October 31, 2012
Last month we looked at aircraft radio frequency and audio frequency interference and measures that we could take to minimize their effects. Let us now turn our attention to the testing and alignment of receivers on the bench. Most of the radios we use are of the superheterodyne type. They utilize an oscillator that is heterodyned or beat against the incoming signal in the mixer and the difference is amplified in the intermediate amplifier (IF). The IF is typically where the gain of a receiver is found. This is also where much of the selectivity is shaped. Using the example of the Allied Signal KX-155 Nav/Com in their test data the selectivity is measured at the 6dB and 60 dB points. You measure the AGC voltage with a 3mV input at a frequency of 126.5 Mhz and call this a reference. Then you increase the RF signal level by 6 dB ( to 6mV) and vary the generator RF frequency above 126.5 Mhz until you have the same AGC voltage as you had with the 3mV input. The AGC voltage is changing because the signal is beginning to fall outside of the bandpass of the IF. The RF input must be increased to compensate for this loss of signal. What you want to know is how far you can vary the frequency of the RF input from nominal (126.500Mhz) until you have the same AGC value as the reference. The same test is performed by lowering the RF input frequency from nominal and finding where the AGC voltage is the same after you have increased the signal generator’s attenuator by 6dB. These are the 6dB points or the IF’s 6dB bandpass. From our previous discussion a 6dB loss means that only about 25% of the signal is getting through the IF at these points. The KX155 Test Data Sheet specifies 16Khz or more of bandwidth at these points for their 25Khz units (126.492 to 126.508 Mhz). In a perfect world we would have no signal through the IF outside the favored bandpass. In the real world the drop-off or skirt is not exactly vertical but it is impressive. If you test for the 60 dB points in the same manner as described above the test data sheet specifies it to be no more than 40 Khz wide (126.480 to 126.520 Mhz) for their 25 Khz units. This means the IF must attenuate a signal an additional 54dB over only a 12 Khz span. Or mathematically the signal is reduced to 0.000004 of the value found at the 6dB point. Why is it important to know this? Because if the IF bandpass should shift or become narrower the desired signals may be attenuated a great deal. It would not be hard for a communication receiver’s IF to pass a simple sensitivity or squelch test using a 1 Khz tone. But our voices contain much higher frequencies. How many times have you tested a customer’s radio for squelch problems that you could not confirm? Check the IF bandpass.
Another example is the Collins VIR32 Nav receiver. Its maintenance manual has you sweep the RF front end, then check for AFC action over a +/- 16 Khz range. In general the VOR receiver’s IF is a much more critical. It must pass the 9960 Hz reference signal without distortion or VOR errors could result. It must also be able to pass both the regular VOR and doppler VOR signals as one although they are produced in entirely different ways and the resulting RF spectrum is different. The doppler VOR transmits voice and the 30 Hz variable signal on both upper and lower sidebands but transmits the 9960 reference on one sideband only. This is in contrast to the standard VOR that transmits all signals on both upper and lower sidebands. The test equipment used in our industry simulates the standard VOR. If you had a VOR receiver with a defective IF bandpass, it may work fine on the bench because there are two sidebands to provide the detected composite video. It may also work on standard VORs but the doppler VOR that utilized the defective sideband in this IF would not perform properly. Our experience is when you have IF bandpass problems that tuning will not correct, the culprit is often the crystal filters. Have you ever had a VOR receiver on the bench that the customer said did not work on some VORs but worked on others? Ascertain if any were doppler VORs and check the IF bandpass!
When testing receivers of all types the signal generators are specified to produce either soft or hard microvolts (mV). A soft microvolt is one measured directly from the generator. A hard microvolt is one that is derived from the output of a 6dB attenuator that has been placed in line with the generator output. Let us look into why this is done. Refer to the figure below:
A signal generator could be considered a voltage source VS with accompanying internal impedance RGEN. The receiver undergoing test has a RF input impedance RREC. The signal generators are typically calibrated with a load attached so that VS = 2 VIN only when RGEN = RREC. If the receiver impedance is not exactly equal to the generator impedance you are going to have errors in the attenuator reading on the generator. By placing a 6dB attenuator in series between the generator and the receiver you calibrate the generator output and isolate the receiver making it less susceptable to RF signal amplitude inaccuracies. The input 2VIN is lowered to VIN after passing through the 6dB attenuator, and connecting the receiver to the attenuator further drops the voltage by half as long as the RREC is 50 ohms. Example: 10mV shown on an IFR Nav 750B produces 20mV unloaded, 10mV is measured when the 6dB attenuator is attached, and 5mV when the Nav receiver is connected. Therefore 10mV soft becomes 5mV hard.
Last month we discussed interference in aircraft. Walter Shawlee 2, of Sphere Research Corp., wrote an excellent article in 1990 called Mysterious Noises that he has graciously allowed the AEA to distribute to those interested. It covers in detail installation tips and plenty of technical information. Contact AEA headquarters for a copy.
Next Month: Omni Bearing Selectors