The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> RF Design >> Help some basic concepts about RF amplifier
https://designers-guide.org/forum/YaBB.pl?num=1374056821

Message started by baab on Jul 17^th, 2013, 3:26am

Title: Help some basic concepts about RF amplifier
Post by baab on Jul 17^th, 2013, 3:26am

Hi, I want to ask some basic question about RF amplifier GNSS architecture.
Here is the overall block diagram of the architecture.

And here is the RF specification of its in the IEEE paper attached:

Now I want to ask two basic questions. Please note that I am only a beginer and not angry!
1. In the paper, the voltage gain of RF amp is 13.3 dB. But in my case, I need to increase the voltage gain to at least 50dB. Do you think this is doable?
Is it so hard to get the voltage gain 50dB?
2. When we said that the RF amp has input power is -110dB, then how can we know about input voltage? I think we need to know about input impedance of RF amp and then use the formula:
V = sqrt (P*Zin)

Title: Re: Help some basic concepts about RF amplifier
Post by aaron_do on Jul 17^th, 2013, 9:38pm

Hi

1. Which amplifier are you designing? Anyway 50 dB gain may be doable (bandwidth will be tight), but it is not advisable. In the first place, the linearity will be very poor as any blocker is liekly to saturate the amplifier, and in the second place, its likely to be unstable. Any time S21 is more than S12 at RF frequencies, you're asking for trouble.

2. Yes to calculate the voltage, you need to know the impedance.

Why do you need to increase the amplifier's gain to 50 dB?

Aaron

Title: Re: Help some basic concepts about RF amplifier
Post by tm123 on Jul 18^th, 2013, 10:20am

Hello baab,

Trying to achieve 50dB in a single RF amplifier stage will be a fruitless effort. You will certainly need to have several cascaded stages to achieve that amount of gain at RF. Depending on whether you are using bipolar or MOS devices, a reasonable amount of gain would be 15dB per stage in my opinion.

There are always tradeoffs involved in design, so you will have to tradeoff some other performance to achieve high gain as Aaron described. Most likely you will be trading off linearity by decreasing the amplifier degeneration. Stability is always a concern at RF as well.

As a side note, there are many ways to define 'gain' of an amplifier including voltage gain, power gain, transducer gain, available gain, etc. Please make sure you understand these definitions and understand what the spec is. Always be clear on how you simulate/measure the 'gain' of any RF/microwave amplifier. I realize you specified voltage gain in this case, but since you are impedance matching at the RF amplifier input and the RF amplifier output is terminated on chip it can become confusing as to what gain you are actually quantifying.

Title: Re: Help some basic concepts about RF amplifier
Post by baab on Jul 23^rd, 2013, 1:29am

Sorry for the late reply!
Aaron,

Quote:

1. Which amplifier are you designing? Anyway 50 dB gain may be doable (bandwidth will be tight), but it is not advisable. In the first place, the linearity will be very poor as any blocker is liekly to saturate the amplifier, and in the second place, its likely to be unstable. Any time S21 is more than S12 at RF frequencies, you're asking for trouble.

I have been said that I need to design an RF amp with voltage gain 50dB. However, I have searched a lot and haven't found any voltage RF amp. All RF amplifier are RF power amplifier. I wonder can I use these RF power amplifier to get voltage gain 50dB.

Quote:

Why do you need to increase the amplifier's gain to 50 dB?

I need about 0.3V at the input of ADC. Maybe, I miss something!

tm123,

Quote:

Trying to achieve 50dB in a single RF amplifier stage will be a fruitless effort. You will certainly need to have several cascaded stages to achieve that amount of gain at RF. Depending on whether you are using bipolar or MOS devices, a reasonable amount of gain would be 15dB per stage in my opinion.

There are always tradeoffs involved in design, so you will have to tradeoff some other performance to achieve high gain as Aaron described. Most likely you will be trading off linearity by decreasing the amplifier degeneration. Stability is always a concern at RF as well.

As a side note, there are many ways to define 'gain' of an amplifier including voltage gain, power gain, transducer gain, available gain, etc. Please make sure you understand these definitions and understand what the spec is. Always be clear on how you simulate/measure the 'gain' of any RF/microwave amplifier. I realize you specified voltage gain in this case, but since you are impedance matching at the RF amplifier input and the RF amplifier output is terminated on chip it can become confusing as to what gain you are actually quantifying.

I am learning basic concepts before designing.
How about if I use multi-stage amplifier? For example, I will use three stages, each one has a voltage gain 17dB.

Title: Re: Help some basic concepts about RF amplifier
Post by aaron_do on Jul 23^rd, 2013, 4:10am

Hi,

If your requirement is 0.3V at the ADC it doesn't mean you need to put all your gain at RF. Usually about 20 dB is enough, and you can put the rest at IF/baseband. You also probably want to do some filtering before any heavy amplification. Otherwise you need very linear amplifiers.

Since you are trying to understand basic concepts, I suggest you try and understand about IM3 requirements, blockers, gain desensitization and filtering. Then you will know what we are talking about.

regards,
Aaron

Title: Re: Help some basic concepts about RF amplifier
Post by baab on Jul 24^th, 2013, 10:21pm

I am reading these specifications. I want to ask one more question. In the picture is the block "RF amplifier" the RF power amplifier? I ask it because most of RF amplifier I have found is RF power amplifier.
Thanks.

Title: Re: Help some basic concepts about RF amplifier
Post by baab on Jul 24^th, 2013, 10:24pm

Title: Re: Help some basic concepts about RF amplifier
Post by aaron_do on Jul 24^th, 2013, 11:44pm

It is a low-noise amplifier (LNA). There is plenty of literature on LNA design. BTW, I recommend you read T. H. Lee's book from start to finish. I also recommend Razavi's RF microelectronics.

regards,
Aaron

Title: Re: Help some basic concepts about RF amplifier
Post by baab on Jul 25^th, 2013, 6:11am

Hi,
Thanks for your recommendation. I bought Razavi's RF microelectronics.

Quote:

It is a low-noise amplifier (LNA).

But there is already a LNA before RF amplifier.
If it is LNA then it really has many materials.

Title: Re: Help some basic concepts about RF amplifier
Post by RFICDUDE on Jul 27^th, 2013, 5:16am

The "rf amplifier" block is a linear amplifier before (or part of) the mixer. It is not, generally, a power amplifier.

The "rf amplifier" driving the mixer is usually defined by the type of input signal the mixer expects to receiver (current or voltage) and the performance specifications (gain, linearity and noise).

Title: Re: Help some basic concepts about RF amplifier
Post by aaron_do on Jul 27^th, 2013, 8:02am

Normally as the second amplifying block in the chain the linearity is more important and the NF less so...but from the specs given, that just seems like a typical LNA to me...

I think I have seen cases where an additional off chip LNA is used to boost the sensitivity.

Title: Re: Help some basic concepts about RF amplifier
Post by baab on Jul 27^th, 2013, 8:53am

Thanks for the helps.
I want to ask some questions. In the case of the transmission line has 50ohm characteristic impedance, then power amplifier and RF amplifier are the same because P = V^2/z0?

Title: Re: Help some basic concepts about RF amplifier
Post by aaron_do on Jul 28^th, 2013, 2:08am

The rf amplifier in you diagram is not the same as a power amplifier. You should do more reading or it will be difficult to explain things without going into a lot of detail.