DE102010007346A1 - Method for operating switch-mode power supply with power factor correction, involves comparing current-measuring value with saw tooth signal for determining measuring time point, and determining current-middle value from measuring value - Google Patents

Abstract

The method involves controlling switches (S1, S2) of a switch-mode power supply with operating frequency, and measuring input current (il) and output voltage per period of the operating frequency. The measurement of the current and the voltage is carried out in the middle of falling or raising flank of the input current. An input current-measuring value is compared with a saw tooth signal for determining a measuring time point, and a current-middle value is determined from the input current-measuring value, where the power supply is operated in a continuous operating condition.

Description

The invention relates to a method for operating a switching power supply with power factor correction, wherein the switching power supply has a digital measuring and control device and at least one switch which is driven at an operating frequency, wherein the input current and the output voltage per period of the operating frequency are measured once.

The 1 shows by way of example a signal flow diagram of a power factor correction according to the prior art. The output voltage v _o and the target output voltage v _o * are in a subtractor 6 subtracted from each other. The result is an error voltage v _e , which corresponds to the deviation of the output voltage from the desired output voltage and to a voltage regulator 2 is forwarded.

The voltage regulator 2 provides a dependent signal in a multiplier 7 is charged with the input voltage v _i . The result is in an adder 8th with the input current i _L and charged to a pulse width control 4 forwarded.

As a result, at least three measured values are required: input current i _L , input voltage v _i and output voltage v _o .

Such a method can be realized analog or digital. The digital realization offers some advantages, including the possibility of non-linear control and execution of calculations, higher flexibility and better reliability.

In the digital implementation, for example with a microcontroller, the measured values must, however, be digitized so that they can be processed by the microcontroller. At an average digitizing speed of 1 μs, at an operating frequency of 100 kHz, less than 10 measured values per period can be determined. Therefore, methods have already been developed in the prior art, which manage with only one measured value per period. The difficulty is to determine the right time to measure and to determine the correct average current from the measured value.

Nevertheless, three different measured variables are to be recorded, which is expensive. However, in the increasingly used for cost reasons bridge-less power supplies, the digital measurement of the input voltage is possible only with increased circuit complexity, which is why so far only analog circuits are known.

The object of the invention is therefore to provide a method of the type mentioned above, which can be used in particular for digital bridgeless circuits and in which no measurement of the input voltage is necessary.

This object is achieved in that the measurement takes place substantially in the middle of the falling or rising edge of the input current, that is compared to determine the measurement time of the input current measurement of the previous period with a sawtooth signal and that from the input current reading a current Average value is calculated.

The method according to the invention can be used in particular with bridge-free switched-mode power supplies with digital control, since no measurement of the input voltage is necessary. As a result, the circuit complexity is reduced, which is why a simple and cost-effective implementation is possible. The calculations and controls can be realized in a microcontroller or microprocessor.

The method according to the invention is suitable both for operation in continuous operation mode (CCM) and in discontinuous conduction mode (DCM).

Since the current edge in the CCM is linear and the measurement time is essentially in the middle of the edge, the current average is calculated from the input current according to the following formula: i _avg (n) = i _s (n).

Although in the DCM the current edge is linear, it is zero at the beginning and / or at the end for a time, therefore a correction is necessary. The mean current value is calculated here according to the following formula from the input current i _s , the output voltage v _o , the switching _instant and the period T: i _avg (n) = i _s (n) * v _o / (v _o - 2 i _s (n) / (T - t _on ))

From the comparison of the input current measurement value of the previous period with the sawtooth signal, a switch-on time for the switches is first determined, from which the measurement time is then derived.

The measurement can take place in the rising or falling edge of the current.

In the rising edge of the input current, the measuring instant is determined according to the following formula: t _sample (n) = (T + t _on (n)) / 2 where T is the period duration and t _{on is} the switching instant of the gate signal.

For the measurement in the falling edge of the input current, the measurement time is determined according to the following formula: t _sample (n) = t _f (n) / 2, With t _f (n) = (T -t _on (n)) * v _in / (v _o -v _in ), v _in = 2i _s (n-1) / (T -t _on ) and
i _s (n-1) is the input current reading of the previous period.

The invention is explained in more detail below with reference to two embodiments with reference to the accompanying drawings.

It shows:

1 a signal flow diagram of a power factor correction according to the prior art,

2 a circuit diagram of an electronic circuit for implementing the method according to the invention,

3 a signal flow plan according to the invention,

4 a diagram with an exemplary waveform and detection of the current in the rising edge and

5 a diagram with an exemplary waveform and detection of the current in the falling edge.

In 2 is an example of the circuit 1 a bridgeless switching power supply shown as is well known in the art.

The circuit 1 serves as a basis for the realization of the method according to the invention. The circuit 1 has two switches S1, S2 arranged in two parallel circuit branches. In series with the switches S1, S2, a diode D1, D2 is connected in each case. Between the switches S1, S2 and the input voltage AC, a coil L is arranged, which is divided into two coil cores. The function of the circuit 1 is sufficiently known and will therefore not be explained here.

The 3 shows a signal flow diagram of a method according to the invention. The input variables here are only the input current i _L , the output voltage v _o and the output target voltage v _o * necessary. The output voltage v _o and the input current i _L are measured in each period of the operating frequency by a measurement. From the two voltage values, an error voltage v _e is calculated as a difference and sent to a voltage regulator 2 forwarded.

From the input current, a current average value i _{avg is} calculated in a digital computation _element , which corresponds to the current that has flowed averaged over a period. This is necessary because only one measured value per period is available.

The error voltage v _e (n) and the current average i _avg (n) are in a pulse width control 4 is processed, which adjusts the gate signal d accordingly, so that a power factor correction is achieved. Conveniently, the voltage regulator, the computing element and the pulse width control can be realized in a digital microcontroller or microprocessor. Due to the programmability, non-linear characteristics can also be achieved.

In comparison to the prior art, no measurement of the input voltage is necessary for this, which considerably simplifies the technical complexity of the circuit and the control, in particular in the case of bridge-less switched-mode power supplies.

The 4 shows an example of a signal diagram to which the inventive method is explained in more detail. In the middle of a gate signal d is shown, which controls one of the two switches S1, S2 of the switching power supply. When the gate signal d is turned on, the switch is turned on and current i _{L flows} through the coil L. The current i _L through the coil L is shown at the bottom. Due to the inductance of the coil L, the current in the coil L does not increase immediately, but substantially linearly. At the top, a sawtooth signal w is shown, which is generated by a separate generator or can be realized digitally.

The switching time t _on (n) of the gate signal d is determined by comparing the switching time of the preceding period t _on (n-1) with the sawtooth signal. As a result, the optimum switching time can be determined at any time, wherein the duty cycle and / or the frequency of the gate signal is variable. The first switching time t (1) is fixed, so that a start of the method is possible.

According to the invention, the measurement takes place in the middle of the rising edge of the stream. Depending on the switching time t _on and the measurement time t _{sample is} determined at which the measurements of the input current i _s (current through the coil L) and the output voltage v _o take place. Preferably, the measurements are made at the same time.

The measurement time is determined by the formula: t _sample (n) = (T + t _on (n)) / 2, where t _on the switching time of the same period n and T is the period. This ensures that the switching time is always substantially in the middle of the rising edge.

From the measured value i _s (n) for the input current at the measurement time t _sample (n), a mean current value i _avg (n) is then calculated.

Will the circuit 1 Operated in so-called continuous conduction mode (CCM), the mean value i _{avg corresponds to} the measured value i _s , since the measured value was determined in the middle of the linear rising edge.

If the circuit is operated in so-called discontinuous conduction mode (DCM), a correction is necessary because the current at the beginning of a period remains zero for a time. The average current value results in i _avg (n) = i _s (n) * v _o / (v _o -2 * i _s (n) / (T -t _on )), where v _{o is} the output voltage at the measuring instant n is.

Instead of the measurement in the rising edge, the measurement can also take place in the falling edge of the current. A corresponding signal diagram is in 5 shown.

The method differs only in the changed determination of the measurement time: t _sample (n) = t _f (n) / 2. Here, t _f (n) = (T -t _on (n)) * v _in / (v _o -v _in ), where v _{o is} the output voltage and v _{i is} the input voltage. Since the input voltage is not available as a measured value, it must be calculated according to v _in = 2i _s (n - 1) / (T - t _on ). Where i _s (n-1) is the input current reading of the previous period.

The current average i _avg is then calculated according to the same formulas as in the rising edge measurement.

LIST OF REFERENCE NUMBERS

1

Circuit switching power supply

2

voltage regulators

3

digital arithmetic unit

4

Pulse width control

v _i

input voltage

v _o

output voltage

v _o *

Output voltage command

v _e

fault voltage

i _L

Input current through coil

i _s

Measured value input current

i _avg

Current mean

t _on

Switching time switch

t _sample

Measuring time

d

gate signal

w

sawtooth

T

period

Claims (8)

Method for operating a switching power supply with power factor correction, wherein the switching power supply has a digital measuring and control device and at least one switch (S1, S2), which is driven at an operating frequency, wherein the input current (i _L ) and the output voltage (v _o ) per Period of the operating frequency are measured once, characterized in that the measurement takes place substantially in the middle of the falling or rising edge of the input current (i _L ), that for determining the measurement time (t _sample ) of the input current measurement (i _s ) of the previous Period is compared with a sawtooth signal (w) and that from the input current measured value (i _s ) a current average (i _avg ) is calculated. A method according to claim 1, characterized in that the circuit in the continuous mode (CCM) is operated. A method according to claim 2, characterized in that the current average (i _avg ) is _calculated according to the following formula from the input current (i _s ): i _avg (n) = i _s (n) A method according to claim 1, characterized in that the circuit is operated in the non-continuous operating state (DCM). A method according to claim 4, characterized in that the current average (i _avg ) is _calculated according to the following formula from the input current (i _s ), the output voltage (v _o ), the switching time (t _on ) and the period (T) : i _avg (n) = i _s (n) * v _o / (v _o -2 * i _s (n) / (T -t _on )). Method according to one of Claims 1 to 5, characterized in that a switch-on instant (t _on ) is determined by comparing the input current measured value (i _s ) of the previous period with the sawtooth signal (w), from which the measuring instant (t _sample ) is derived. Method according to one of claims 1 to 6, characterized in that the measurement takes place in the rising edge of the input current (i _L ) and the measurement time (t _sample ) is determined according to the following formula: t _sample (n) = (T + t _on (n)) / 2 Method according to one of claims 1 to 6, characterized in that the measurement takes place in the falling edge of the input current (i _L ) and the measurement time (t _sample ) is determined according to the following formula: t _sample (n) = t _f (n) / 2, With t _f (n) = (T -t _on (n)) * v _in / (v _o -v _in ), v _in = 2i _s (n-1) / (T -t _on ) and i _s (n-1) is the input current measurement value of the previous period.

Images