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

"Remeasure the time constant for me."

Jiang Ming's words landed beside the recorder, which was still feeding paper. Huang Kun looked at the furnace temperature drop curve after the shutdown, and then left the No. 3 furnace key on the control panel.

"Resume operation only after the furnace temperature drops below 200 degrees Celsius. Archive all data from today, and submit the retest plan to me tomorrow morning."

Fang Xudong cut off the entire paper tape and numbered it according to the three parts: heating, oscillation, and cooling. Da Liu then cut off the power to the actuator, leaving only the temperature measurement circuit to continue recording.

Old Sun installed the newly machined split nut back into the propulsion mechanism. After turning the handwheel ten times, the axial movement was already within the range that the dial could distinguish.

Jiang Ming stayed at the recording table, matching each change in heating current with the temperature response. The calculated delay was always longer than the old record, and the gap in the high-temperature range was still widening.

The old records come from the early acceptance of Furnace No. 3. At that time, the two ends of the quartz tube were open, the heating coil only heated the empty tube, and the temperature was read intermittently by an external high temperature gauge.

The test that day was equipped with a dual platinum resistance protective sleeve, end caps, and transmission brackets. The addition of these components altered the heat transfer process, but this difference still couldn't explain the twelve-minute lag.

Lin Lanying brought over the records of all the quartz tube replacements. The current tube is 0.8 millimeters thicker than the old one, and its weight has increased by more than 10%.

"We had a thin-walled tube blown up last year, and we replaced it with this batch of thick-walled parts. We didn't have time to redo the old curves."

Jiang Ming added the mass of the quartz tube to the heat capacity column and included the weights of the ceramic protective sleeve and end caps separately. The re-estimated time constant was still lower than the measured value.

Before leaving, Huang Kun glanced at the calculation sheet and only re-highlighted the question mark.

"I need to retest the boundary tomorrow morning. Simply changing the position of a set of knobs won't solve the problem."

Back in 502 that night, Fang Xudong pinned the paper tape to the blackboard, connecting the time differences between the peaks and troughs with red lines, forming a gradually expanding closed loop from the three rounds of oscillations.

After checking the amplifier's zero drift, Da Liu confirmed that the offset of the two-stage DC amplifier during the entire test was below the allowable value, and the actuator's hysteresis error was also within half a division.

Old Sun was sitting at the workbench, adjusting copper nuts. After listening to the data from the three men, he looked up and asked a question.

"Are the pipes in the furnace usually left unused or sealed before use?"

Jiang Ming turned to the process page of the Institute of Physics. When the regional purification is in operation, both ends of the quartz tube need to be sealed, the purified protective gas is introduced into the inside, and the flow is maintained at the specified pressure.

On the day of the test, in order to facilitate the adjustment of the sensor and the inspection bracket, the end cap was only mechanically fixed, and both the exhaust port and the air inlet were in an open state.

"The no-load test follows the open conditions, while the formal operating conditions require a protective atmosphere."

Old Sun put the file back in the toolbox and reached out to tap the cooling curve on the blackboard.

"The furnace you tested today is different from the one you used when it was being loaded with materials."

Fang Xudong took the process sheet, found the protective gas flow rate and pipe pressure, and then wrote down the gas conditions on the edge of the recording paper.

"When the gas participates in convective heat transfer, the path of heat dissipation inside the tube will also change after the end cap is sealed, so the old time constant cannot be used directly."

Jiang Ming divided open quartz tubes, sealed quartz tubes, and loaded quartz tubes into three types of thermal boundaries, and the equivalent heat capacity and heat dissipation terms on the paper changed accordingly.

The formula can indicate the direction of the difference, but the actual parameters still need to be obtained from the step response, especially the convection coefficient brought about by the protective gas, which cannot be determined by existing data alone.

As the night shift bell rang from the end of the corridor, Jiang Ming put away the No. 3 furnace record, entered the confidential compartment, and checked the curtains and door latches one by one.

The bronze pages unfolded deep within his consciousness, and the outline of the long table in the precision calorimeter laboratory reappeared, with Kruger placing a quartz thermometer on its stand.

Jiang Ming explained the open test, end sealing and protective atmosphere conditions of Furnace No. 3 in turn, and wrote the twelve-minute delay and amplified oscillation in the center of the blackboard.

Kruger picked up the sketch of the quartz tube, first looking at the wall thickness and length, and then drawing two boundary lines at the end cap.

"You are measuring the heater's response in air, but you are using it to control the temperature inside the sealed quartz tube. The heat capacity and heat dissipation boundary of these two have been reversed."

"The mass of the quartz tube increased by more than 10%, and even after I added that to the calculations, the difference still doesn't match."

"You only added solids."

Kruger moved the chalk inside the quartz tube and wrote down three terms: gas density, isobaric heat capacity, and flow heat transfer. He also marked radiative heat dissipation on the outer wall.

"The sealed protective atmosphere participates in heat storage, the gas flow changes the heat exchange inside the pipe wall, and the gland cuts off the original natural convection path at the end."

Jiang Ming asked, "Gaseous gases have a limited heat capacity, yet they can extend the delay by so much?"

"Looking at the heat capacity of the gas alone, the proportion is indeed limited. However, the measurement point you control is hidden in the ceramic sleeve. The heat has to pass through the coil, the quartz tube, the gas, and the protective sleeve before it is transferred to the platinum resistance thermometer."

Kruger plotted the four elements as continuous thermal resistance and thermal capacity, and several sets of interconnected delays appeared in front of the measurement point.

"The twelve minutes you see include the furnace body response, the tube wall response, and the sensor response, but the controller treats them as the same inertial link."

Jiang Ming looked at the four sets of parameters on the blackboard.

"When re-measuring, the tube must be sealed in the formal protective atmosphere, the sensor must be kept in its original installation position, and then the heating power must be changed in a step."

"We also need to record the intake air temperature, flow rate, and pressure."

Kruger drew a mark next to each item.

"If the atmospheric conditions change once, the equivalent time constant will also change. If you are pursuing zero or one degree, the test boundary must be written into the operating procedure."

"The differential term amplifies the high-temperature noise of the platinum resistance thermometer, and fine, intermittent movements already appeared in the latter half of the first test."

"Add a low-pass filter to the differential channel to filter out measurement noise before allowing it to determine the temperature change trend."

Kruger drew a first-order RC network and required that the filtering time be much shorter than the furnace thermal inertia, while also being longer than the high-frequency measurement jitter.

How should the ratio and integral be adjusted?

"Find the proportional term again based on the critical gain with atmosphere, slow down the integral, and don't let the previous error continue to push the actuator after the furnace temperature changes."

When he turned to the last page of the experimental record, Kruger wrote the open and sealed conditions on the same line, with a clear dividing line in between.

"The measurement boundary is part of the model. Once the boundary changes, the old parameters will remain in the old model."

The laboratory in his mind faded into darkness. Jiang Ming returned to the cubicle, where the lamp wick still burned at its original flame, and the paper strips on the table remained at their peak position.

He immediately drew up a step test plan with a protective atmosphere, listing the intake temperature, flow rate, pressure, quartz tube wall thickness, and sensor position as fixed conditions.

The second sheet describes a corrected circuit: the proportional element is re-evaluated to find a critical value, the integral time is extended, and a first-order RC filter is added to the micro-branch.

Before dawn, Fang Xudong entered the cubicle again with the paper strip. Jiang Ming pushed two plans to him and asked him to check all the measurable parameters first.

"We can get a gas flow meter from the Yuanlei No. 1 exhaust line, and the pressure gauge can be obtained from the original equipment of the Institute of Physics. We are still missing a measuring point for the inlet temperature."

Fang Xudong took a pencil and drew a copper resistor next to the intake pipe, then looked at the thermal boundary model from the beginning.

"Yesterday's test at least left a complete oscillation cycle. Although the old parameters are obsolete, the critical range can still be used to limit the starting point."

"Retain it as a safety boundary, and retest all the tuning data."

At 8:00 a.m., Jiang Ming and his group re-entered the Institute of Physics. The No. 3 furnace had cooled to room temperature, and Huang Kun and Lin Lanying were standing by the control panel.

Jiang Ming laid out the retest plan, the first line of which stated that the quartz tube should be sealed according to the formal process and a protective gas of the specified flow rate and pressure should be introduced.

After reviewing the thermal resistance and thermal capacity equivalent diagram, Huang Kun moved his pencil along the coil, quartz tube, gas, and ceramic protective sleeve layer by layer.

Why did we overlook the right atmosphere yesterday?

"I brought the time constant from the old open test into the new loop, and the boundary conditions were wrong, so the subsequent parameters were naturally wrong as well."

Jiang Ming kept the original error on the first page of the solution, with a note next to it indicating the specific consequences that would cause response delays, integral accumulation, and oscillation amplification.

Lin Lanying turned to the operation page and saw that there were record columns for inlet air temperature, flow rate and pressure. She then took out the formal process card for furnace No. 3 from the data cabinet.

"The protective gas should be applied according to this card. The flow meter needs to be recalibrated, and the pipeline was just leak-tested last night."

Huang Kun kept looking at the equivalent model for a while before putting the pencil on the table.

"Where did you get the basis for dividing the open furnace, sealed furnace, and measuring sleeve into four groups of thermal inertia overnight?"

"Last night, we recalculated layer by layer along the path of heat entering the temperature measurement point, and then used yesterday's step curve to deduce the result. A single time constant cannot explain the double delay in the record."

Jiang Ming pointed to two turning points in the cooling section. The first point corresponds to the heater reducing its power, while the second point is where the quartz tube and protective sleeve continue to release heat to the sensor.

Huang Kun took out the notebook with question marks drawn on it yesterday, looked at the two turning points several times, and then added a line next to the question marks to retest the sealing conditions.

When he looked up at Jiang Ming again, his gaze, for the first time, left his age and resume and fell on the two sheets of paper filled with boundary conditions.

"Install the tubes according to the official atmosphere, seal all measuring points, and begin measuring the step response this afternoon."

The end cap of furnace No. 3 then closed, and the purified protective gas entered the quartz tube, causing the flow meter float to rise to the specified mark on the process card.

Fang Xudong pressed the recorder, and Jiang Ming set the heating power to the first step. The temperature line on the paper tape took a long time to rise.

Huang Kun stood in front of the recording table, looking at the curve that was slower than yesterday, and reached out to pull the old parameter table from the control panel.