Thermal environment during descent

The temperature of the air around a probe may change dramatically as it descends. For the deep atmospheres of the giant planets, and the lower atmospheres of Earth, Venus and Titan, the temperature rises almost linearly with depth.

In the deep atmospheres, the heat transfer by convection from the gas around the probe becomes prominent, and the infrared opacity of the gas prevents the probe from 'seeing' the cold of deep space. As a result, the probe will tend towards the ambient temperature.

In certain locations, on the mid-latitude surface of the Earth and on Venus at an altitude of around 50 km, this ambient temperature is close to the desired operating temperature of electronic equipment. Elsewhere, however, the equipment must be held at a temperature offset from the ambient, and thus requires temperature gradients to be sustained within the probe itself.

For short-duration missions such as descent probes, the system may rely on the thermal transient. If it starts at a suitable temperature, and the exterior of the probe is suitably insulated, the probe will only warm or cool to unacceptable temperatures after the mission is over. This is the approach adopted in the short missions to date.

Passively sustaining a temperature gradient requires heat to flow across that gradient (like current through a resistor R with a potential held across it). In the case above, the temperature of the probe interior may be considered as the voltage on a capacitance, with the capacitance C being the analogue of heat capacity. The time constant with which this interior voltage approaches the externally applied voltage is simply RC. An acceptably long mission duration can sometimes be obtained by increasing either R or C or both - applying thicker or higher-performance insulation, or increasing the heat capacity of the interior, respectively.

As an example, the Pioneer Venus small probes had beryllium structural plates (e.g. Hennis and Varon, 1978; Lorenz et al., 2005), to minimize mass and maximize heat capacity. Pressurizing the interior with xenon gas also reduced the gas conductivity. Note that the Pioneer Venus spacecraft were sealed entirely (which presented significant development challenges, particularly for the instrument seals) while the Galileo probe was unsealed, but individual equipment boxes inside were pressure-tight.

In practice there are limits on both of these design trends. While arbitrarily good insulation can be made at a cost, there are usually practical limitations to the utility of doing so - penetrations through the insulation are usually required to allow sensor access to the environment, for example. Increasing the mass of the interior is obviously ultimately limited by the mass capability of the launch and entry system; other limiting factors include the practicable density that can be achieved, such that a mass increase requires a volume increase too (and thus an increase in surface area and hence insulation mass).

Another approach that can delay the onset of unacceptable temperatures is the addition of thermal ballast exploiting phase changes (Russian papers usually refer to these as thermal 'accumulators'). The change of phase from solid to liquid or liquid to vapour is accompanied usually by the absorption of a large quantity of heat - ice being perhaps the most familiar example. A somewhat more convenient material than ice is lithium nitrate trihydrate. This material has a transition temperature of 303 K (29°C) and a latent heat of 296kJkg_1. Soviet Venus landers (Figure 8.2) also incorporated 'sublimators' to reject heat. Similarly, the Apollo spacesuits rejected heat by evaporating water.

In the long term, the only way to prevent failure by overheating is to pump out the heat that leaks into the probe (or is generated within it - especially true of components with high local power dissipation such as transmitters). Since heat is being transported from cool to warm reservoirs, the expenditure of work is required. The generation of that work will itself require energy. On a small scale that might be achieved by stored energy inside the probe, but the tradeoff

- Ground tests

+ Flight data

- Ground tests

+ Flight data

40 60

Time (mins)

Figure 8.2. Temperature evolution of the gas inside the instrument container of the Venera 14 descent capsule. (Zelenov et al., 1988b) Curves show the region determined during tests in ground-test chambers; crosses indicate the records from the descent on Venus.

between that approach and the application of phase-change ballast is unlikely to be favourable except in special instances, such as where only a small detector needs to be cooled.

To generate the work required for heat pumping would therefore require an external energy source. On the Venusian surface, with little available light (and ambient temperatures in any case too high for photovoltaics to function) this would require mechanical or thermoelectric conversion of heat flow from a heat source at a higher temperature than ambient. Thus a radioisotope power source outside the probe could be used to drive a heat pump to keep the interior cool. Ironically, the maintenance of long-term low temperatures requires parts of the system to be at exceptionally high temperatures.

Many of the principles discussed here are familiar to people who go on picnics or camping trips. Usually it is desired to keep food or beer at a suitably low temperature, and hence it is placed in a cooler. The thicker the insulation of the cooler, or the more cold beers placed with in it, the longer is the time before the beer gets too warm. A common enhancement is to deposit bags of ice in the cooler. Finally, for long durations, some kind of heat pump or refrigerator is required, driven by an external energy source.

One further approach worth mentioning parallels the design of low Earth orbit satellites which jump from high to low equilibrium temperatures as they go from day to night, but never encounter either extreme because of suitably long time constants.

Insulation requires some discussion. Much terrestrial insulation functions largely by suppressing convection. This effect is of little utility in very thin atmospheres (where the air cannot transport much heat anyway) or in very thick atmospheres, where the conductivity of the gas is large, allowing efficient heat transport even when convection does not occur. Venera landers used a porous silica material, machined into blocks around the outside and the inside of the pressure hull. Although the material itself was not airtight (to allow the material to breathe and prevent it from being crushed by the increasing pressure), the outside was coated to minimize forced convection. (Zelenov et al., 1988a,b).

Solid plastics are an obvious approach, and were used on Pioneer Venus, although thermosoftening polymers like polyethylene and PTFE have very poor mechanical properties at temperatures above around 100 °C. Polymer foams may have rather better insulating properties, but these foams may have significant volatile contents, so outgassing may be an area of concern. The Huygens probe used a polyurethane foam Basotect (Figure 8.3).

Aerogel and fumed silica are very light foams with excellent insulation properties. These materials have so little solid material that they are often translucent, and to minimize the radiative transfer of heat through them may be doped with absorbing material (the Sojourner rover (Eisen et al., 1998) used a doped aerogel as an insulator). Note that despite the name, aerogel is quite rigid and brittle.

Appropriate choice of structural materials can influence thermal performance significantly. The thermal conductivity of stainless steel is much lower than that of aluminium; titanium is similarly a poor conductor. Beryllium, while a very

Aluminium top Seals at experiment / platform structure interface

Aluminium after cone + fore dome

Internal foam 45 to 70 mm thick

Mechanism brackets conductively decoupled from F.S. _

Boxes

Black paint, except batteries

Batteries

Radiatively + conductively decoupled

Titanium struts to orbiter

SED + ring: 15 layers of MLI Labyrinth foils

Front shield exterior

Rear face exterior: MLI 15 to 16 layers

HTP/Prosial: 2.1 mm thick CFRP/Honeycomb structure HTP/AQ60: 18.2 mm thick

Front face exterior: MLI 15 to 16 layers

RHUs

27 on main plate Radiative window of HTP/AQ60: 17.4 mm thick 8 on top plate white paint (0.17 m2 ) External MLI 15 layers

Figure 8.3. Thermal design features of the Huygens probe (from Jones and Giovagnoli, 1997).

Aluminium top Seals at experiment / platform structure interface

Back cover

External MLI :15 layers HTP/Prosial: 0.5 to 2.7mm thick Al structure 0.8 to 1.6 mm thick

Aluminium after cone + fore dome

Titanium struts to orbiter

Internal foam 45 to 70 mm thick

Titanium horizontal and vertical struts

SED + ring: 15 layers of MLI Labyrinth foils

Mechanism brackets conductively decoupled from F.S. _

Front shield exterior

Rear face exterior: MLI 15 to 16 layers

Boxes

Black paint, except batteries

HTP/Prosial: 2.1 mm thick CFRP/Honeycomb structure HTP/AQ60: 18.2 mm thick

Front face exterior: MLI 15 to 16 layers

Batteries

Radiatively + conductively decoupled

Front shield central section

CFRP/Honeycomb structure

RHUs

27 on main plate Radiative window of HTP/AQ60: 17.4 mm thick 8 on top plate white paint (0.17 m2 ) External MLI 15 layers

Figure 8.3. Thermal design features of the Huygens probe (from Jones and Giovagnoli, 1997).

difficult and expensive material with which to work (owing to the toxicity of beryllium dust), has excellent stiffness and heat capacity and was therefore used in the internal structure of the Pioneer Venus small probes.

One design for a Venus probe pressure vessel (38 cm diameter, able to hold a 15.6 kg payload) uses concentric spheres, the outer one of titanium alloy, the inner of stainless steel. The inner sphere distorts during entry deceleration, transferring loads via six short titanium struts that disengage (reducing heat leak) when the loads are removed. The space between the spheres is filled with a fibreglass felt and xenon gas. The effective thermal conductivity was 0.014 WmT KT1 at 20 °C and 0.054WmT KT at 460°C, leading to a total heat leak to the payload of 84 W (Hall et al., 1999).

0 0

Post a comment