Wireless Waffle - A whole spectrum of radio related rubbish

How not to design transmitters and receivers: part 15 (receiver gain distribution)signal strength
Monday 22 November, 2021, 20:32 - Amateur Radio, Broadcasting, Licensed, Pirate/Clandestine, Electronics
Posted by Administrator
Before addressing each of the building blocks required for a receiver, it is first important to consider the necessary gain distribution of the circuit and the various elements that make it up, that is to say the amount by which the signal needs to be amplified before it reaches the de-modulator. Set the gain too low and the receiver will lack sensitivity. Set the gain too high and not only is there a chance that it may self-oscillate due to the excessive gain, but the de-modulator may be driven with unnecessarily high signals which impede its performance.

The age old CA3089 (and the later CA3189) is still a good building block for a wide-band FM receiver as it has acceptably good noise performance, is easy to use, and has all the features needed (such as a signal strength indicator output) for most applications. The datasheet for the CA3089 shows that it begins to develop an audio output when the input to it reaches 12 microVolts. At this point it is 3dB down (half power) and rises to maximum output very quickly.

Now for some maths:
  • Let us assume that the bandwidth of the receiver is 250 kHz (this is set by the IF filters) which is about correct for a wideband FM receiver.
  • The amount of thermal (background) noise generated in 250 kHz of bandwidth into a 50 Ohm load is 0.45 microVolts - this is the amount of noise that would be naturally present at the input to the receiver based on the kTB formula where 'k' is Boltzmann's constant, T is the temperature of the receiver in degrees Kelvin (usually regarded as 290K) and B is the bandwidth in Hz.
  • Note that man-made noise and other factors will increase this level at the antenna and thus the input level of the receiver will be somewhat higher. In addition, unless the receiver is noiseless (which is impossible) it will, itself, generate additional noise above and beyond this level, the amount of this is known as the noise figure or noise factor. However, let's use the thermal noise as the theoretical minimum input.
  • The necessary voltage gain of the receiver from its RF input to the input of the CA3089 therefore needs to be a factor of 26.6 (12/0.45), or a power gain of 28 dB.
Next we need to consider the gain (or loss) of the various elements of the receiver:
  • The input band-pass filter may have a gain (actually a loss) of -2 dB (lower loss filters can be generated, but we shall be cautious in this regard).
  • Mixers have not yet been discussed, however the NE602 which will be introduced later and used to perform the job of the mixer has a gain of +17 dB.
  • This means that the IF stage which usually comprises a filter and an amplifier needs to have a gain of +13 dB to make the overall gain correct.
  • A typical ceramic 10.7 MHz IF filter will have a loss of around -6 dB, meaning that to correct the overall gain, the IF amplifier needs approximately +19 dB of gain.
This arrangement is illustrated in the figure below.

receiver gain distribution

This arrangment of gain means that the de-modulator will see the background (thermal) noise as strong enough to begin to de-modulate audio. Any actual received signal will therefore be above this level and will be strong enough for the device to work correctly.
add comment ( 253 views )   |  permalink   |   ( 2.5 / 957 )


<<First <Back | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | Next> Last>>