2.1 Hard wired

Hard wired means that a subsystem is directly connected to the power system, and this way there is no protection. For the subsystem this is the most reliable method. For the power system, however, this method of connection is potentially fatal. A short circuit on one of these lines will severely damage the power system, and therefore the satellite. On certain satellites, some designs have current limiting resistors, which will act as a fuse should the line stay shorted to ground.^{(Reference Ismail, Bakry, Selim and Shehata1,Reference Bekhti11–13)}

2.2 Fused lines

For fused lines, the subsystem is connected to the power system via a fuse, thereby protecting the power system from short circuits down the line. The disadvantage of fuses is that when they break, they cannot be reset. The reason for using fuses is dictated by space requirements as they take little space compared with power switches. Subsystems that must be switched on all the time, are low power, and have enough redundancy to survive an “accidental” break if one fuse are powered by fused lines.^{(Reference Ismail, Bakry, Selim and Shehata1,Reference Bekhti11–13)}

2.3 Resettable fuses

Resettable fuses have become an attractive circuit protection against overcurrent conditions due to their ability to cycle back to a conductive state after the current is removed, acting as circuit breakers. This allows the circuit to function again without having to replace it. Next-generation resettable fuses meet the standards and requirements for secure and reliable overcurrent protection within rated limits. Once system requirements are defined and circuit analysis has determined the operating parameters, selecting the right fuse for a specific design is rather simple.^{(Reference Grasselli, Schirone and de Luca5)}

2.4 Power switches

Power switches are the most flexible way of distribution. The power switch has two main functions. First, it can switch the subsystem ON and OFF by means of a telecommand. This is useful when it is not necessary to have a subsystem powered up all the time. It can also be a necessity to switch off a subsystem when the power budget cannot support the power demand, e.g. the high-power transmitter. Second, the power switch is a protection. The switch is also an electronic fuse that automatically switches OFF when the current drawn by the subsystem becomes larger than a pre-set value. There are two basic types of switches:^{(13–Reference Mohan, Undeland and Robbins16)}

• Bipolar junction transistor (BJT, positive and negative) switches

• Field effect transistor (FET) switches

2.5 The bipolar junction transistor switch

There are basically two types of bipolar switches in use: a positive switch and a negative switch. The positive switch can be subdivided into two categories:^{(13–Reference Mohan, Undeland and Robbins16)}

• Low power switch, up to 200mA trip current

• Medium power switch, up to 1A trip current

• High power switch, higher than 1A trip current

The negative switch has only a low power version because the −10V line can only support low power. ^{(13–Reference Mohan, Undeland and Robbins16)}

2.6 The positive bipolar transistor switch

The positive power switch, shown in Fig. 3, works as follows: In the OFF state, the command line is low. The base of Q₂ is low, and there is no current out of the base of Q_1. Therefore, Q₁ is OFF and the output of the switch is low. Diodes D₁ and D₃ ensure that the switch stays OFF when the telecommand is low.

Figure 3.

The positive bipolar power switch.

When the command line becomes high, a pulse is generated through the capacitor, which switches transistor Q₂. Transistor Q₁ is ON, and the collector goes to the same level as the emitter. When the output of the switch is ON, the resistor bridge divider (R₅, R₆) sets the base of Q₂ at 3.5volts. This keeps Q₂ ON and keeps Q₁ ON, so the switch stays ON.

Switching OFF uses the same principle as switching ON. When the telecommand line goes low, a negative pulse from the capacitor forces the base of Q₂ to go down to zero; thereby, the switch goes OFF. D₂ ensures that the base does not go below zero (−0.6V).

The automatic switch OFF of the switch is obtained as follows: when the switch is operating nominally (normal output current) transistor Q₁ is in saturation. When the current through Q₁ increases, the transistor, at a certain point, will come out of saturation. When that happens, the voltage drop across the collector and emitter (Vce) of Q₁ increases; this causes the output to go down and the voltage at the base of Q₂ to go down. Because of that, Q₂ will be less saturated, so the base current of Q₁ decreases, which causes Q₁ to be less saturated until the switch goes OFF.

Because a subsystem requires a higher current than normal (higher than the trip current of the switch) when switched ON (capacitor charging), the switch must be capable of large initial (in-rush) currents. This is done by the capacitor C₁. As long as the capacitor’s pulse keeps the base of Q₂ above the nominal 3.5V, the in-rush current can be higher than the trip current. By increasing the capacitor, the in-rush capability is increased. A 1F capacitor is judged enough for most subsystems. Some subsystems require larger capacitors; therefore, a 10F electrolytic capacitor can be used.^{(13–Reference Mohan, Undeland and Robbins16)}

The rest of the components are estimated as follows: The transistors, Fig. 3, are chosen according to the current handling of the switch. The trip current of the switch and its components are highly dependent on the H_fe of the transistors. The calculation gives only an estimate, and the real values of the resistors must be selected for each switch. The resistor values of the divider chain R₅ and R₆ are chosen such that the base of transistor Q₂ is at 3.5V. To keep the trip current of the switch as independent as possible from the H_fe of Q₂, the current through R₅ and R₆ must be larger than the base current of Q₂. If this current is about 5 times higher and the H_fe of both transistors is about 100, the current through R₅ and R₆ is given in Equation (1):^{(13–Reference Mohan, Undeland and Robbins16)}

(1)\begin{align}{\text{I = trip current/500}}\end{align}

So, the resistance value of R₆ is equal to, Equation (2):

(2)\begin{align}{{\text{R}}_{\text{6}}}{\text{ = \{3}}{\text{.5V * 500\} /trip current}}\end{align}

And the resistance value of R₅ is equal to, Equation (3):

(3)\begin{align}{{\text{R}}_{\text{5}}}{\text{ = \{voltage drop }}{{\text{R}}_{\text{5}}}{\text{*500\} /trip current}}\end{align}

The voltage drop over R₅ is different for a 5V, 10V or 14V switch. The value of the resistor R₄ sets the trip current of the switch and depends greatly on the H_fe of Q₁. A lower H_fe requires a larger base current and thus a lower resistor value for R₄.

An approximate value of R₄ is given in Equation (4):

(4)\begin{align}{{\text{R}}_{\text{4}}} = \{\{\text{Base voltage} {{\text{Q}}_{\text{2}}}-\left({\text{2* diode drop}}\right)\} * {{\text{H}}_{{\text{fe}}}}{{\text{Q}}_{\text{1}}}\}\text{/ trip current}\end{align}

Table 1 summarises values for the resistors R6, R5 and R4 for different positive BJT switches. The resistors values are produced for different voltages, i.e. +5V and +10V and different trip currents. Note that positive BJT switches are designed in this case for a large interval of trip currents.^{(13–Reference Mohan, Undeland and Robbins16)}

Table 1

Bipolar positive power switches

2.7 The negative bipolar transistor switch

The negative switch, Fig. 4, works according to the same principles as the positive one except for the current flowing in the other direction. Also, the buffer is of the inverting type so switching ON is a negative going pulse and switching OFF as a positive pulse. The telecommand remains the same.

The power switches have one problem – they are temperature sensitive. Because the H_fe of the transistors drops with lower temperatures, the trip current is set approximately two to two-and-a-half times the nominal expected current. This is to compensate for the temperature coefficient of the trip current, which is about 0.5%/°C. It also incorporates the decrease in H_fe of the transistor and an increase in power consumption by the subsystems.

Figure 4.

The negative bipolar power switch.

Table 2 summarises values for the resistors R6, R5 and R4 for different negative BJT switches. The resistors values are produced for different voltages, i.e. −5V and −10V and different trip currents. Note that negative BJT switches are designed in this case for lower power.^{(13–Reference Mohan, Undeland and Robbins16)}

Table 2

Bipolar negative power switches

2.8 The field effect transistor switch

Since bipolar transistor switches become inefficient at higher currents, a power switch based on a FET has been developed (see Fig. 5). The principle is the same as for the bipolar power switch. The P channel FET is ON when the gate is low (zero volts). The FET is OFF when the gate is at the same voltage as the source.

Figure 5.

Field effect transistor switch.

In the OFF state, the output is low, and the input is high. The positive input of the comparator is kept at approximately 3.5V by the voltage bridge divider chain R₁, R₂ and is therefore higher than the negative input. Thus, the output is high, and the FET is OFF.

When the command line goes high, a 5V pulse via the capacitor C₂ makes the negative input higher than the positive, the comparator output goes low and the FET switch is ON. The switch settles in normal operation with the negative input just above the positive input.

Switching OFF works the same as the bipolar switch. The low going pulse from the capacitor forces the negative input to go below the positive input, causing the switch to turn OFF.

The trip current is obtained in the following way: The FET has a fairly constant internal resistance and so with increasing current drawn through the FET, the voltage drop across the FET increases linearly. When the voltage drop increases, the output voltage decreases so does the voltage at the negative input of the comparator. When the negative input falls below the positive, the switch turns OFF.

The trip current can be set by adjusting select-on-test resistor R₁₄. By making the voltage drop across these resistors larger, the voltage on the negative input becomes higher than the voltage on the positive input and this will make the trip current larger.

A low pass filter from the positive input of the comparator to ground is incorporated to avoid quick variations on the input tripping out the switch, the cut-off frequency of the low pass filter is given in Equation (5):^{(13–Reference Mohan, Undeland and Robbins16)}

(5)\begin{align} \text{f}_{{\text{cut - out filter}}}=1/[2p^\ast(\text{R}_{1}//\text{R}_{2}+\text{R}_{12})\ast\text{C}_{3}]\end{align}

With R₁ = 68kΩ, R₂ = 22kΩ, R₁₂ = 100kΩ and C₃ = 470 nF, and assuming that any resistance between the battery and the switch is of no significance, the cut-off frequency of the low pass filter is given:

\[{{\text{f}}_{{\text{cut - out filter}}}}{\text{ = 2905 Hz}}\]

By connecting the op amp to the 14V supply, the FET is switched OFF with 14V op amp output. The drain-gate voltage is then 0V. To switch ON the output of the op amp goes to 0V so the drain/gate voltage is −14V. The addition of R₁₅ will increase the stability of the switch as the voltage difference between the comparator input increases but has the disadvantage of decreased efficiency. The resistor R₁₅ is also used as a current monitor.

Table 3 summarises values for the resistors R10 and R14 for different FET power switches. The resistors values are produced for the voltage 14V and different trip currents. FET power switches in this case are designed to support current values up to 3500mA (trip current).^{(13–Reference Mohan, Undeland and Robbins16)}

Table 3

FET power switches