Being a ham operator I'm always on the lookout for new radio related projects. This time I investigated the properties of direct conversion receivers using the cheap and widely available NE612 balanced mixer IC. There are four different versions of this chip NE602, NE612, SA602 and SA612. From what I understand the main difference between NE and SA is that SA has extended temperature range (primarily made for military use I guess). For normal indoor hobby projects NE are most likely sufficient. Then there is the 602 vs 612 question. Both versions are very similar and I've heard the 602 was the first version and then came the slightly improved 612 version. I heard a rumor that the masters for the 602 was actually destroyed in a fire and only 612 chips are made nowadays but because people keep requesting 602 chips they are still being sold but are actually relabeled 612 chips. I have used both NE602 and NE612 interchangeably and not noticed any obvious difference. The main part of a direct conversion receiver is a single mixer stage. A local oscillator is mixed with the frequency one wants to receive and the product of this mix is the sum of the two sidebands around this frequency. Hence a direct conversion receiver can be used to receive CW, LSB, USB, DSB and AM signals with very simple means. At first I thought of using a passive balanced diode mixer stage and no preamp on the input but quickly discarded this method since there is a high risk that the LO will travel back out the antenna and cause interference to other stations. This is when I seriously began looking at the NE612 instead. The NE612 is very sensitive and does not need any preamp and there is minimal risk that LO will exit the antenna. The NE612 also contains all needed active circuitry for implementing a simple oscillator. I did not build an oscillator for this project but instead relied on my signal generator for generating the LO needed. A Fluke 6060A, which is a big piece of equipment but produces a clean and stable signal so that I could focus on the receiver part without having to worry about the quality of the VFO. That's another project. The main focus on this design was trying to handle the limited dynamic range of the NE612 which easily gets overloaded. Overloading causes nasty distortion and inter-modulation. To minimize that risk and also increasing the dynamic range of the receiver I designed a simple automatic variable attenuator. Below is the final schematic of the receiver. This is a junk-box build so component values are not optimal. Looking at the signal path from left to right there is a variable attenuator at first using a simple potentiometer. After that there are two protection diodes. One of which is also used for the variable attenuator. I did not have any PIN diodes at home but regular 1N4148 worked well. Next is a matching transformer. The input impedance of the NE612 is 1500kOhm according to the datasheet. I had no spare toroids at home so the core of the transformer is actually an RFI choke from an old VGA cable. Remember: N1/N2=sqrt(Z1/Z2) The output from the mixer in a direct conversion receiver is low level AF and is prone to pick up noise. One really good thing about the NE612 is that it is balanced. I kept the balanced signal for the first op-amp AF amplifier. That improves the signal to noise ratio quite a bit. The first AF amplifier also serves as a low pass filter. The AF output is then routed back to the second half of the op-amp that is working as a comparator. The purpose of this comparator is to detect high level signals and adjust the variable attenuator at the RF input. This makes up a crude "AGC" like function. The aggressiveness of the automatic attenuator is set using the comparator reference potentiometer. External LO from my signal generator is fed to the NE612 through a decoupling capacitor. Signal level is fixed at +1dBm. Going any higher does not improve signal receive quality and may damage the NE612. Also note that no BPF is used on the RF input. I initially planned to make this a single band 40M receiver but after some experimenting I realized that this design would work well for a general coverage receiver. If strong broadcast transmitters are close I suspect there would be trouble but that is not an issue at my location. The output AF signal also drives the simple power amplifier seen below. Some additional low pass filtering is added here as well. I know the amplifier could be a lot better but I found the amplified signal from this circuit to be more pleasing to the ear than the signal from an LM386. Below is the final receiver. Power amplifier is not included on this board. Note that the potentiometer on the RF input is not mounted. I did not need it after implementing the variable attenuator. At least not using my relatively simple antenna (tuned 15m wire). Also present on the board is a 7805 regulator that is not included in the schematic. Having a clean power supply is really important. After using the receiver for a few weeks I must say that I'm quite pleased and surprised. This thing actually works. And it works quite well. Most signals I can hear just as good on my Icom IC-735 as on this simple receiver. Of course the fact that both sidebands are always received reduces the selectivity of the direct conversion receiver but in case of both sidebands being present in the received signal it may also work in its favor. To sum it up. A fun and useful project. Now I just have to add a simple DDS VFO for a general coverage HF receiver.