I started with a ship full of max-level veterans, and I'm just a complete noob.

The next morning, the innermost worktable in Workshop No. 1 was cleaned up.

A piece of white cotton cloth was laid on the table, and the four corners of the cloth were taped to the edge of the table so that it would be visible if something fell.

Two 100-watt incandescent bulbs were installed directly above the workbench, with homemade spotlights covering the bulbs, so that the light was focused on a palm-sized area in the center of the table.

The issue with the local oscillator was resolved last night, and today all our energy is focused on the mixer diode.

Old Sun sat on a high stool with his back straight, a pair of binocular magnifying glasses hanging on his chest. The lenses were modified from high-powered eyepieces borrowed from the Institute of Physics, capable of magnifying twenty times.

He held a micro-motion clamp in his right hand, with a tungsten wire clamped at the head of the clamp. The tungsten wire was so thin that it was almost invisible under the light, and only when it was turned to a certain angle could a sliver of reflection be captured.

Da Liu stood half a step behind Lao Sun, in front of him was a homemade constant current source. There were two knobs, one controlling the current magnitude and the other controlling the pulse width. The terminals were attached to two copper rails on the worktable by alligator clips.

Jiang Ming stood to the side of the workbench, holding a pencil and a notebook in his hand, the pencil tip resting on the first blank line.

"The first one, let's begin."

Old Sun took a deep breath and used a suction cup in his left hand to fix a round germanium chip onto the stage. The chip was one millimeter in diameter and its surface had been ground and polished, making it look like a drop of solidified mercury under the light.

The micro-motion fixture in the right hand slowly descends, and the tungsten wire needle tip is aligned with the center of the wafer.

This tungsten wire was made by Jiang Ming last night using an electrochemical corrosion method. First, the tungsten wire with a diameter of 0.1 mm was vertically immersed in a potassium hydroxide solution and then an alternating current was applied. The tungsten at the surface of the liquid was corroded into a cone shape with a tip diameter of about 30 micrometers and a cone angle of about 15 degrees.

Thirty micrometers, which is half the thickness of a human hair.

Old Sun pressed the magnifying glass down, and the tungsten filament needle tip in his field of vision turned into a huge silver cone, while the grain patterns on the surface of the germanium wafer were clearly visible, like a patch of gray frozen soil.

The needle tip was about ten micrometers away from touching the germanium surface. Old Sun stopped, placed his thumb on the micro-motion knob, and turned the knob one full turn, moving the needle tip down five micrometers.

He turned the multimeter two notches, and the needle tip touched the germanium surface. The current pointer on the multimeter jumped, indicating a contact resistance of about 800 ohms.

"Current preparation".

Da Liu adjusted the current knob of the constant current source to the preset position and set the pulse width to ten milliseconds.

"put."

Da Liu pressed the pulse button, and a three-milliampere current pulse was injected into the contact point of the germanium wafer through the tip of the tungsten wire. Within ten milliseconds, the temperature of the contact area rose sharply, and the trace impurities at the tip of the tungsten wire melted and diffused on the contact surface, forming a PN junction with a diameter of about five micrometers.

The forward resistance on the multimeter jumped from 800 ohms to 120 ohms, and the reverse resistance jumped from 800 ohms to over 5,000 ohms.

"One hundred and twenty in the forward direction, five thousand in the reverse direction." Jiang Ming wrote the numbers down in his notebook, then shook his head. "The reverse current is not high enough, the leakage current is too high, it's not up to standard."

Old Sun lifted the micro-clamp and used tweezers to remove the germanium chip and put it into a cardboard box labeled "waste".

"The second one."

With the new germanium chip installed and the new contact position, Liu adjusted the pulse current from 0.2 mA to 2.8 mA.

This time the forward resistance dropped to 90 ohms, but the reverse resistance was even worse, at only 3,000 ohms.

废。

Thirdly, adjust the contact pressure by turning the micro switch knob only one and a half turns to make the needle tip press more lightly.

The forward voltage was 150 and the reverse voltage was 8,000, which seemed acceptable, but when Jiang Ming used a high-frequency signal generator to measure the junction capacitance, the result was above the standard.

废。

If the junction capacitance is too large, it means that the contact area between the needle tip and the germanium surface is too wide, and the high-frequency signal will be shunted by the capacitor, resulting in a sharp decrease in mixing efficiency.

To reduce the junction capacitance to below one picofarad, the contact area must be as small as a few micrometers, the needle tip must be finer, the pressure must be lighter, and the current pulse must be shorter.

From the fourth to the eighth, Lao Sun tried different combinations: the pressure increased from one bar to half a bar to a quarter bar, the current increased from three milliamps to two milliamps to one and a half milliamps, and the pulse increased from ten milliseconds to five milliseconds to three milliseconds.

Each one has different problems: some have too high forward resistance and cannot form an effective junction, some have excessive reverse leakage current, and some still cannot reduce the junction capacitance.

On the ninth chip, the tungsten filament slipped upon contact, and the needle tip left an invisible scratch on the germanium surface, rendering the chip unusable.

The tenth one, when Da Liu pressed the pulse button, his hand trembled slightly, and the current lasted for about twenty milliseconds. The doubled duration caused the contact area to overheat, and a black spot appeared on the germanium surface. It burned through.

From the eleventh to the thirteenth, Lao Sun began to adjust his breathing rhythm, making sure to operate the knobs at the end of each exhale, so that the movement of the chest cavity would have the least interference with his fingers.

The parameters of the thirteenth one are very close: 110 ohms in the forward direction, 12,000 ohms in the reverse direction, and a junction capacitance of 1.1 picofarads.

The difference is 0.1 picofarad.

"The contact point might be off-center." Jiang Ming looked at the changing trends of the thirteen sets of data in the notebook. "The surface of a germanium wafer is not uniform everywhere; the crystal properties will differ at different locations with different crystal orientations. Try the area about one-third of the way from the edge; the density of crystal defects is lower there."

Old Sun raised the magnifying glass, rubbed his eyes, and rested his fingers on the teacup next to him. The water in the cup had long since gone cold, but he took a sip without paying any attention.

"The fourteenth."

The new germanium chip, with the needle tip aligned with the off-center edge, applies pressure at a quarter-division mark, a current of 1.8 mA, and a pulse of 4 milliseconds.

Forward resistance is 95 ohms, reverse resistance is 9,000 ohms, and junction capacitance is 0.95 picofarads.

It finally dropped below one picofarad, but the reverse resistance dropped, falling from 12,000 back to 9,000.

Old Sun stared at the multimeter pointer for five seconds, pursed his lips into a thin line, and put the tube into the recycling bin.

"The fifteenth one."

Da Liu handed the new germanium chip to Lao Sun's left, and at the same time adjusted the pulse width of the constant current source from four milliseconds to three and a half milliseconds, a judgment he made after observing the previous data.

Old Sun didn't comment on Liu's adjustments; he simply pressed the magnifying glass down again and aligned the tungsten filament with one-third of the edge of the new chip.

The micro-knob turned a little more than a quarter of a turn, roughly between one-fifth and one-quarter of a turn. This amount cannot be represented by a scale; it relies entirely on the memory of the fingers.

When the needle tip fell, the multimeter's current pointer deflected slightly, indicating a contact resistance of 700 ohms.

"put."

Da Liu pressed the pulse button, and his finger steadily released it. Three and a half milliseconds later, the current was automatically cut off.

Old Sun didn't move, his hand still holding the clamp in place, his eyes fixed on the contact point through the magnifying glass. He waited a full five seconds before releasing his right hand, allowing the clamp to lift slightly.

Multimeter switched to positive mode: 105 ohms.

Switching to reverse: Above 15,000 ohms, the pointer is already close to full deflection.

Jiang Ming connected the probe of the high-frequency signal generator, set the frequency to 2.9 GHz, and measured the junction capacitance.

The meter pointer is stable at 0.8 picoseconds.

"0.8." Jiang Ming wrote the number into his notebook, paused on the paper with his pen, and then drew a circle next to it.

Fang Xudong had been standing at the other end of the workbench since 5 p.m., holding the frequency conversion loss measuring device brought from physics.

He clamped the fifteenth tube into the test socket, connected the local oscillator to 2.9455 GHz, the signal terminal to 2.95 GHz, and the intermediate frequency output terminal to a 4.5 MHz selective amplifier.

The pointer on the level meter at the intermediate frequency output jumped up. Fang Xudong turned the local oscillator power knob to find the optimal injection point and noted down the frequency conversion loss value.

"6.2 decibels."

He then removed one of the original Soviet-made germanium diodes from the captured mixer board and tested it under the same conditions.

"10.7 decibels."

The difference is 4.5 decibels, and the loss of the domestically made hand-rubbed tube is nearly two-thirds lower than that of the original Soviet-made tube.

Fang Xudong placed the two tubes side by side on the table. One was a standard product machine-packaged in a Soviet factory, with a smooth and regular glass shell. The other was one that Lao Sun had just manually dotted on a one-millimeter germanium sheet with a thirty-micrometer tungsten wire. It didn't even have a proper outer packaging, and the leads were temporarily glued on with silver solder paste.

But the parameters don't lie.

Old Sun got off the high stool, took off his magnifying glasses and hung them around his neck, then wiped the sweat from his forehead with his sleeve.

"These hands of mine have spent my whole life working with vises and files, and today I'm actually handling something thinner than a hair."

He smiled, his smile lines appearing at the corners of his eyes. He reached for the mug of cold tea on the table, took a big gulp of cold water, and his Adam's apple bobbed twice.

Jiang Ming closed the notebook, walked to the welding table, and pulled out a stack of graph paper and a sharpened pencil from the drawer.

He flattened the paper and wrote a few words in the upper left corner: Double-balanced mixing.

The pencil tip lands on the paper. The first line is a ring structure with four ports evenly distributed in the four quadrants of the ring. Each port is labeled with its function name: RF input, local oscillator input, intermediate frequency output, and isolation port.

Inside the ring structure, the symbols of four diodes are drawn on four arcs, connected end to end, with alternating directions, forming the core topology of the double-balanced bridge circuit.

Jiang Ming drew quickly, with clean lines and no corrections, as if the picture had been in his mind for a long time.

Fang Xudong walked behind him, looking at the gradually taking shape circuit diagram, without making a sound to disturb him.

The pen tip continues, extending outward from the mixer core to the winding structure of the broadband balun transformer, the intermediate frequency filter network, and the pre-amplifier low-noise stage.

A complete circuit diagram of a dual-balanced anti-interference receiver preamplifier is being unfolded stroke by stroke on graph paper.