When working for a firm that made thick film amplifiers, I was given a task of designing a low-noise amplifier with a noise figure of 2.5dB, a gain of 15dB, and an input and output VSWRs of two to one or 9.54dB return loss. The frequency range was from 10 MHz to 300 MHz. That meant I could not use a double resistor series-shunt feedback resistor since that circuit configuration would raise the noise figure. The amplifier was to operate on a 15V supply drawing about 20 milliamps.
I decided to design the amplifier using a non-dissipated feedback consisting of a directional coupler incorporating a dual aperture core with number 32 bifilar wire color coded red and green. I built the directional coupler on a PC board and adjusted the turn's ratio to give the required tap for the gain of 15dB. I calculated the thick-film resistors to bias the chip transistor for the correct Class A DC operating voltage and current. I then had someone install the coupler on the thick-film ceramic carrier that was to be mounted on a 3-pin TO-8 header using silver epoxy.
After the unit was assembled and prior to installing the cover that was to be hermetically sealed, I tested the unit using an HP network analyzer. The gain and VSWRs all met specifications, so I approved the design for production. Three days later, 50 units were ready to be tested by the technicians in the quality department prior to sealing the caps.
Early in the testing, the technician came into my office and said that the gain was 13dB and the VSWRs were four to one and out of specification. I asked the tech to bring the units to my network analyzer where I tested them and they all met specs. I went to his network analyzer and sure enough, they exhibited high VSWRs and 13dB gain.
Looking over the analyzer setup, I discovered the output level of the network analyzer was set to its maximum of +20dBM. This meant the input signal was so high as to saturate the amplifier, and cause the gain to be low and the VSWRs to be high. I showed the tech how to lower the output level to -20 dBM wherein he tested the units and found them to be in spec.
I guess we all assume test equipment setup is automatic, but one has to be careful concerning the levels in and out of test equipment.
This entry was submitted by William J. Garner and edited by Rob Spiegel.
William J. Garner is an RF microwave engineering consultant with 47 years design experience. He has published papers in trade journals and holds seven patents.
Tell us your experience in solving a knotty engineering problem. Send stories to Rob Spiegel for Sherlock Ohms.