voltage gain power gain

Hi all:

I am confused about the power gain and voltage gain used in the RF design, especially when designing LNA and mixer. Which one should draw more concern?

Can anyone help me out? Thanks a lot!

At RF and above it is the signal power that counts so power gain is the one to use.

still not get it. Why does the signal power count at RF and Above?

Thanks a lot!

Because it is the power signal to noise ratio that determines performance. Also power is easier to measure than voltage at these frequencies.

Hello,
i think the power gain is used in NF calculation while the voltage gain is used in IP3 calculation,
i mean total NF and IP3.
note the Ap=Av**2 when the source and load resistance are equal.

I know they are the same in a matching conditon. But why do we use power gain more often than voltage gain? Is it bacuse efficiency? since "maximum power transfer=highest efficiency"

Eactly!

On the other hand, power gain(in dB)=2*voltage gain(in dB).

At RF it is hard to create a good high or low impedance. Therefore people do not think in terms of voltages and currents for RF. The 50 ohm comes from the characteristic impedance of optimized coax.

i think is power gain

Hi Ic_fan,
At Rf frequencies power gain is used. The main reason is at this frequency range reflections (both current and voltage) become to occur in the transmission lines. So, characterization of the circuit is done by s-parameters. And s-parameters are based on incoming and outgoing powers.

best regards,

S parameters are based on forward and backward traveling wave amplitudes, which are much more like a voltage than a power!

One thing is that when you down convert (a Mixer) the output
will be at some IF Freq and the Impedance will most likely not
be 50 Ohms so Voltage gain might be better.

dB = 10*LOG(Pout/ Pin)
dB = 10*LOG([{Va^2}/Ra] / [{Vb^2}/Rb])
(@ RF & above Ra~Rb~50 Ohms)

biff44:

S parameters based on the ratios of reflected/incident power waves and transmitted/incident power waves; an's and bn's


So at this case they are based on power!
But the results gives the forward and reverse voltage gains and reflection coefficients!

In analog cirucit, people like to use voltage gain but in RF circuit , power gain is more used because people concern more about power deliver efficiency

I agree with the explanation given by Mr.Edata.....in RF measurement due to transmission line effect Power Gain is preferred

voltage signal are quite important during circuit design (due to device constraints).

But power is our concern during transmission.

I believe people get confused with this PowGain of 3dB = 10log(Pout/Pin), where a power ratio is 2. By converting to voltage gain VoltageGain in dB = 20log(Vout/Vin) must also be 3dB, not 6dB. because to have power ratio of 2 or 3dB, a voltage ratio is not 2.
Just do simple math converting log units to linear on the same resistance.

So, I believe Power Gain in dB == Voltage Gain in dB, no doubling

Doubling will apply only if the ratio of linear units is equal, i. e. power ratio of 2 is 3dB, while Voltage Ratio of 2 is 6dB

Or am I missing something basic.:D

If you say xx dB gain, it's unequivocal without addendum. But the calculation is different for power or voltage ratios.
20 dB means factor 10 in voltage but factor 100 in power.

yes, the calculation is different. Maybe I wasn't 100% clear in my statement.
As you FvM mentioned a gain of 20dB in linear term is ratio of 100 for Pout/Pin, or for voltage it is 10.

That means; if you increase your voltage 10 times, the power will increase 100 times, assuming on the same load, so the gain is always 20dB, no metter what you measure voltage or power

I hope this will at least lower cofusion if somone says my RF amplifier has a gain of 10dB, is it power or voltage?
it is the same for both, you just calculate it differently, this why a dB is introduced, to eliminate big numbers and have one figure for all of them, voltage current or power.

This is the main reason s-parameters were developed. You would have an almost impossbile task trying to characterize a RF system by treating it like a typical 4 port system, opening and shorting the nodes and trying to measure voltage and current. The ports would have to be shorted at a distance much less than a wavelength from the device. S-parameters require only that the ports are terminated to some impedance, which is easy to do.