Ancillary Controls for the Cooling System

FC-17 liquid cooling system
System failure mode	What the system sees	Automatic system response	Consequences of failure
Leak in tubing to the outside world (impossible for leak to be inside MVD)	Drop in fluid pressure (timing depends on location of sensor)	Shutdown of fluid cooling system Shutdown of all electronics Air cooling system remains on	Without electronics, no data can be taken until leak is fixed. Electrolytic fluid will not cause electrical shorts, but will require cleanup.
Pump failure or change in pump power level	Flowmeter shows change in flow	Immediate shutdown of cooling system and electronics in case of failure If change is gradual, temperature sensors will sense trouble and issue a warning/shutdown	Without electronics, no data can be taken. Pump will have to be accessed and repaired or replaced.
Chiller failure or unexpected change in power level	Rise in temperature of LDOs and in flow loop	Warning when temperature spec is exceeded Shutdown when highest acceptable operating temperature of electronics is reached	Temperatures requiring only a warning will not impact data or its acquisition. In the case of failure and subsequent shutdown, no data can be taken. The chiller may need to be accessed and repaired or replaced.
Loss of "dirty" power supply	Cooling system goes down: Flowmeter shows loss of flow as it turns off Temperature of LDOs begins to rise	Turn off all electronics Air cooling system is down already	No data can be taken until power comes back up. There is no physical damage to cooling systems.
Loss of "clean" power supply	Electronics are down, but flow, temp sensors are up	Leave cooling system running uncontrolled (it will turn itself off if limits are exceeded)	No data can be taken until power comes back up. There is no physical damage to cooling systems. If cooling system shuts down, it will need to be restarted.

Unified Air Cooling System
System failure mode	What the system sees	Automatic system response	Consequences of failure
Blower failure or change in power level	Sudden drop in air flow Slow rise in temperature of MCMs	Immediate shutdown of cooling system and electronics if air flow drops too much If change is gradual, temperature sensors will see it and issue a warning/shutdown	While electronics are down, no data can be taken. The blower will need to be repaired or replaced, or the filter changed.
Fan bearing overheats	Signal from the built-in temperature sensor	Warning at one temperature Shutdown of fan and electronics of the affected half at a higher temperature	In the case of slight overheating, water supply to the bearing should be confirmed to be at a standard level. In the case of shutdown, no data can be taken, and water flow will need to be investigated. The problem could also be in the blower itself.
Chiller failure or unexpected change in power level	Rise in temperature of MCMs	Warning when temperature spec is exceeded Shutdown when highest acceptable operating temperature is reached	Temperatures requiring only a warning will not affect data or its acquisition. In the case of failure and subsequent shutdown, no data can be taken. The chiller may have to be accessed and repaired or replaced.
Loss of efficiency in heat exchanger	Very gradual rise in temperature of MCMs and flow loop	Not applicable to ancillary controls; should be dealt with by making sure air is very clean in the first place, filters, and/or regular maintenance	Electronics will be operating at temperatures higher than initially specified. There is no physical damage to the cooling system. If radiation damage to the silicon detectors has occurred, a worsened signal-to-noise ratio may result from the higher temperature.
Leak to the outside world	Rise in temperature of main enclosure of MVD Eventual rise in temperature of MCMs Rise in humidity of flow loop	Warning when temperature inside enclosure rises above spec Warning if MCM temperature rises above spec Notification if humidity rises above spec Highly unlikely that this would bring about shutdown conditions	Vertex finding algorithms may need to be adjusted to account for change in dimensions of Rohacell components. The leak should be found and repaired at the earliest convenience.
Failure of chiller used for humidity control or leak in that chiller's water lines	Rise in humidity in flow loop	Warning when humidity spec is exceeded in either direction This should never cause shutdown, since we can recover good data if we know how much the Rohacell is changing	Vertex finding algorithms will need to be adjusted to account for change in dimensions of Rohacell components. Humidity chiller will have to be accessed and repaired. Note: this failure mode affects both halves of the MVD.
Buildup of ice, mold, or other obstruction on the dehumidifying coils	The liquid level sensor in the heat exchanger will signal at a different rate, which will be noticed in software Rise in humidity in flow loop	We will try to fix the problem by instructing the chiller to cool the coils further. If the problem is uncontrollable, the temperature will rise until temperature sensors notice trouble and turn the system off.	If the problem is severe, the heat exchanger will have to be accessed and cleaned out.
Loss of "dirty" power supply	Air cooling system goes down: Flowmeter shows loss of airflow Temperature of MCMs begins to rise	Turn off all electronics Cooling system is down already	No data can be taken until power comes back up. There is no physical damage to cooling systems.
Loss of "clean" power supply	Electronics are down along with detailed sensors, but F, T, H sensors are up	Leave system running uncontrolled (it can turn itself off if limits are exceeded)	No data can be taken until power comes back up. There is no physical damage to cooling systems. If system shuts down, it will need to be restarted.

This page last modified on 10 July 98 by Rachel Cunningham.