All waves RF generator
built and modified by sv3ora

An RF signal generator is a special piece of test equipment, which allows you to test a wide variety of RF circuits and perform many other interesting RF experiments. For quite a few years, I was trying to perform the RF experiments that required an RF source, using purpose-built narrow band LC or even crystal oscillators. After the tests were completed, I dissasembled some of these since I needed the components and I had to rebuild them in the future to test new circuits. When the oscillators were not dissasembled, they ended up in the circuits junk box. More or less, I ended up with a whole bunch of narrow band RF oscillators, most of them not usable for my next projects that required other frequencies of operation. If you are really into RF, soon or later you are going to need a wideband RF signal generator.

Nowadays, with the improvements in DDS technology, a digitally-synthesized frequency-stable wideband RF signal generator with good characteristics can be made at very low cost. It almost makes no sense to try to build an analogue type RF signal generator, commonly composed of many resonators, so as to cover a wide frequency range. But for the homebrewer, the situation may be quite different. There are various reasons for that. The DDS requires a microcontroller in order to set it's frequency. The microcontroller requires a programmer hardware and software, in order to be programmed. A PC is also required for the programming operations. Quite a few of the homebrewers do not know how to write a program to control the DDS and learning MCU programming is difficult for many. Thus, they rely on programs others have built and they cannot alter their operation to their needs. The DDS also requires a stable very high frequency clock oscillator for it's operation although the situation becomes a bit better (lower frequency needed) if there are internal frequency multipliers within the DDS chip. Apart from these things, soldering a DDS chip is a nightmare for the homebrewer with little or no experience in SMD and it requires SMD equipment. The tiny pinout of the DDS chips can be proven to be challenging to solder, even for more experienced people and prototyping is almost out of question. Finally, a DDS chip is a "black box" module and there is no real satisfaction to the RF experimenter, since he is not building any RF circuit, but just using a chip to produce the RF without any chance to change it's RF characteristics. Take the time to read the list of requirements in this paragraph and you will immediatelly see why a DDS-based generator is not always the best solution, practically speaking. Even the cost of the DDS that might initially be thought as low, can be proven to be much higher at the end, with all those mentioned requirements.

An analogue RF signal generator instead, can be built with a few discrete components soldered together with a common soldering iron and no previous experience in SMD components or MCU-related things. Depended on the generator circuit and the way of building it, the total cost may be lower than that of a DDS generator and the resonators can be home-built. In a DDS, the output waveform result is always the same (guaranteed by the DDS chip specifications), if properly built. However in an analogue type RF generator, the quality of the output signal depends on many factors, such as the circuit topology, the RF shielding, the mechanical stability, the Q of the resonators etc. But you have a full control over these things and it is really up to you to decide the best way to match your requirements and that is the beauty of it.

There are poorly designed analogue RF generator circuits or poorly built, as well as good ones or properly built. I have seen and tested quite a few signal generators that appear on the net and on magazines but I wasn't really satisfied with the end result of any of these. It is ok to build a poor analogue RF signal generator if you only intend to do simple things with it, but if you are going to use it for more serious work, choose a circuit and method of construction that meet your tighter requirements. Since you are going to do the effort and spend your time in building an RF signal generator anyway, I would advise you to start with a better quality circuit and build it as properly as you can, based on your available time, your patience and the end result you try to achieve.

For the average home lab, the main characteristics that distinguish a good analogue RF signal generator from a poor one, as far as concern the quality of the generated signal, are:

• Wide frequency range.
  You certainly want a wide frequency range so you can test more circuits in more frequencies. The widest the better, that is the main purpose you are making the signal generator. However, keep in mind that as the frequency gets significantly higher, the frequency stability suffers more and the need for better shielding gets important.
• Variable output level.
  The output level of the generated signal must be able to be varied over a broad range, so that you can test equipment at both very low and high levels. This is usually accomplished with a resistive attenuator at the output of the signal generator. This attenuator is better to be a switched attenuator and not just a potentiometer (voltage divider), so that the exact attenuating values are known and the input and output impedances are kept constant regardless of the attenuator setting.
  
• Constant output level.
  If you use an output attenuator, you could set the level of the generator at any value you wish. However, a constant output level from the generator (prior to the attenuator) is desirable. This is because, regardless of the attenuator setting, you would expect in many of your tests to see a constant output level as you tune accross the frequency of the generator, regardless of it's level. Imagine for example that you want to sweep the generator to test the responce of a wideband filter. The last thing you want, is for the generator level to change as you sweep across the filter bandpass.
  
• Constant 50 ohms output impedance.
  The output impedance of the RF generator must remain constant near 50 ohms at all frequencies, which is the standard impendance value of most of the RF circuits we use today. It is sometimes ok to drive a higher impedance load with a 50 ohms source, but the opposite may cause problems from the excessive loading of the generator. The constant impedance must be maintained at the different settings of the output attenuator (if you use one). A switched attenuator designed for 50 ohms, fulfills this requirement.
  
• Low distortion (harmonics content).
  Every oscillator, even sinewave, produces harmonics (distortion). In the case of single band oscillators, these harmonics can be very easilly attenuated with a low pass filter at it's output. However, in wideband oscillators, harmonics attenuation cannot be easily and cheaply done using multiple low pass filters for the different ranges, so harmonics handling has to be taken care from the core design of the oscillator and the buffer amplifiers. It is important to have low distortion, the lowest the better, especially if you intend to use the RF generator as a local oscillator to mixers in direct conversion receivers.
  
• Adequate frequency stability.
  This is very important and most unlocked (VFO) analogue RF generators are not capable of adequate frequency stability. You want your output signal frequency to stay put for as long time as possible if you want to do jobs like testing narrow band filters and many others. It is difficult to achieve that in an unlocked VFO, without an oven to keep the temperature stable. However, you can minimize frequency drift by using better quality inductors and capacitors as frequency determining components and NP0 capacitors in the RF places of the circuit. Frequency drift compensation using different PPM capacitors is possible but this is difficult if different resonators are switched in the circuit.
  
• Fine tuning step
  I have seen quite a few wideband RF generators that use just a single variable capacitor without any reduction gear to set their frequency. This is ok for general receiver alignment, but for narrow band work (eg. SSB and CW filters alignment), you need to be able to precisely set the frequency. In other words, you need to have a fine "tunning step" if you cant to call it like this, since there is no such thing in analogue terms. If a variable capacitor is used to vary the frequency, fine tuning step can be accomplished with the use of mechanical reduction drives. If very fine tuning step is required, you can even connect two such drives in series.
  
• Good shielding
  Good shielding is mandatory to a good quality RF signal generator. Shielding ensures that frequency pulling from hand effects is eliminated when you come close to or touch any of the chassis and knobs in the front panel of the generator. It also provides isolation within the generator. If you use a switched attenuator make sure you shield it, so no external signal can pass through it. If you want to go extreme, each attenuator switch of the attenuator can be independently shielded from the others. The high voltage and the transformer that supplies power to the generator must be shielded from the rest of the circuit as well. You can also use RFI filters in the mains, to prevent the mains cable from radiating RF leaked from the generator. The whole RF generator circuit must be enclosed inside a metal enclosure and connected to ground, but do not place air-core inductors too close to the metals if you can.
  

The trick for quickly engraving channels onto the pcb, is to make thin side lines on each trace or pad first, making sure that the copper is cut all the way through, down to the fiberglass. Then peel a small part of the copper out of the pcb at one side of the trace. Finally, use tweezers to pull this peeled part and the copper will be pulled all the way along the channel, isolating the trace from the copper ground around it.

Make sure you cover the red LED with a small piece of black thermal shrink tube, so as to create a lightproof enclosure for the LED. Heat the thermal shrink tube and when it is still hot, hold it's both ends with pliers and let it cool as you hold it. When it is cooled the shrink tube will stay in place with both ends closed, preventing any light from affecting the LED.

Here are more pictures from my implementation of the RF generator

Back to main site