powered off if necessary. The Receiver Module is quite sensitive, even when just using the onboard PCB trace antenna, picking up all transmissions that the ZW3102 could with an external antenna. Our design still allows for an external antenna if SC that is preferred. Since it is a module, the usage will depend a lot on your intended project. In general, you should connect a supply of 3.3-5V to one or more of the Vcc pins and one or more of the GND pins. If you have not enabled the onboard antenna, an external antenna should be connected to the ANT pin. As we noted earlier, a 173mm-long wire (including the length of the headers and traces back to the matching network) works well as a quarter-wave antenna for 433MHz. It can be curled or corkscrewed to save space if necessary. We have found that the Receiver Module is capable of receiving nearby signals without an external antenna; you might try this for testing purposes. In general, you should have no trouble using it to replace a receiver anywhere we have specified the Jaycar ZW3102. Conclusion Fig.5: this simple circuit can be used to generate an indication that an RF transmission is being received based on the RSSI. The resistors set the threshold to about 1.2V, which we found to be a suitable level for distinguishing a valid signal from none. Our unit varied around 0.9V to 1.1V when no intended transmission was occurring. This level might be lower in a less urban area than the location of our lab. With an active transmitter, we saw values between 1.3V and 2.0V. This could be measured by a microcontroller’s ADC (analog-to-digital converter) peripheral to detect the presence of a signal. Another option is a comparator set to an appropriate threshold. Some micros (including the 8-bit PIC16F18146) include a comparator The RSSI voltage peripheral that could be used for this The RSSI pin delivers an analog purpose. The micro could then be provoltage between 0.5V and 2.0V, so a grammed to ignore any transitions on microcontroller with an analog-to-­ the DATA pin unless the RSSI indidigital converter will be well-suited cates that a strong enough signal is to monitoring the RSSI. present. We closed JP1 by bridging it with solder (you can also use a small 0W SMD resistor). You will need a generous amount to bridge the gap between the pads. 68 Silicon Chip Using a socket strip as a guide will ensure that the pins are soldered with the correct separation even though they are in two groups. Australia's electronics magazine siliconchip.com.au