Microwave Amplifier design

[hammooooody]

Hello everybody,

i am suppose to design a simple microwave Amplifier (1 - 3Ghz) with input and output microstrip matching networks by simulation and also building the hardware ...

I am still a noob with RF and i don't know where to start from.

In MATLAB simulink for example, there's an S-parameters Amplifier Block which requires S-parameters values which i need to bring from an actual transistor's data.
My Questions:
1- What type of transistors should i look for?(Component code would be great)
2- do i have to obtain the S-parameters values for that transistor myself? or they are provided in its datasheet or something?
3- which simulation tool do you recommend?

 

[martine1212]

1.-Read Gonzalez book "Microwave tansistor amplifiers"
2.-Read again and make sure you understand it.
3.-If you want a 1-3 (or 4) GHz amplifier probably you need a transistor with and ft of about 9 to 16GHz. More is not always better. Bipolar would be easier and cheaper to find.....
4.-Most manufacturers can provide you with S data for their transistors for an specific set of bias conditions.
5.-Agilent's ADS or microwave office would be good for designing the stuff.

 

[srikanthvajjala]

You need S2p parameters of the devices. Select rftool in matlab and load the input matching circuits,device S2p file and output matching circuits into the rftool. Then see the response at your required frequency. It is very easy to simulate but you need strong fundamentals abt the amplifiers.

Hello everybody,

i am suppose to design a simple microwave Amplifier (1 - 3Ghz) with input and output microstrip matching networks by simulation and also building the hardware ...

I am still a noob with RF and i don't know where to start from.

In MATLAB simulink for example, there's an S-parameters Amplifier Block which requires S-parameters values which i need to bring from an actual transistor's data.
My Questions:
1- What type of transistors should i look for?(Component code would be great)
2- do i have to obtain the S-parameters values for that transistor myself? or they are provided in its datasheet or something?
3- which simulation tool do you recommend?

 

[ltrlspree]

if you need to cover 1-3 GHz range and you do not have a lot of experience, you may think of an RFIC as well. It is more integrated and you have less effort for designing in the device.

You can check for example the ABA-3xxxx or ABA-5xxxx families from Avago Technologies (former Agilent) as long as you have an easy NF requirement.

Cheers

 

[hammooooody]

Thank you for the great responses ...

 

[eee100]

Hi,

I would suggest Agilent ADS for simulation. Its a very good software. For 1-3 GHz range you can use MMIC amplifiers. For example I used MSA-0886 for my RF amp. At your frequencies you will get 15dB gain on average with a good impedance matching. These MMIC modules exist in Agilent ADS's libraries so its easy to simulate. Keep in mind that the PCB design is also very important at those freqs. You will need to design your PCB carefully or the circuit may be completely useless in real life even it works perfectly in the simulation tool.

Good luck

 

[jimiblues]

1.-Read Gonzalez book "Microwave tansistor amplifiers"
2.-Read again and make sure you understand it.
3.-If you want a 1-3 (or 4) GHz amplifier probably you need a transistor with and ft of about 9 to 16GHz. More is not always better. Bipolar would be easier and cheaper to find.....
4.-Most manufacturers can provide you with S data for their transistors for an specific set of bias conditions.
5.-Agilent's ADS or microwave office would be good for designing the stuff.

Hi martine1212,.

I agree with your lists, but have difficul in understanding 'for ft, more is not always better'. It will be very nice if you can explain more on this.

Cheers,
jimiblues

 

[hammooooody]

I know it's quite late to say that, but i wanna thank everybody. And special thanks to martine1212
Gonzalez book is brilliant. it has answered many of my questions. Agilent ADS was easy to follow and user friendly, i managed to do most of the work without referring to tutorials or anything like that. Thanks again.

 

[martine1212]

Sorry for the late reply, if you pick up a transistor with a very high ft (say 50GHz for a 3GHz project) then it is very easy to generate an oscillator due to simple and unavoidable parasitics (add the board, the solder, the pads, the small transmission line interconnect). The transistor has enough gain at much higher frequencies so you do not need much to make it oscillate or hit resonances.

Usually I pick transistor with and ft with a factor of 5-7 from what I need. Unless I need every bit of speed then I try to keep the gain low above the range of frequencies I am interested.

Hi martine1212,.

I agree with your lists, but have difficul in understanding 'for ft, more is not always better'. It will be very nice if you can explain more on this.

Cheers,
jimiblues