DE102006018771A1 - Electric component e.g. power semiconductor, overload protecting method for frequency converter, involves measuring temperature of coolant at source and another temperature of coolant at coolant inlet - Google Patents

Abstract

The method involves measuring an actual temperature (T1) of a coolant e.g. liquid, at a power dissipation source (12) and another actual temperature (T2) of the coolant at a coolant inlet (20). A temperature difference between the actual temperatures is determined, and an actual coolant volume flow rate is determined from the temperature difference, where the actual coolant volume flow rate flows at the power dissipation source during determination of an instantaneous limit temperature (Tj).

Description

The The invention relates to a method for overload protection electrical Components, in particular for liquid-cooled electrical Components.

It is known that electrical components, such as power semiconductors in frequency converters, cooled Need to become. This must be during the the dissipation of heat dissipated by the operation of the electrical components. It is known, the electrical components on, on or to arrange in heat sinks. For discharging the power loss heat be the heat sink with a Coolant, For example, air or a fluid (water), applied.

It is known to actively influence the cooling by an actual temperature is measured in the vicinity of the electric power loss generating heat generating element and in dependence of this actual temperature affects a coolant flow, for example, is increased. For example, this is off DE 100 48 704 A1 a method for a device with power semiconductors, in which a temperature of at least one component of the device and / or the environment and / or the supply air is determined directly or indirectly and in which actions are started when the temperature exceeds a predefinable or predetermined threshold, and in which a temperature model is used which takes into account at least the temperature of the environment of the heat sink and / or supply air for the heat sink, an operating parameter and a device parameter and calculates temperature differences and / or temperatures. Calculated setpoint temperatures are compared with measured actual temperatures. In this case, it is disadvantageous that the calculated setpoint temperatures are determined exclusively from parameters resulting during construction, production and / or assembly of the electrical components. Thus, only static output parameters are taken into account.

Of the Invention is based on the object, a method of the generic type by means of protection in a simple and efficient manner possible of electrical components is.

According to the invention this Task by a method having the features mentioned in claim 1 solved. Characterized in that an actual temperature at least one power loss source is measured, an actual temperature of a coolant at a coolant inlet is measured, a temperature difference between the at least an actual temperature at the power loss source and the actual temperature at the coolant inlet is determined, from the temperature difference of the actual coolant volume flow is determined and this actual coolant volume flow in determining the current limit temperature at the power loss source flows, is advantageous possible a safe overload protection of electrical components, with the current, included in the power loss Parameters are calculated dynamically. This is almost an adaptive Adjustment of cooling the electrical components to the given operating situation possible. So results in a better management of the electrical components, as an intervention in the management, in overload of electrical Components, only in the really necessary case takes place. Opposite the known static approach this results in a operating margin, in which the components still with their current maximum power can be driven without to fear that damage is. Too early Throttling the power, as in the prior art, can thus be avoided become.

It Thus, only an overload protection reaction, For example, a warning signal, a shutdown signal, a power reduction signal or the like, generated when the actual refrigerant flow rate below of a value for the actual power loss is needed.

Further preferred embodiments of the invention will become apparent from the others, in subclaims mentioned features.

The The invention will be explained in more detail in an embodiment with reference to the accompanying drawings. It demonstrate:

1 a block diagram of a method for overload protection of electrical components;

2 in a diagram, the course of a thermal resistance as a function of a coolant volume flow and

3 in a diagram, the dependence of the thermal resistance of a heat sink of a Temperature difference.

1 shows a total with 10 designated arrangement for overload protection of an electrical device, such as a frequency converter. The order 10 includes power semiconductors not shown in detail 12 , which are arranged on a heat sink. The heat sink has a coolant channel 14 through which a coolant, such as water, can be conveyed. The coolant quantity and thus the coolant volume flow can be adjusted by a controllable displacer unit.

By means of a temperature sensor 16 an actual temperature T1 in the immediate vicinity of the power semiconductor 12 or the power semiconductor 12 measured yourself. By means of another temperature sensor 18 becomes an actual temperature T2 of a coolant at a coolant inlet 20 of the coolant channel 14 measured.

By means of a routine 22 the temperature difference dT between the actual temperature T1 and the actual temperature T2 is determined.

For a thermal resistance Rth generally applies: Rth = dT / P = f (dV / dT), where P is the power loss of the electrical component 12 is. This means that the larger the coolant volume flow dV / dt, the lower the temperature increase of the coolant, and thus also the temperature of the heat sink at the position of the electrical component generating the lost heat 12 , The temperature difference dT is thus lower. With a constant power loss P, the volume flow dV thus results in proportional temperature increases. According to the arrangement 10 the reference value for the temperature increase dT is the actual temperature T2 of the coolant at the coolant channel inlet 20 , that is without entry of a power loss P.

During the intended use of the arrangement 10 However, there are no stationary conditions. Because the power loss P fluctuates, the actual temperature T1 changes at the power semiconductor 12 and thus the temperature difference dT between the actual temperature T1 and the actual temperature T2, which is substantially constant.

Herewith This results in transient, ie short-term loss of power P and the coolant volume flow dV / dt dependent Thermal Resistance Zth.

In determining these transient thermal resistances, fixed, design-related parameters are taken into account, which are the coolant channel 14 and the power semiconductor 12 Wear the heat sink. The heatsinks are clearly characterized by measured variables which show the dependence of the thermal resistance Rth as a function of the coolant volume flow dV / dt. 2 shows an example of such a characteristic. In addition, the heat sinks are clearly characterized by measured variables which show the dependence of the thermal resistance Rth as a function of the temperature difference dT. In 3 an example of such a characteristic for a heat sink is shown.

by virtue of this known parameter of the heat sink, the known temperature difference dT of the known coolant volume flow dV / dt let yourself the instantaneous transient thermal resistance Determine Zth.

These transient thermal resistance values Zth are via an adaptation 24 a thermal model 26 fed. By means of the thermal model 26 becomes the junction hot spot temperature T _{j of} the power semiconductor 12 calculated. This temperature T _j represents the instantaneous load temperature of the power semiconductor 12 T _{j is} calculated according to the following relationship:

In this case, the correct instantaneous transient thermal resistances Zth are taken into account. The transient thermal resistance Zth is taken into account in the form of a matrix Zth, i, j, where i is the power half ladder 12 and j represents the index used for the currently available volume flow. This matrix Zth is theoretically infinitely large. Conveniently, however, three to five elements are considered for i and three elements for j. In this example, the heat sink with three different cooling volume flows would then have to be measured in order to determine the respective transient thermal resistances Zth, i.

The with the thermal model 26 calculated barrier hotspot temperature T _j is calculated as a function of the given transient thermal resistance Zth and the power loss P based on the measured instantaneous temperature difference dT _IST . This in the module 28 calculated actual value T _{j, IS} is compared in a comparator 30 with one of the power semiconductors used 12 dependent limit value T _{j, LIMIT} . If the comparison shows that T _{j, IST} ≤ the limit value T _{j, BORD} is, is via a controller 32 the order 10 with a new value for the junction hotspot temperature T _j and the power loss P driven. For this purpose, the coolant volume flow dV / dt within the arrangement 10 changed, for example, increased.

Returns the comparison 30 that the temperature T _j, Is greater than the threshold T _{j, Border} , becomes an action 34 triggered, for example, a warning signal, a throttling of the power supply to the power semiconductors 12 or a shutdown of the arrangement 10 may include.

From the above explanation, it is clear that by the method found an inherently safe behavior of the arrangement 10 given is. Because of the dependence of the calculated junction hotspot temperature T _{j, IST} is the difference between the actual value T _{j, IST} and the threshold T _{j, LIMIT} for smaller power dissipation also smaller. In order to reach the limit value T _{j, LIMIT} , the coolant _volume flow dV / dt must change significantly. If the power loss P is high, the temperature reaches T _{j, IST} the limit value T _{j, LIMIT} with a correspondingly larger coolant volume flow dV / dt at the limit volume flow, which must be at least present. If the power loss _exceeds the nominal value, the limit value T _{j, LIMIT is reached immediately} by the actual value T _{j, IST} . Thus, the action becomes 34 only triggered if the coolant volume flow dV / dt is below the value required for the currently available (switched-on) power loss P. With increasing power loss P, the action 34 increasingly triggered faster, so the arrangement 10 inherently safe.

Claims (6)

Method for overload protection of electrical components, in particular in liquid-cooled electrical components, wherein in a coolant channel, which comprises a coolant inlet and at least one coolant outlet and which affects at least one power loss source, a coolant volume flow is generated, characterized in that an actual temperature (T1) at the at least one power loss source ( 12 ), an actual temperature (T2) of a coolant at the coolant inlet ( 20 ), a temperature difference (dT) between the at least one actual temperature (T1) and the actual temperature (T2) is determined, and from the temperature difference (dT) the actual coolant volume flow (dV / dt) is determined and this is -Kühlmittelvolumenstrom (dV / dt) in the determination of the instantaneous limit temperature (T _j ) at the power loss source ( 12 ). A method according to claim 1, characterized in that the instantaneous transient thermal resistance (Zth) from the actual coolant volume flow (dV / dt) is determined and flows into the determination of a junction hotspot temperature (T _j ). Method according to one of the preceding claims, characterized in that the transient thermal resistance (Zth) in the form of a matrix is a thermal model ( 26 ) is supplied, by means of which the junction hotspot temperature (T _j ) is determined. Method according to one of the preceding claims, characterized in that the junction hotspot temperature (T _{j, IST} ) is compared with a limit temperature (T _{j, LIMIT} ). A method according to claim 4, characterized in that when the limit temperature (T _{j, LIMIT} ) through the barrier hotspot temperature (T _{j, IST} ), the actual coolant volume flow (dV / dt) is changed, so that a changed Junction hotspot temperature (T _j ). Method according to Claim 4, characterized in that when the limit temperature (T _{j, LIMIT} ) is _exceeded by the junction hotspot temperature (T _{j, IST} ), an overload protection action ( 34 ) is triggered.