GB2447875A

GB2447875A - Short Circuit Protection of DC to DC converter

Info

Publication number: GB2447875A
Application number: GB0706267A
Authority: GB
Inventors: David Dearn
Original assignee: Wolfson Microelectronics PLC
Current assignee: Cirrus Logic International UK Ltd
Priority date: 2007-03-30
Filing date: 2007-03-30
Publication date: 2008-10-01
Anticipated expiration: 2027-03-30
Also published as: GB0706267D0; GB2447875B

Abstract

Under normal conditions a switch 10 connecting a power supply to an inductor 30 of a power converter is switched in opposition to a diode bypass switch 20. Under conditions such as a short circuit switch 20 may be kept open and the inductor current passes through diode 21 so that there is a voltage across the inductor 30 and the inductor current decays sufficiently to compensate for a rapid increase in inductor current when the supply is applied, when switch S1 is on. A controller 70 may determine that switch20 is to be kept off by detecting when a voltage Vout on the output is less than a threshold or when the current returning to the supply exceeds a threshold. Controller 70 may determine that switch 10 is turned off and switch 20 is turned on when the inductor current exceeds a limit value, adjusted according to the voltage error and switch 20 may be kept open if the limit value exceeds a threshold. The first switch may be closed for a minimum blanking period to allow for transient surges. The power converter may be used to power an audio player, computer, mobile communicator, satellite navigator, a camera or a hand held device in a vehicle.

Description

Short Circuit Protection The present invention relates to power

converters and in particular, protecting against short circuits in DC-DC power converters.

DC-DC power converters typically operate by repeatedly connecting and disconnecting an input power supply to a reactive circuit to provide an output power supply. The output power supply's voltage and current can be controlled by varying the timing of the switching of the input supply. However, a number of factors need to be taken into account in providing suitable control of such converters. In particular, it is important to ensure that the power output from the converter stays within certain parameters to provide the required output voltage and current. It is also important to monitor and thus be able to control the converter under short circuit conditions in the load to prevent : * . possible damage to the converter or other components of the system coupled to it. * * I...

* : Figure 1 a shows the layout of a basic fixed frequency buck power converter 1. The S...

basic converter I includes an inductor 30 and a reservoir capacitor 40. The converter I operates by closing a switch 10 to connect the input power source V to the inductor 30 :.: : during a first phase ( 1. Current L in the inductor 30 starts to rise causing the capacitor to charge and current to be supplied to a load (not illustrated) connected to the output (Vour). The current I in the inductor 30 continues to rise until the end of the first phase pl when the switch 10 is opened.

The current L in the inductor 30 continues to flow into the load and the capacitor 40 but there is no return path for it back to the inductor 30 once switch 10 is open. Therefore, when switch 10 is open, the current L through the inductor 30 starts to decrease and the voltage across the inductor 30 starts to drop thereby increasing the voltage across the diode 50, eventually forward biasing the diode 50 and bringing it into conduction. With the diode 50 forward biased, this forms a current path through the inductor 30, load and diode 50 during a second phase p2 when the switch 10 is disconnected. Under normal operating conditions, the inductor current decreases due to the voltage across it, which is equivalent to the sum of the voltages across the load and diode.

Figure lb illustrates the waveforms associated with the switch 10, inductor current IL and voltages V, V0 during the operation of the converter I described above in connection with Figure Ia.

The converter 1 arrangement of figure Ia does suffer from the problem of having the load current passing through the diode 50 throughout the second phase (p2. This means that power is dissipated in the diode 50. Where the output voltage V0 of the converter 1 is relatively low compared to the diode voltage, this wasted power may be a significant proportion of the output power which means the converter has poor efficiency.

To overcome this, the diode 50 may be replaced or supplemented with a switch. Figure 2 shows an alternative converter 2 arrangement, where an additional switch 20 is : arranged to switch on, i.e. close, when switch 10 is opened. A switch control unit 60 * provides a control signal SI to switch 10, as before. However, the signal S I is also : provided to an inverter 100 which provides an inverted signal S2 to switch 20. In *.e.

practice somewhat more complicated circuitry may be used to ensure there is a dead-time between each of switches 10, 20 being turned on so as to ensure switches 10, 20 :.: . are never both closed at the same time.

The converter 2 operates in a similar way to the converter 1 of figure 1. During the first phase cpl, switch 10 is closed and the input supply V provides current IL to the inductor 30. At the end of the first phase (pi, switch 10 opens and switch 20 closes to provide the current path during the second phase (p2. The advantage of using the second switch 20 is that the voltage drop across the switch 20 can be engineered to be lower than that of the diode 50 of Figure Ia. The switch 20 is typically a MOS device which can have a low on-resistance RON. A low RON means that the power dissipated by the switch 20 and thus converter 2 is lower, which helps to improve efficiency of the converter 2.

Figure 3 shows the typical operating conditions i.e. waveforms, for the circuit of figure 2. During phase I spl, the current 1L in the inductor 30, rises as the applied input power supply V drives current into the load through the inductor 30.

During normal operation, the inductor current change LML during phase I would be: Change in inductor current LIL = - TdI

L

where V is the input supply voltage, V0 is the output voltage, L is the inductance of the inductor 30 and TdJ is the delay before switch 10 opens.

Once switch 10 is turned off and switch 20 is turned on, the inductor current L decreases at a rate corresponding to the voltage across the load (plus any voltage across the switch 20 and other circuit components, although these should be comparatively small relative to the load voltage). In steady state operation, the inductor current L will rise and fall as shown in figure 3 around the mean output current JAy. * I I...

In steady state, the duty cycle of switch 10 will be V:V0, but usually V may vary *.,.

within some range whereas V0 is desired to be close to some predetermined value, * regardless of output current demand. To achieve the correct duty cycle, switch control * ** unit 60 is usually embedded in a control loop, receiving signals representing the output : * :* voltage Vo and often a second signal representing the input current through switch 20 I..

to implement a "current-mode" converter. The signal representing the current may be derived from a series sense resistor or by a current mirror arrangement for example.

These feedback paths and current sensor are shown in figure 2.

in operation, a current-mode converter may, during p1, compare within the switch control unit 60 the sensed input current L with some limit value LIM, and turn off switch 10 when the input current L has increased to this value. This cycle-by-cycle limit value LIM is modulated over timescales of several cycles by Von-, or more exactly by the difference between V0j, and the desired output voltage. Thus, the duty cycle is adjusted to give the duty cycle and peak current required to provide the desired value of Vo11 under the prevailing load current.

It is normal and good practice for the current sensing to be disabled for a short period after the switch 10 is turned on, to allow for transients to decay away and thus to avoid false triggering and premature turn-off of switch 10. Also there will be some delay from when the sensed current reaches the threshold value and when the resulting signal propagates to switch 10 and starts to turn it off. Thus, there will be a minimum delay Idly between switch 10 turning on and switch 10 turning off, i.e. a minimum pulse width.

With the above converter 2 arrangement, it is difficult to provide short circuit protection if the high-side output (Vo) of the converter 2 becomes connected to ground. For example, if a short 35 occurs between the converter 2 output and ground, it is difficult to detect and control the current IL in the converter 2. As shown in figure 4, when a short occurs, during phase 1 (p1), the current in the inductor 30 rises rapidly as the full : .. supply voltage V is dropped across it. Since the output voltage Voijr has fallen well below the nominal value, the control loop will increase the "target" current I to some maximum value At some point in time, the inductor current L exceeds the * threshold Ij after a delay due either just to the propagation delay or also to the blanking delay, i.e. a minimum of TdJ into phase 1 p1, switch 10 is turned off and switch 20 turned on. S..

S

As switch 10 opens, switch 20 closes to provide the current path during p2. However, since the output is shorted to ground and switch 20 has only a very small voltage across it, the inductor 30 has virtually zero volts superimposed across it. Again, this ignores the voltage due to resistance in the circuit but that should ideally also be relatively very small.

Since the rate of current change in the inductor 30 is proportional to the voltage across it, the current L in the inductor 30 remains virtually constant throughout phase 2 p2.

However, during phase 1 p1, the inductor current L rises rapidly: Change in inductor current zI during short Circuit = VN X Idly

L

where V is the input supply voltage, L is the inductance of the inductor 30 and Idly is the delay before switch 10 opens.

In the case of a short circuit, the inductor current L rises rapidly during the initial period Tdl and then remains virtually constant throughout the second phase. It then rises again in the next cycle and so on. The inductor current L continues to increase until the resistance of the inductor 30, switches 10, 20 and other circuit components becomes significant enough to limit the further rise of the inductor current L. However, this may be at a high value. Also there is a risk that the current L through the inductor 30 will be high enough to saturate the core of the inductor 30, in which case, its inductance will decrease, further exacerbating the cycle-by-cycle increase in current L. : Another way of controlling the inductor current IL during a short circuit is to reduce the :..::: converter's switching frequency. Reducing the converter's switching frequency would ***.

still allow the inductor current IL to build up when switch 10 is closed as before but, the :r' current L will have longer to decay when switch 10 is off and switch 20 is on. One * * drawback associated with this method of controlling the inductor current is that the * "lower frequency" inductor current may have an affect on other components in the system, such as audio components for example wherein audible artefacts are produced *** as a result of the "lower frequency" inductor current.

Another way of controlling the inductor current L during a short circuit condition is to operatively increase the resistance of either or both of the switches 10, 20 for the duration of the short circuit. However, one drawback associated with this method of controlling the inductor current L is that the switches may be required to dissipate a substantial amount of heat which will have knock-on effects in the circuit/system design.

There is therefore a need to provide a convenient means for controlling the converter under short circuit conditions but without unnecessarily complicating the converter and associated system design.

Therefore according to the present invention, there is provided a power converter comprising: a first switch coupled between a first power supply rail and a first node; a second switch coupled between said first node and a second power supply rail; an inductor coupled between said first node and an output node; a diode connected in parallel with said second switch; and a controller arranged, in a first mode, to control said second switch in opposition to said first switch and, in a second mode, to keep said second switch open.

The present invention also provides a method of controlling a power converter, said power converter comprising a first switch coupled between a first power supply rail and a first node; a second switch coupled between said first node and a second power supply rail; an inductor coupled between said first node and an output node; and a diode connected in parallel with said second switch, said method comprising: controlling said second switch in :. opposition to said first switch, in a first mode; and controlling said second switch to remain :..::: openinasecondmode. * S..

: The power converter preferably comprises a comparator arranged to compare the voltage on * the output node to a reference voltage. In this way, the controller changes to operating in said second mode if said comparator determines that the voltage on said output is below : * :1.. said reference voltage. *..

Optionally, a current monitor can be arranged to determine if the current returning to the second power rail from a load connected to the power converter exceeds a predetermined threshold current. This helps to identify a short circuit situation. In a short circuit situation, the controller changes to operating in the second mode if the monitored current is above the threshold current.

Preferably, the controller controls the first switch to close during a first phase and open during a second phase. The controller determines the end of the first phase and the start of the second phase to occur when the current in the inductor exceeds a limit value. This limit value is adjusted from cycle to cycle based on the output voltage error. If the output suffers a short circuit condition, this value will rise due to the output voltage error. Eventually if it keeps rising, it will rise above a limit value threshold and the controller will change to operating in the second mode.

Advantageously, the first switch remains closed for a minimum blanking period after said first switch closes. l'his ensures that transient surges or start-up surges can be catered for with the short circuit protection operating which may prevent normal operation of a powered load.

The controller is preferably arranged so that in the first mode it opens the second switch and closes the first switch during a first phase and opens the first switch and closes the second switch during a second phase.

The first and second switches can be selected from any suitable switching device. More specifically, the switches may be PMOS devices, NMOS devices, bipolar transistors, or silicon controlled rectifiers. The switches do not necessarily need to be the same. I. * *

:..::: The present invention is ideally suited to operation in portable electronic devices such as, but not limited to: an audio player (MP3 player, radio, mobile phone); computing unit (PDA, laptop etc.); mobile communications unit (mobile phone, PDA, two-way * radio); satellite navigation unit; digital still camera or digital video camera. Furthermore, the power supplies can also be used in hand held devices as well as devices used or installed :.:: * in vehicles such as: cars; trains; boats or planes. ***

The present invention will now be described in more detail by reference to specific examples and with reference to the following drawings, in which: Figure la shows an example of a basic power converter; Figure lb shows waveforms explaining the operation of the converter of figure Ia; Figure 2 shows a modified power converter; Figure 3 shows waveforms explaining the operation of the converter of figure 2; Figure 4 shows waveforms explaining the operation of the converter of figure 2 when a short circuit occurs; Figure 5 shows a power converter according to the present invention; and Figure 6 shows waveforms explaining the operation of the power converter of figure 5.

The embodiment described below uses a modified control arrangement to independently control the switches 10 and 20 in the converter circuit 3 shown in figure 5. Like elements have been given the same reference numerals throughout. The switch control in this embodiment allows independent control of the switches 10 and 20 such that, in particular, they can both be turned off at the same time.

In this arrangement, under normal conditions, the circuit 3 operates in a similar way to the arrangement of figure 2. In an initial phase, p 1, the switch control 70 turns on i.e. closes, switch 10 to increase the current IL in the inductor 30. During this phase, p1, switch 20 is open i.e. is turned off. At the end of phase 1 p1, switch 10 opens and * . switch 20 closes thus providing the low resistance current path during phase p2. S.. *

* Converters of this type are normally used to provide an output voltage which is :.: * either fixed or variable over a predetermined range. In either case, the designer will be aware that the output voltage V0 should not, under normal conditions, fall below some predetermined minimum voltage V. The present invention makes use of this aspect by monitoring the output voltage Voy. The arrangement of figure 5 is therefore provided with means for monitoring the output voltage V0 to determine if it is above or below a predetermined minimum value V. This is in addition to the normal feedback of output voltage to the switch control unit described above with respect to figure 2.

A comparator 80 compares the output voltage V to the predetermined value V and provides an output signal 81 indicative of whether the output has fallen below V or not. This output is provided to the switch controller 70 for controlling the switching of switches 10 and 20.

When a short circuit 35 occurs between the output Vo and ground, the capacitor 40 is quickly discharged and the output voltage V0 rapidly falls to ground potential. As the output voltage drops below V, the output 81 of the comparator 80 changes state. The controller 70 responds by inhibiting the operation of the switch 20. Switch 20 therefore remains open throughout the converter cycle, whilst the short-circuit condition remains.

As shown in figure 5, a diode 21 is provided in parallel with switch 20. This may be a separate, i.e. discrete, diode or the parasitic diode of the switch 20 device itself, assuming an NMOS type transistor switch is employed as the switch 20. Once the comparator 80 has determined that a short circuit is occurring by reference to the output voltage V0 and subsequently the operation of the switch 20 is inhibited, then during phase 2, (p2, the inductor current L cannot pass through switch 20 and instead passes through the diode 21. Thus, the circuit operates in a similar manner to the circuit of figure 1 during a short condition. As a result, the voltage VD produced across the diode 21 ensures there is a voltage, typically 0.6v -0.7v, across the inductor 30 such that the current IL through the inductor 30 will decrease during phase 2, (p2.

S S...

:. The operation of the circuit 3 of figure 5 will now be described in more detail with * reference to figure 6. Figure 6 assumes that the short circuit occurs at a time t5, just :.: before the switch 20 turns off at the end of phase 2, 92. Of course, the short circuit could happen at any point in the cycle and the operation of the circuit would be fundamentally the same.

Before the time t5 at which the short circuit occurs, the inductor current 1L is ramping down as normal. Then the short circuit occurs, and the output V0 is effectively connected to ground so the output voltage V0 becomes zero. There is now no voltage across the inductor (Voiyr V = 0), so the current L becomes constant. At the start of phase (pi, switch 20 turns off and then, a relatively short time after, switch 10 turns on.

The full supply voltage V is therefore placed across the inductor 30. Consequently, the inductor current IL rises rapidly. The over-current detection detects the high current and switch 10 is turned off. The rate of rise of current L is likely to be high enough that the measured current passes the threshold value iLnx within the minimum pulse width Tdly so switch 10 will not be turned off until after a time TdJ. By then, the current 1L in the inductor 30 will have risen by: = V T

L

The output voltage Vo drops to ground potential and this is detected by the comparator 80. The switch control 70 has turned off switch 10 since IL exceeded 1u, but because the signal 81 from the comparator 80 is high, indicating a low output voltage Vo, switch 20 is not turned on. The inductor current L therefore flows through the diode 21 and the forward voltage VD of the diode 21 is expressed across the inductor 30. This forward voltage VD will tend to decrease the inductor current L* Consequently, over the rest of the cycle, i.e. during phase 2, p2, the inductor current L decreases by an amount: S. * S Al VDx(TTdJY) 2 L *5*5

S *...

* : where Al is the decrease in current through the inductor 30, V is the diode forward : ** voltage, L is the inductance value, and T is the switching period of the converter.

S

Thus, if the short remains, the current IL will initially increase at a rate VfL for a time I giving an increase in current IL and then tend to decrease at a rate VD/L for the remainder of each clock cycle giving a decrease in current. If T11 is such that Al1 is equal to Al2, then the increase in current IL will be exactly cancelled by the decrease in current 1L, and the circuit 3 will reach a steady state in which the current L reaches a constant value at the start of each clock cycle. If T is smaller, so that Al1 is smaller than A!2, then the current will collapse to zero before the end of phase 2, 4)2, causing also the diode voltage VD to collapse to zero, and the current L will stay at zero until the start of the next clock cycle (as shown in the second full cycle of Figure 6). If TdI is larger, so that M1 is greater than 12, then the current L will increase by an amount Ìi -L.12, each clock cycle, which is clearly undesirable.

To obtain an expression for the minimum value of Tdly, set = 12, that is: VxTdY YDX(T.TdJ

L L

The above can be arranged to determine the minimum duty cycle.

(v+vD)TdlY = T.V0 TdI = VD.IVID.

T (v+vD) vu :. Thus, the minimum duty cycle T,jfl' should be less than VWV to provide a stable control of the short circuit current. For an application with a 6V supply and a diode forward voltage of 0.6V, this would give a minimum duty cycle of approximately 10%.

This assumes a maximum supply voltage and a minimum diode voltage. In practice, these values may vary from the ideal due to process and temperature variations.

* Consequently, a shorter minimum duty cycle of perhaps 7% may be preferable in :.: practice, to ensure stable operation under short circuit conditions and extremes of tolerances on the diode forward voltage VD and supply voltage V. Even if this were not possible, the arrangement would still provide better control of short circuit current than the arrangement of figure 2.

More detailed analysis of this circuit would include effects such as LIM actually taking many cycles to increase to Ij because of the time constant of the output voltage control loop, and the possibility that the short might occur at any time in the cycle, but in the steady state the above conditions are valid.

By using the arrangement of Figure 5 described above, the normal current limit block used in normal operation of a current-mode converter can be used to effectively control the short circuit condition with little additional circuitry. As a current limit block is a normal requirement for a converter 3 of this sort then the invention can provide short circuit protection with only minimal modification of the switch control and the additional comparator 80.

This provides efficient operation of the converter 3 under normal operating conditions but effective control of short circuit currents during fault conditions.

The switches 10, 20 are preferably MOS devices which provide low resistance in use.

Switch 10 is preferably a PMOS device and switch 20 is preferably an NMOS device.

In the above embodiments, the voltage comparator 80 monitors the output voltage for use in determining if a short circuit has occurred causing the output voltage to drop below a threshold. However, other means may be employed to determine when a short circuit occurs. As stated above, under short circuit conditions, the voltage control loop would increase 1uii up to a maximum value k. By monitoring LIM and determining if it rises above a threshold value LthSPSI the control can determine that a short circuit condition is occurring. The control can then operate to disable the operation of switch S. :.: * As another alternative, the current in the load's return ground path can be monitored. The * average current would be monitored to determine if it exceeds a threshold level indicative that a short circuit has occurred. Again this can then be used to control the operation of the switch 20.

The above techniques could be applied to other forms of converter, for example converters with voltage mode control loops, though the current sense circuit might not already be there, so would need to be added.

Where current monitoring is used to determine if a short circuit has occurred, it is preferable to provide a blanking period such as that as mentioned above. This allows for initial surge currents or transient surges. If no blanking period were provided there is the possibility that normal operation would be interrupted due to the operation of the short circuit protection. The blanking period provides a minimum period, i.e. a delay, before which the short circuit protection begins to operate.

The invention has been described above in terms of specific embodiments. It should be noted that the above-described embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design many alternative embodiments without departing from the scope of the appended claims and drawings. The word "comprising" does not exclude the presence of elements or steps other than those listed in a claim, "a" or "an" does not exclude a plurality, and a single element may fulfil the functions of several elements recited in the claims. Any reference signs in the claims shall not be construed so as to limit their scope. I. * * * * * * **.

S * * S. * * . S.. *

S *55

Claims

CLAIMS: 1. A power converter comprising: a first switch coupled between

a first power supply rail and a first node; a second switch coupled between said first node and a second power supply rail; an inductor coupled between said first node and an output node; a diode connected in parallel with said second switch; and a controller arranged, in a first mode, to control said second switch in opposition to said first switch and, in a second mode, to keep said second switch open.
2. A power converter according to claim 1 further comprising a comparator arranged to compare the voltage on said output node to a reference voltage, wherein said controller changes to operating in said second mode if said comparator determines that the voltage on said output is below said reference voltage.
3. A power converter according to claim I further comprising a current monitor :. arranged to determine if the current returning to the second power rail from a load * .1* ***.* connected to said power converter exceeds a predetermined threshold current, wherein said controller changes to operating in said second mode if said current monitor determines that *** *..: the returning current is above said threshold current.

I

*
4. A power converter according to claim 1 wherein said controller controls said first switch to close during a first phase and open during * a second phase, and said controller determines the end of the first phase and the start ofthe second phase when the current in the inductor exceeds a limit value, said limit value being adjusted during operation according to the output node voltage error of the power converter, and said controller changes to operating in said second mode if said limit value exceeds a limit value threshold.
5. A power converter according to any one of the preceding claims wherein the first switch remains closed for a minimum blanking period after said first switch closes.
6. A power converter according to any one of the preceding claims wherein the controller is arranged, in said first mode, to open the second switch and to close the first switch during a first phase and open the first switch and close the second switch during a second phase.
7. A power converter according to any preceding claim wherein said first switch is selected from the group comprising: PMOS devices, NMOS devices, bipolar transistors, and silicon controlled rectifiers.
8. A power converter according to any preceding claim wherein said second switch is selected from the group comprising: PMOS devices, NMOS devices, bipolar transistors, and silicon controlled rectifiers.
9. A method of controlling a power converter, said power converter comprising a first switch coupled between a first power supply rail and a first node; a second switch coupled : between said first node and a second power supply rail; an inductor coupled between said first node and an output node; and a diode connected in parallel with said second switch, said method comprising: *.** : controlling said second switch in opposition to said first switch, in a first mode; and *. controlling said second switch to remain open in a second mode. * *. * . S
10. A method according to claim 9 further comprising switching from said first mode to said second mode when a voltage on said output node falls below a predetermined level.
11. A method according to claim 9 further comprising monitoring the current returning to the second power supply rail from a load connected to said power converter to determine if it exceeds a predetermined threshold current, wherein said controlling of said second switch changes to operating in said second mode if said monitoring determines that the current returning is above said threshold current.
12. A method according to claim 9 further comprising: controlling said first switch to close during a first phase and open during a second phase, and determining the end of the first phase and the start of the second phase when the current in the inductor exceeds a limit value, said limit value being adjusted during operation according to the output node voltage error of the power converter, and switching to operating in said second mode if said limit value exceeds a limit value threshold.
13. A method according to any one of claims 9 to 12, further comprising controlling the first switch to remain closed for a minimum blanking period after said first switch closes.
14. A method according to any one of claims 9 to 13 further comprising, in said first mode, closing the first switch and opening the second switch during a first phase and opening the first switch and closing the second switch during a second phase.
15. A method according to any one of claims 9 to 14 further comprising switching from said first phase to said second phase after said inductor current exceeds a predetermined : current level. * *I*. * * * *..
16. A method according to any one of claims 9 to 15 further comprising determining if **.

said power converter output has a short circuit and, if so, switching to said second mode. * *
17. A system comprising a power converter according to any one of claims ito 8. * ** * * * 1* *
18. An electronic apparatus comprising a power converter according to any one of claims 1 to 8.
19. An electronic apparatus according to claim 18 wherein the electronic apparatus is one selected from the group comprising: an audio player; a portable computing unit; a mobile communications unit; a satellite navigation unit; a still camera and a video camera.
20. A vehicle including a power converter according to any one of claims I to 8.
21. A power converter substantially as described herein with reference to the drawings.
22. A method of operating a power converter substantially as described herein with reference to the drawings. S. *1 * *.. * I **I.

I S... I. *.

I I S.. *

I II.

I