Interview with the ICOM lead engineer of 756PROII

Japanese CQ magazine November 2001 issue has an article on IC-756PROII. It is an interview to the lead engineer of ICOM IC-756PROII. Of course it is written in Japanese. Here is my English summary for non-Japanese HAMs who are interested in it.

The major improvements of 756PROII are:

In the development of 756PROII, ICOM has focused on the improvement of the RX performance. The RX of 756PROII is made up of the BPF, first IF, second IF, third IF, A/D converter, DSP, D/A converter, and AF. It is very important to get them matching well all together, particularly the matching between analog and digital circuits are essential for the good RX.

It is not an easy job to design a good AGC loop. Each component of the loop should affect the signal quality. In 756PROII, the third IF signals are converted into the digital signals using a 24-bit A/D converter. The analog circuit must carefully be designed so that it can make full use of the wide dynamic range of the A/D converter. During the development, ICOM experimentally used a couple of OP amplifiers, which are too expensive to use in HAM gears, in order to try the performance limits of the RX. These experiments have been fed back to the design of 756PROII.

ICOM has redesigned a couple of RX stages for 756PROII. To put it concretely, the third mixer was completely redesigned. The third mixer converts 455kHz signals to 36kHz signals. 756PRO used a mixer IC; 756PROII, on the other hand, uses fast analog switches. The conventional receivers have a mechanical or crystal filter before the third mixer, and therefore the input to the third mixer is limited in bandwidth. 756PROII, however, does IF filtering with the DSP after the third mixer, and the input to the third mixer is limited only in 15kHz bandwidth. This is the reason the third mixer is one of the key components that dominate the RX performance of 756PROII. The third mixer should be designed to have as small distortion as possible.

Another refinement in the RX is the BPF in the front end. Inductors and capacitors used in the BPF would generate distortion. To keep the signal purity, the bigger the inductor, the better. For the BPF, 756PROII has much bigger inductors than 756PRO. In addition, the first mixer and the I/O circuits around it are redesigned for better IMD with less noise. In addition, 756PROII uses new Motorola’s PIN diodes for signal switching in the RX front end to reduce the distortion.

For the preamp 1, 756PRO used paralleled grounded gate FETs. 756PROII, on the other hand, uses a push-pull FET amplifier to reduce the second IMD.

The crystal filter installed in the first IF is carefully sorted out in the production phase.

The noise blanker is completely redesigned.

Quite a few users requested to improve the 50MHz sensitivity to capture the noise-level signals. The DSP filter with rounded shoulder shape gives a solution to the request. Theoretically, the filter with the rounded shoulder shape degrades the signal quality, but it sure gives better readability for the noise-level signals. People who like rag chew on 40 or 80 meters would dislike this shape factor. So 756PROII allows users to select the shapes of the DSP filter.

The DSP filter is technically interesting, but another reason to use the DSP for the IF filter is that the availability of mechanical and crystal filters for HAM use. It is skeptical that those kinds of filters are continuously available in the market forever in the future. The DSP filter is implemented by software and is easy in the quality control. IC-746 was the last transceiver that used the conventional analog circuit for the RX. ICOM would like to expand of the DSP use to the products in the lower price ranges.

(Oba JA7UDE’s IC-756Pro II page.)