DD290372A5 - PROCESS FOR WELDING SUBSTITUTES WITH THE RENEWAL THROUGH A LIGHT BOW MOVING INTO A MAGNETIC FIELD - Google Patents

Abstract

The invention relates to a method for press welding of parts to the heating by a magnetic field moving arc which consists in coaxially arranging the welded parts with a working air gap d between their faces, generating a magnetic field in the gap d and ending an electric arc which causes heating of the Stirnflaechen the Schweiszteile. The direct current component of the arc is measured and, once it has reached its minimum value corresponding to the achievement of the temperature of plastic deformation of the material of the welded parts by the temperature of the faces of the welded parts, a sudden increase in the arc current is achieved with subsequent swaging of the welded parts , Fig. 1 {Preszschweiszen; parts; Arc current; magnetic field; heating; Stirnflaechen; Schweiszteile; Temperature control; DC component}

Description

For this 2 pages drawings

Field of application of the invention

The present invention relates to pressure welding, and more particularly relates to a method of pressure welding of parts with heating by an arc moving in a magnetic field.

This invention can be applied in various fields of mechanical engineering in the manufacture of hollow parts with a closed contour as well as in construction or in the manufacture of pipes of cardan shafts, reaction rods, exhaust pipes and other parts as well as in boiler construction in the production of steam pipes.

Characteristic of the known state of the art

A particular problem in the welding of small pipes, where it is not possible to turn them in the welding process, for example in the installation of pipelines, represents the assurance of stable quality of the welded joints with high process performance. At present, such pipes are essentially of Hand in use

by jacketed electrodes or under protective gas welded. The quality of the obtained welded joints hangs in

significantly depending on the qualification of the welder, the material quality and the welding conditions. The change in the spatial location of the weld zone at the transition from the top to the bottom of the joint complicates the process in connection with the need for a corresponding correction of the welding parameters.

To avoid errors that are characteristic of a manual arc welding, are used in a number of cases Welding systems used with a microprocessor control. Here, the quality of the welded joints is improved

and the performance is somewhat increased, but the cost of the scooter equipment is substantially increased, the demands on the operator's qualifications increased, and certain restrictions placed on the operating conditions of the operators

Equipment (for example, in terms of temperature, dustiness and humidity) imposed. In a number of cases, cold resistance butt welding is used, where the stability of the Quality of the welded joints is ensured at a high Autorr.dtisierungsgrad the process. Such a thing Welding process can be realized under both operating and field conditions. The in the known However, welded joints produced by the process are characterized by the presence of a ridge of irregular shape

pointed lugs and beads formed on both the outer and on the inner surfaces of the pipe. In this case, the burr must be removed from the inner surface of the pipeline because it reduces the passage cross-section, whereby a resistance is opposed to the conveyed via the pipe media. Such burrs are usually removed by means of a cutting tool attached to a rod, which can complicate the process and can be realized only on straight pipe runs.

Recently, welding of small pipes has been done by welding in a magnetic field

used moving arc, which are the most promising and powerful.

It is a method of pressure welding with heating by an arc moving in a magnetic field

(DD-PS No. 129179) known, consisting in that the welding parts, in particular tubes, coaxially with a

Arranges working air gap between their faces, in which a DC magnetic field is generated, an electrical Arc ignites, which moves under the action of the DC magnetic field on the circumference of the end faces. At the Beginning of the process, the arc current is adjusted so that the faces of the weldments to the temperature

a plastic deformation are followed, followed by a jerky increase in the arc current and upsetting.

In the described method, the time of heating the end faces to the temperature of the plastic deformation

preset, without any deviations of the arc current value or the dimensions of the welded parts, such as the diameter or the wall thickness of the pipes to be welded, are taken into account. This has unfavorable effects on the stability of the quality of the welded joints. Such deviations are not taken into account even with a fixed value of the duration of the burst-like increase of the arc current. In addition, the well-known

It does not process to determine local changes of the gap caused by the non-parallelism of the faces and their Bumps are conditional. As a result, the obtained weld joint may have local defects. At the transition to welding parts of other types sizes, corresponding changes are made to welding parameters

which are usually adjusted after a series of trial welds, resulting in an additional time and

Energy expenditure leads. The known method of welding with heating by a ~. * - in a magnetic field moving arc

So can not guarantee the stability of the quality of the welded joint.

It is also a method of pressure welding of parts with heating by moving in a magnetic field Arc (ZIS messages [DDR], No. 10,1982, pp. 1051 to 105S) known, consisting in that one welds the parts

arranges coaxially with a gap between their faces, in which an electric arc is ignited in the

Working air gap generates a magnetic field, under the action of which the electric arc on the perimeter of the end faces

moving the welding parts to the temperature of plastic deformation of the material of the welding parts, determining the time of reaching this temperature, since which a jerky increase of the arc current is realized to a value at which the temperature of the faces of the welding parts becomes high * Point reaches the material of the welded parts, and performs an upsetting of the weldments.

In the known method, the time of heating the end faces to the temperature of the plastic deformation

determined according to the amount of heat energy stored at the end faces of the welded parts, namely the tubes. In turn, the amount of heat energy stored is determined by means of an electronic computer by measuring the arc current and arc voltage and reading the heat energy amount every 0.1 sec.

It should be noted, however, that due to the measured values of the arc current and the arc voltage

determined amount of energy is practically an energy generated by the arc. Only a part of this energy will go to the

End surfaces of pipes to be welded stored, which is largely dependent on the conditions of heat exchange

the environment (ambient temperature, presence of air movement, possible contact of the pipe to be welded with the high thermal conductivity having structural elements) dependent and not always safe.

The amount of energy stored at the end faces of the welding parts can therefore not depend on the amount of energy detected

be clearly evaluated.

In addition, the duration of the surge current increase in the known method is preset, with the eventual Deviations in the degree of heating of the faces of the weldments are disregarded. When changing the type sizes of the weldments are in the electronic computer to enter data that the new Heat energy value correspond, when exceeded a jerky current increase is realized. Despite the application of a complicated computing technology in the system of control of the welding process sees the process

no possibility to measure a gap size before and does not guarantee high stability of the quality of the welded joints.

In addition, to operate the equipment on which the known method is carried out, a highly qualified personnel

while the operating conditions are limited by a certain ambient temperature range

Object of the invention

The present invention provides a method of pressure welding of parts with heating by a magnetic field moving arc which ensures a stable quality of the welded joints and enables a correction of welding parameters in accordance with changes in welding conditions and dimensions of the weldments.

Explanation of the essence of the invention

The invention has for its object to develop a method for pressure welding of parts with the heating by a moving in a magnetic field arc, in which the determination of the time at which the temperature of the end faces of the welded parts practically the temperature of a plastic deformation of the material the welded parts are equal due to the determination of a change in longitudinal dimensions of the welded parts, which makes it possible to effect a correction of welding parameters according to the changes of the welding conditions and the dimension of the welded parts.

The above object is achieved in that in the method for pressure welding of parts with the heating by a moving in a magnetic field arc, consisting in that one arranges the welds coaxially with a working air gap between their faces in which an electric arc ignited is generated in the working air gap, a magnetic field, under the action of the electric arc moves on the circumference of the faces of the welds, wherein it heats up to the temperature of a plastic deformation of the material of the welds, the time of reaching this temperature determines, since which jerky increase of the arc current is realized to a value at which the temperature of the faces of the welds reaches the melting point of the material of the welds, to perform a swaging of the welds, according to the invention the determination of the time at the temperature of the end faces of the welded parts is practically equal to the temperature of the plastic deformation of the material of the welded parts, by measuring the DC component of the electric arc, wherein for this DC component a minimum value is set at the temperature of the end faces of the Schweißte'le practically the same the temperature of the plastic deformation of the material of the welded parts is, and upon reaching the minimum value by the DC component of the electric arc, a jerky increase of the arc current is realized.

It is according to the invention that a maximum value of the DC component of the electric arc is set, at which the end faces of the welded parts are melted, and when the maximum value is reached by the DC component of the electric arc, a swaging of the welded parts is carried out. It is further according to the invention that the size of a local gap on the circumference of the end faces of the welds in a time interval from the arc ignition to the moment of reaching the minimum value determined by the DC component of the electric arc and the working air gap is exceeded by the size of the local gap of the electric arc is removed, the weld parts are brought together in the axial direction until their faces contact each other over time, which is required for their cooling to a temperature below the temperature of the plastic deformation of the material of the welded parts, after which the weld parts by the size of originally set working air gap, in which an electric arc is re-ignited.

Likewise erfindungsgomär> is that the determination of the size of the local gap on the circumference of the faces of the welds durie- h h Measurement of the AC component of the electric arc is carried out, for which a limit is set, the exceeding of the working air gap by the size of the local Gap corresponds. The inventive method for pressure welding ensures high stability of the quality of the welded joints, taking into account any deviations of the dimensions of the welded parts and the pressing parameters. This eliminates the need for complex automatic control systems based on computer technology, which reduces the operating conditions for the equipment and the qualifications of the operating personnel. At the same time, the presetting of welding parameters is facilitated, and there is no need for a large number of natural models for processing operating modes. So it is a quick conversion of the system on which the urfindungsgemäße method is performed, guaranteed at the transition to the welding of parts of different sizes. The procedure automatically corrects the operating modes when the cross-sectional dimensions of the welded parts change by 1.5 times.

Ausführungrbeisptele

The invention will be explained in more detail below on a specific embodiment with reference to exemplary embodiments and drawings. Show it

Fig. 1 shows a system schematically, on which the method for pressure welding of parts with the heating is carried out by a moving in a magnetic field arc, with a control unit and

Fig. 2: (a, b, c) timing diagrams of the change of the DC component, the electric arc current and the movement of one of the welding parts.

The method of pressure welding with the heating of parts by an electric arc moving in a magnetic field can be carried out on a known system comprising two clamping devices 1 (Figure 1), in which coaxial welding parts 2; 3 are arranged with a closed circle in cross section, for example tubes. Outside the weld parts 2; 3 are on the periphery and in the vicinity of the end faces magnet 4 (permanent or

Electromagnet) mounted. The thrust member 3 is arranged with respect to the suture holder 2 with possibility of displacement in the axial direction, which is performed by a drive 5 for reciprocating movement. Between the end faces of the welded parts 2; 3, an electrical arc is shown schematically. The system includes a welding power source 7, as which a welding rectifier can be used with a falling current voltage characteristic whose output terminals are connected to dio clamping devices 1. The system is provided with a control unit 8 whose output signal acts on the drive 5 for reciprocating movement and on the welding power source 7. The control unit 8 includes a voltage generator 9, which is connected to the end faces of the welded parts 2; 3 is guided. In the voltage generator 9, the AC and DC voltage components U and Ud of the electric arc 6 are separated. The these DC components Ud; U, associated signals reach the inputs 10; 11 of a comparison circuit 12. An output 13 of the comparison circuit 12 is connected to the input of an arc current regulator 14 and to the first input of a switch 15, at the input of an output 16 of the comparison circuit 12 is placed. The output of the switch 15 is connected to a control input 17 of the welding power source 7, with the other control input 18 of the output of the arc current regulator 14 is connected. To an output 19 of the comparison circuit 12, the third input of the Scnalters 15 and the first input of a delay circuit 20 are connected, c'eren second input is connected to the output 13 of the comparator 12. The output of the delay circuit 20 is connected to an input of a starter circuit 21 of the drive 5, whose other tingang is connected to an output 22 of the comparator 12. The output of the starter circuit 21 is connected to the control input of the drive 5 for reciprocating movement.

For a better understanding of the nature of the method according to the invention, FIG. 2 shows time diagrams of a change: a) the DC component Ud of the electric arc; b; c) the arc current I (curve A) and the displacement S (curve B) of one of the welding parts 2; 3 without or with respect to the AC voltage component U, shown. The method of prefetch welding with the heating of parts by an arc moving in a magnetic field is realized by the described device as follows:

In the tensioning devices 1 (Figure 1), the welding parts 2; 3 with a working air gap δ between the end faces coaxially clamped in which by applying a voltage from the welding power source 7 to the clamping devices 1, an electric arc δ is ignited.

It has been found experimentally that the size δ of the working air gap is in a range of 1.5 to 2.0 mm and is determined by the value of the open circuit voltage of the welding power source 7. When the upper limit is exceeded by the value δ of the working air gap, a break in the arc and a short circuit occurs when the lower limit is exceeded. The adjusted working air gap δ corresponds to L aim Welding of the welded parts 2; 3 low-carbon steels without protective gas, a value of the DC component Ud of the arc 6, which value is in the range of 24 to 30 V. At the same time as the ignition of the arc 6, a magnetic field is generated in the working air gap 6 by means of the magnets 4. Due to the interaction of the current of the arc 6 with the magnetic field, the arc 6 moves on the circumference of the end faces of the welding parts 2; 3 whereby they are heated.

It has been found experimentally that at the start of the process, the DC component U _d of the arc 6 increases with the start of the arc 6 and then begins to decrease. This decrease takes place before the end faces of the welded parts 2; 3 start to melt. On the other hand, the DC component U _{d of} the electric arc 6 is directly proportional to the length of the arc 6, the size of the working air gap δ between the end faces of the welded parts 2; 3 is determined. In the present case, the decrease of the value Ud with the reduction of the working air gap δ due to a thermal expansion of the welded parts 2; 3 together when heated by the arc 6. The type of change Ud in time is identical for different materials, is determined by the proportion of heating the welding parts 2; 3 applied heat energy and depends on the linear coefficient of thermal expansion of the material of the welded parts 2; 3 off. The decrease of the DC component U _d due to the linear expansion can reach 20% of the value U _d in the steady process (after the stabilization of the movement of the arc 6). Since the extension of the weld parts 2; 3 when they are heated and consequently the reduction of the working air gap δ depends on the heating temperature, the temperature of the end faces of the welded parts 2; 3 are judged according to the current value of the DC component Ud of the arc 6. At the voltage generator 9 a signal corresponding to the current value of the voltage at the arc 6 arrives, where this signal is separated into two signals carrying information about the AC and DC components U, U _d .

In the case of a mechanical pre-machining of the end faces of the welded parts 2; 3 which ensures a constancy of the working air gap δ on the circumference, the AC component U _{1 is} not large, and it can be neglected, while the control of the welding process after the DC component U _d can be realized. The current value of the DC voltage component Ud of the electric arc 6 reaches the input 10 of the comparison circuit 12, which generates reference signals whose values are preset and assigned to the minimum value U _dm tn and the maximum value Ud _n , * of the ballast voltage component U _d .

The values U _dm i _n and U _dmlx are determined experimentally, wherein U _dm i _n of a heating of the end faces of the weld parts 2; 3 testifies to the temperature of the plastic deformation and U _dni ax of the melting of the end faces. At the same time, the comparison circuit 12 generates other reference signals U ₁ ; U ₂ whose preset values (24 to 30 V) correspond to a range of variation of the DC component Ud of the arc 6 in which a stable arc of the arc 6 is observed without tearing or shorting, the range of the signals U ₁ ; U ₂ within limits Ud min, U _{d ma) t} (20 to 40 V) moves.

1 to 2 s after the ignition of the arc 6 (time t, Fig. 2) and after reaching a stable movement therethrough, a comparison of the current value of the DC component U _d with the reference signals U ₁ ; U ₂ started. If Ud has departed from said range, which is indicative of exceeding the allowable range (1.5 to 2.0 mm) by the size of the working air gap δ, the comparison circuit 12 (FIG. 1) outputs a control signal from the output 16 am second input of the switch 15 and from this at the control input 17 of the welding power source 7 arrives and the latter shuts off. Then one of the welding parts 2; 3 released, set the predetermined working air gap δ, the arc 6 re-ignited and built up a magnetic field.

If the current value U _d does not leave the said range (24 to 30V), the current value of the DC voltage component U _d with its minimum value U _dm becomes 2 to 7 s (time t ₂ ; FIG. 2 a) after the ignition of the electric arc 6 i _n compared. As a result of thermal expansion of the end faces of the Schwsißteile 2; 3 b>'their heating by the arc 6, the working air gap δ and, accordingly, the running-Je value of the C "nspannungskomponente Ud of the arc 6. At the moment of the addition of the value Ud _m tn by the value U _d (Ze \ point t ₃ Fig. 2), what of the heating of the end faces of the Schwcißteile 2; 3 testifies to the temperature of the plastic deformation, the comparison circuit 12 generates a control signal which arrives from the output 13 of the Vergleichsschaltun 112 at the input of the arc current controller 14. In this case, the signal from the output of the arc current regulator 14 arrives at the control input 18 of the welding current line 7, upon which a sudden increase in the current of the electric arc 6 (FIG. 2 b, curve A) is realized. Such a stoll-like increase in the current of the electric arc 6 leads to the melting of the end faces of the welded parts 2; 3, whereby the working air gap δ between them is increased.

It has been found experimentally that the layer thickness of the on the end face of each welding part 2; 3 molten metal before upsetting the weldments 2; 3 must be 0.5 to 0.8 mm. If the layer of molten metal is below 0.5 mm, the oxides are not completely removed from the end face and are not displaced from the joint during compression. If the layer thickness of the molten metal above 0.8 mm, the gradient of the temperature field increases excessively, and when upsetting it is not possible, the weld parts 2; 3 to deform a value sufficient to crush any bumps and displace liquid metal from the joint. In the first case as well as in the second case, the quality of the welded joint is low. As the working air gap δ increases, the current value of the DC component U _d increases. As soon as the voltage corresponding to the current value of the Gleiphspannungskomponen'.e Ud signal 10 reaches the maximum value U _dm , _x from the output of the comparator 12 at the second input of the starter circuit 21, a control signal a, with the drive 5 is turned on, the upsetting of the weld 2; 3 realized. When upsetting the welding part 3 shifts in the direction of the welding part 2 by a distance S ₁ (Figure 2 b, curve B), which exceeds the working air gap δ and for a cooperation of the welding parts 2; 3 (Fig. 1) until its plastic deformation required for the formation of the welded joint.

The welded parts 2; 3 are held in the compressed state in the course of 2 to 3s. During this period, the welding current source 7 switches off (time t ₅ , Fig. 2 b, curve A), whereupon the welding parts 2; 3, and the drive 5 (Fig. 1) brings the tensioning devices 1 into the initial state (time t _e , Fig. 2 b).

From what has been said above, it is obvious that the method according to the invention makes it possible to take into account the actual heating of the welded parts and, by virtue of this, to ascertain, with a high degree of accuracy, the moment of the sudden increase of the arc current and the upsetting. Thus, conditions for the consideration of possible random fluctuations of welding parameters and dimensions of the welded parts are created, which increases the stability of the quality of the welded joints.

If for welding welding parts 2; 3 are used with any local elevations or depressions, the AC component U, the arc 6 is measured, which carries information about the unevenness of the local gap δι and for a limit U, is set, exceeding the working air gap δ by the size of local gap δ; equivalent. From the voltage generator 9, the U, associated signal passes to the input 11 of the comparison circuit 12. This signal is compared with a reference signal which corresponds to the limit U ₁ , is determined experimentally and by exceeding the working air gap δ by the local size of the gap δι (in the drawing is not shown) testifies. If the reference signal is exceeded by the current value of the Wechsdkomponente U, corresponding signal, so comes from the output 19 of the comparison circuit 12 at the third input of the switch 15, a control signal in preparation for the shutdown of the welding power source 7. As soon as the signal corresponding to the current value of the DC component Ud drops to the level of the reference signal, which corresponds to U _{d min} , the comparison circuit 12 outputs a control signal from the output 13 to the first input of the switch 15 and to the second input of the delay circuit 20. In the presence of the two control signals, the switch 15 turns off the welding power source 7, and the delay circuit 20 generates after expiration of a time interval (ta to t ₇ , Fig. 2c) preset and for additional oxidation of the faces of the weldments 2; 3 (Fig. 1) is needed to avoid its welding, a control signal arrives at the first input of the starter circuit 21. Here, the starter circuit 21 turns on the drive 5, which shifts the welding part 3 in the axial direction by a value S ₂ (Figure 2c, curve B) until the end faces of the welding parts 2; 3 in the course of a time interval (t ₇ to t ₈ , Fig. 2c) touching each other, which is necessary for their cooling to a temperature below the temperature below the temperature for the plastic deformation of the material of the welded parts. As a result of such interaction, the unevenness on the end faces of the welded parts 2; 3 (Fig. 1) balanced. Then, the welding parts 2; 3 is separated by the drive 5, and a preset working air gap δ is set between its end surfaces by re-igniting an electric arc 6, and the welding process is performed in the order described above.

Monitoring the uniformity of the gap by measuring the AC component of the arc makes it possible to eliminate the dependence on the determination of the welding group, either to sort out the welding parts or to reduce the consumption of material to eliminate the rejection by balancing the faces of the parts before carrying out their welding.

For a better understanding of the nature and advantages of the invention will be given below an embodiment of the invention.

The 57 mm diameter, 5 mm wall thickness tubes to be welded made from low carbon steel were taken up by the fixtures of a welding machine with a working air gap δ = 2.0mm. In the working air gap, an electric arc was ignited and a magnetic field was generated by means of permanent magnets. At the end of 2s, when the arc was stably moved by the arc, the arc voltage was controlled. The DC component had to be within the limits of 24 to 28 V according to the pre-tuning. The value of the direct voltage component of the arc left the said range, which is an incorrect setting

-6- 290 272

of the working air gap δ, the control unit switched off the dimming rectifier.

For the AC component of the arc, a limit U £ = 6V has been set. The values of the reference signals were also preset: Ud min = 21V and Ud _mIX = 35V.

If the DC component Ud was within the limits of the set range, the faces of the parts were heated by the electric arc until the DC component dropped to 21V. The arc current was equal to 250A during the entire heating process. This was followed by a jerky arc current rise to 1000A, which is why the faces of the weldments melted, the gap δ increased and Ud grew. At a value Ud = 35V was compressed. The speed of approach of the parts was at least 1b0mm / s and the compression force was 6OkN when compressed. The average weld time was 12 seconds from the moment of arc ignition.

An artificial local enlargement of the gap by 0.8 mm caused the welding rectifier to switch off after lowering the value of the DC component to 21V, the arc extinguished, and the faces of the parts pressed against each other after 1 second had elapsed. The subsequent diverting of the parts by the value of the working air gap δ = 2.0 mm and an arc ignition caused a normal course of the welding process.

The welding method according to the invention makes it possible to automatically change the welding time when the arc current changes. Thus, with a decrease in the arc current at the stage of the end faces, the welding time was extended to 200A up to 15s. The duration of the bursty current increase was 0.2s in both cases. In this way, the process of presetting the welding parameters is facilitated, and accordingly, the required number of natural models needed for such adjustment is also reduced.

The method described ensures a high stability of the weld quality, wherein the committee does not exceed 4% of the total number of welded joints compared to 6% in the known method.

Claims (4)

1. A method for pressure welding of parts with the heating by a moving in a magnetic field arc, comprising arranging the welds coaxially with a working air gap between their faces, in which an electric arc is ignited in the working air gap, a mag generated etfeld , under the action of which the electric arc moves on the circumference of the faces of the welded parts, heating them to the temperature of plastic deformation of the material of the welded parts, determining the time of reaching this temperature, after which a jerky rise of the arc current to a value is realized, in which the temperature of the end faces of the welded parts reaches the melting point of the material of the welded parts, and performs a compression of the welded parts, characterized in that the determination of the time, bai the temperature of the end faces of the welded parts (2 3) practically the temperature of the plastic deformation of the material of the welded parts (2; 3) is the same, by measuring the DC component of the electric arc (6), wherein for ciese DC component is set a minimum value at which the temperature of the end face of the welded parts (2; 3) practically equal to the temperature of the plastic deformation of the material of the welded parts (2; 3), and upon reaching the minimum value by the DC component of the electric arc (6), a jerky increase of the arc current is realized. 2. The method according to claim 1, characterized in that a maximum value of the DC component of the electric arc (6) is adjusted, in which the end faces of the welding parts (2, 3) are melted, and upon reaching the maximum value by the DC component of the electric arc ( 6) a jam is performed on the welded parts (2, 3). 3. The method according to claim 1 or 2, characterized in that the size e ^; of a local gap on the circumference of the end faces of the welding parts (2; 3) in a time interval from the ignition of the arc (6) to the moment of reaching the minimum value by the DC component of the electric arc (6) and if the working air gap (δ) is exceeded. the electric arc (6) is extinguished by the size of the local gap, the welds (2; 3) are brought together in the axial direction until their faces contact one another over time, which cools them to a temperature below the temperature of the plastic one Deformation of the material of the welding parts (2, 3) is required, whereupon the welding parts (2, 3) are separated by the size of the originally set working air gap (δ), in which an electric arc (6) is re-ignited. 4. The method according to claim 3, characterized in that the determination of the size of the local gap on the circumference of the end faces of the welded parts (2, 3) by measuring the AC component of the electric arc (6), for which a limit value is set, the exceeding the working air gap (δ) by the size of the local gap.