This is the kind of amplifier design that gives you the most design flexibility. You could compare this level of design with programming a computer in assembler. You have to deal with many details that a higher level language shields you from, but on the other hand you can do anything imaginable and with much higher execution speed.

There are easier alternatives. You can use a pre-designed amplifer like an RF gain block IC. Or you can use a CMOS inverter with added negative feedback.

There are many RF gain block ICs available. These are ICs in which they have done all the work for you and you only need to make A simple pcb with a few components and it will work. Here are some examples:

http://www.minicircuits.com/products...onolithic.html
http://media.digikey.com/pdf/Data%20...20MSA-3186.pdf
http://www.sirenza.com/documents/pro..._Datasheet.pdf

These kinds of amplifiers have much much more bandwidth than you need which can be a danger. Because the bandwidth is so large, you have to design your circuit with a layout and parts suitable for 1 GHz operation so that the amplifier doesn't oscillate.

I have seen people make amplifiers successfully at 1 MHz using simple CMOS (HCMOS) logic gates. If you put a feedback resistor from output to input, these gates will operate in a linear way and make an easy amplifier. You still would have the problem that the gain might be too high and the amp will oscillate. The lower the feedback resistor value the better I think.

__________________
RadioRon

 

So, should I implement the BJT transistor amplifier circuitry on the PCB for result and troubleshooting? What if we implement the op-amp in the amplifier design? Do you have any suggestion on the op-amp that we may use in the design? By the way, I am running out of time... (5 weeks to go before presentation)

 

An op amp would work too, but might be somewhat noisier. The key to the op amp approach is to choose one with good bandwidth. You need at least 8 MHz gain bandwidth product to get reasonable gain at 1 MHz. This one is typical:

http://www.national.com/ds/LM/LMV851.pdf

but there are many to choose from. Any more bandwidth than 50 MHz and you are risking problems with stability.

In most projects there are many ways to solve a problem. You are faced with choosing from several alternatives, none of which you have built before. In this case, you should assume that the first prototype that you build won't work well and will need some adjustment. For this reason it is often best to build something very early in your schedule to get a feel for how far away from completion you are. It is not wise to use up most of your time available studying the design on paper and in simulation and assume it will work when you finally build it. It won't. You will make minor mistakes, and with little time left risk panic and further errors.

In your case, you should build rough prototypes of each stage using hand cut pcb and then test them individually. Using a piece of un-etched copper pcb, you can sketch out a simple block layout in pencil right on the copper, and using a sharp knife or a rotary tool with a grinding head (ie. dremel) you can cut simple straight strips out of the copper to isolate islands of copper which are the connection nodes between components. Do this for the BJT amplifier and you can have it built in less than one hour. If you use double sided copper board, the uncut side can become a good ground plane which is useful in this kind of design. (always connect the ground plane to the top side ground) Then you can test it using a generator and an oscilloscope. Once it is adjusted and working, you can consider this little pcb a building block to use in your system. It might look ugly compared to a commercially designed product, but that is not important.

If you are able to work quickly like this, it won't be hard to make a choice of which type of amplifier to use because you can choose any one type, build it and decide quickly if it will be OK or if you need to try another. Working quickly in building prototypes takes some pressure off of how you make your choices.

Of course, to work quickly may mean to make the wrong choice and waste a few dollars on the wrong parts. This is a normal risk in our world and we can take some solace in the fact that the electronic parts for prototypes are the least expensive part of the game.

__________________
RadioRon

 

Thank you for your advices, I know what should I do now.

For the rx ferrite loop, I am thinking of making it to be a single block, where terminals from primary loop are connected to the tuning capacitor, and the terminals from secondary loop are connected to the BJC connector. Is there any thing that should i pay attention to?

My rx rod antenna is about 1 meter high, is it too big for the buffer amp?

 

Nothing comes to mind about the ferrite loop. Go ahead and build it. I think you have the right idea.

The rod antenna is not too big for the buffer.

__________________
RadioRon

 

I have designed the phase detector circuitry, but I wonder if those are the suitable value for the components used in the circuitry. Any comment on this circuitry?

Attached Files

PhaseDetector.pdf (38.4 KB, 4 views)

 

I would reduce the bypass and coupling cap values to 0.1 uF instead of 1 uF. Otherwise, seems OK. You would be wise to copy the application pcb layout in the data sheet, page 21.

__________________
RadioRon

 

Well, I am thinking of providing the power to the Phase Detector using a power supply, thus I have made changes on the diagram... Do I need to add extra capacitors to the power supply part as what have been done in previous design (as shown in the picture below)?

Attached Images

PhaseDetector.JPG (100.1 KB, 5 views)

Attached Files

PhaseDetector.pdf (39.1 KB, 3 views)

 

The only critical point where you must have a ceramic bypass capacitor is from Vcc to ground as close as possible to the chip.

__________________
RadioRon

 

Well, for the transmitter part, the output swing of the crystal oscillator is around 4Vpp, so, is there any suitable amplifier for this situation? This is because I want to increase the transmit power of the transmitter... Or should I use voltage divider on the output of the crystal oscillator and implement the small signal amplifier (just like the type that i build for the receiver) to increase the transmitter power?

 

I would use no voltage divider or perhaps a modest voltage division ratio, but you don't need to attenuate the signal. A better approach is to use the same transistor in your amplifier but alter the bias point by adjusting the resistor values. When you have a lot of voltage swing available like this and you aren't carrying modulation on the carrier, a simple class C bias is appropriate. You can achieve this by removing the base bias resistors and simply letting the input signal turn the transistor on and off. You could also reduce the emitter resistor a little bit too so that more output power can be had.

__________________
RadioRon

 

Is there any example of this type of amplifier?

 

When I looked on the web, most of the examples of class C operation are at much higher power levels and a different type of output circuit is used. So I cannot find a good example of the type you are doing. It is not much different from the amp we have discussed before.

If you google on Class C RF Amplifier you will see examples of high power amps and you may get some inspiration from those.

__________________
RadioRon

 

Well, I have done etching the pcbs for the system, however, the ground plane is over-etched where the ground plane seems to be a little bit faded out. I have tried several times etching the pcb, but this still happens to the ground plane. Is it alright to have over-etched ground plane? Will this reduce the performance of the block or cause the block to be malfunctioned?

 

I have solved the etching problem, the ground plane in all the circuits are perfect in shape as well as the connectivity... Well, regarding the Class C Amp, I have designed an Class C Amp bu referring to a example model on internet. But the circuit is not working... Did I do any mistakes in designing this amp circuit?

Attached Images

Drawing1.jpg (357.1 KB, 4 views)

Attached Files

ClassCAmp.pdf (40.3 KB, 3 views)