Final Design

Final Test Setup:

Side View

The final design of the test setup consists of a 1' by 1' panel structure with heating tape as the heat source, dry ice and a fan as the cold source, and thermocouples and a DAQ as the measurement system.

Heater Element:

Figure 1. Heater Blanket

This system utilized the heater tape's ability to ramp up at specific temperature increments to test a wide range of models. The heating blanket used can reach the maximum in flight operating  temperature of 411°C (772°F). The actual temperature used was a maximum of 121°C (250°F) because the aluminum panels were bound together with an epoxy that could withstand a maximum temperature of roughly 121°C (250°F). The blanket was purchased with a temperature controller that was tested to reach the specified temperature within ±5°C (±9°F).

Thermal Sensor:

Figure 2. Thermocouples

Figure 3. Thermocouple Wiring Setup

In order to measure a proper temperature gradient across the honeycomb panel, several temperature sensors were used to measure the temperature field across the face of the hot and cold sides of the specimen and the thermal insulation. The thermocouples were fabricated from 24 gauge J type thermocouple wire: the wires were welded together into beaded thermocouples and placed inside the test setup in accordance with our design standard. In order to obtain reliable readings from the thermocouples, the signals they produce must be amplified and filtered; for this purpose, thermocouple breakout circuitry will be used. After steady state was achieved, the data was monitored and collected through Arduino software.

The thermocouples have an operating range of -210oC to 260oC with an absolute accuracy of ±2.2oC. The Adafruit Universal Thermocouple uses a 3.3V to 5V power supply and requires 4 SPI pins in addition to Voltage in and Ground. The Arduino MEGA2560 has 53 digital pins to allow use of many thermocouples concurrently.

Sandwich Panels:

Figure 4. Honeycomb Panel Views 

With the intention of creating models similar to UTC Aerospace System’s nacelles, honeycomb sandwich panels were used. Heat transfer patterns through the sandwich panels are different than of solid Aluminum sheets because the honeycomb structure drastically reduces the thermal conductivity.

    Hot side:

The hot side of the panel was layered with thermocouples, a thermal pad, a heater blanket, and calcium silicate insulation to enforce heat transfer in one direction. The insulation also kept the panel from ambient effects of the room.

    Cold Side:

The cold side was perforated to 5% percent open area (POA) using a CNC machine. The holes allowed for acoustic dampening.

Figure 5. Honeycomb Panel Specifications

 Wind Tunnel :

Fan

Figure 6. Fan placed inside wind tunnel and sealed with duct tape.

The fan was 0.508m x 0.508m (20” x 20”) fan with the ability to provide a flow rate of 1.18 m3/s (2500 CFM).

Flow Straighteners and Insulation

In Figure 7, flow straighteners were used in order to prevent eddies forming inside the wind tunnel. Another benefit with these flow straighteners allowed for the turbulent flow to be converted to a laminar flow. This allowed for a more streamlined flow to interact with the top of the perforated aluminum panel which allows for consistent and steady cooling of the panel by forced convection. The material used for the flow straighteners were plastic straws.

Figure 7. Photo of flow straighteners placed inside the wind tunnel.

The material used for the wind tunnel was cardboard, and the material for the insulation around the wind tunnel was polystyrene. Polystyrene was used due to its cost effectiveness, but the insulation used around the panel and heating blanket itself was Calcium Silicate which is a far more effective material in insulation and capable of operating at a much larger temperature range (-20°C-925°C). The insulation was utilized in order to prevent the ambient room temperature from interfering with the data gathered.

Final Performance Results:

Figure 8. Forced Convection Comparison

Figure 9. Free Convection Comparison

Figure 10. Simulation and Empirical Data