JPS6121797B2 - - Google Patents

Description

[Detailed Description of the Invention] This invention uses a digital calculation device to perform torque method, rotation angle method, torque/rotation angle method, axial force management method,
The present invention relates to a bolt tightening device that allows you to freely select one of the stress point method and the load-control washer-based tightening method.

Conventionally, various methods have been known for tightening bolts or nuts, such as the torque method, rotation angle method, torque, rotation angle method, axial force management method, stress point method, and tightening method using a load control washer. The torque/rotation angle method detects the torque or torque change rate within the elastic range of the bolt, and starts from the tightening position where these values reach a predetermined value, and then calculates the predetermined rotation angle θ _s to correspond to the predetermined axial force of the bolt. This is for tightening. The rotation angle method starts from a state where there is no protruding space between the stacked objects to be fastened, and finishes tightening by tightening a predetermined rotation angle θ _s corresponding to the predetermined axial force of the bolt. It is. The torque method is
Tightening is completed by tightening to a predetermined tightening torque value corresponding to a predetermined axial force of the bolt. Further, the axial force management method is to perform tightening by calculating the torque or rotation angle corresponding to the final target axial force from the relationship between the axial force, the rotation angle, and the tightening torque.

The stress point method determines the rate of increase in the tightening torque of the electric motor with respect to the tightening rotation angle or tightening time, and stops tightening when this rate of increase falls below a certain predetermined value after the elastic region.

Furthermore, a tightening method using a load control washer that plastically deforms under a predetermined tightening load within the elastic range of the bolt is also widely known (for example, JP-A-58-56772, 57-
163073, 57-163074, etc.). This method utilizes the fact that the tightening torque remains approximately constant during plastic deformation of the load control washer, and detects the end of the plastic deformation region of the washer from fluctuations in the rate of change of the tightening torque, and then stops tightening. It is.

However, while the rotation angle method has the disadvantage that the error in the starting point directly becomes a tightening error, it has the advantage that tightening can be easily performed when strict control of the tightening force is not required. In addition, the torque method is
Because it uses the proportional relationship between torque and axial force,
Compared to the rotation angle method, it has the advantage of being able to check the torque with a torque wrench, but it ignores variations in the proportionality constant between axial force and torque due to differences in the friction coefficient of each bolt, which causes variations in axial force. This has the disadvantage of causing Furthermore, since the axial force management method takes into account the difference in the friction coefficient for each bolt, it is possible to accurately match the axial force of each bolt to the target axial force and perform accurate tightening. Since it was composed of analog circuits, it was difficult to input and maintain various constants determined by the bolt's nominal diameter, material, etc., making it difficult to improve accuracy.

In addition, the stress point method and the method using a load control washer have a problem in that malfunctions may occur when the rate of change in torque varies minutely depending on the condition of the friction surface.

As described above, each tightening method has its own advantages and disadvantages, so it is necessary to use different tightening methods depending on the required tightening accuracy of the bolt. However, since conventional bolt tightening devices are made to adapt to a single tightening method, there is an inconvenience that it is necessary to prepare a separate tightening device for different tightening methods. Ta.

This invention was made in view of these inconveniences, and allows the user to freely select the torque method, rotation angle method, torque, rotation angle method, axial force management method, load-bearing point method, and tightening method using a load control washer. It is an object of the present invention to provide a bolt tightening device that can adopt an optimal tightening method depending on tightening conditions.

In order to achieve such an object, this invention detects the tightening torque and tightening rotation angle of an electric motor as digital signals. A memory that stores calculation programs for each method of tightening using a load control washer and a selection switch that selects one of the calculation programs stored in this memory; and a selection switch that selects one of the calculation programs stored in this memory; It is equipped with a digital calculation device that performs calculations according to a program and outputs a tightening stop signal, and allows the tightening method to be selected.

The present invention will be described in detail below based on the illustrated embodiments.

FIG. 1 is a block diagram showing a first embodiment of a bolt tightening device according to the present invention. In FIG. 1, reference numeral 1 denotes an AC power source, and power from this power source 1 is supplied to a closed circuit consisting of a switch 2, a semiconductor switch 3, and a DC series-wound motor 4. 5 is a rotation angle detector that outputs an angle pulse P (△θ) for each tightening rotation angle △θ of bolts and nuts as the electric motor 4 rotates;
Reference numeral 6 denotes a torque detector, and this torque detector 6 is equipped with a strain gauge attached to a part of the electric motor 4 that receives the tightening reaction force, and converts the strain caused by the tightening reaction force into an electric signal to perform tightening. Detect torque. 7 is an AD converter that converts this electrical signal indicating the tightening torque into a digital signal T. 8 is a digital arithmetic unit, and this arithmetic unit 8 sequentially performs predetermined arithmetic operations according to a flowchart described later. 9 is a memory storing an arithmetic program for this arithmetic unit 8 to perform predetermined arithmetic operations; 10;
1 is an input/output device for setting various setting values such as threshold value Tth, data regarding bolts, and inertia rate of the rotating portion of the tightening device; 11 is a phase control circuit; and 12 is a gate pulse generation circuit. Based on the tightening stop signal S generated by the arithmetic unit 8, the phase control circuit 11 causes the gate pulse generation circuit 12 to stop generating the gate pulse G, and by opening the semiconductor switch 3, the power source of the motor 4 is changed. cut off. A power supply voltage detector 13 detects the closing of the switch 2 and outputs a calculation start signal to the calculation device 8. 14 is a selection switch. The memory 9 contains the torque method calculation program PRG.
(T), Rotation angle method program PRG (R), Torque, Rotation angle method program PRG (TR), Axial force management method program PRG (J), Force point method program PRG (Y), Load control washer A program PRG (LC) for the tightening method is stored in advance, and the arithmetic unit 8 operates according to the program selected by the selection switch 14.

FIG. 2 is a flow chart showing the overall operation of this embodiment.

First, the selection switch 14 is set at a predetermined position to select a tightening method. Next, when the switch 2 is closed, the power supply voltage detector 13 detects the closing of the switch 2 and sends a calculation start signal to the calculation device 8. The arithmetic unit 8 first determines whether the selection switch 14 has selected the torque method (switch 100), and if the torque method has been selected, sequentially reads the torque method program PRG(T) from the memory 9 and performs calculations. Thereafter, the selected tightening method is similarly determined in steps 102, 104, 106, and 108, and the calculation program corresponding to the selected tightening method is read out and calculations are performed.

The following torque method, rotation angle method, torque, rotation angle method,
Programs PRG (T) and PRG for the axial force management method, load-bearing point method, and tightening method using load control washers
(R), PRG (TR), PRG (J), PRG (Y),
PRG (LC) will be explained sequentially using a flowchart.

<Torque method> FIG. 3 is a flowchart of the torque method, and FIG. 4 is a diagram of its tightening characteristics. When the selection switch 14 selects the torque method, the arithmetic unit 8 sends a tightening operation start signal to the phase control circuit 11 at the beginning of the program PRG (T), and the phase is controlled by the gate pulse generation circuit 12. Generate gate pulse G. For this reason, the semiconductor switch 3 is fired in phase synchronized with the trigger pulse G, and a drive current begins to flow into the motor 4, so that the motor 4 begins to rotate. As the bolt is tightened, the rotation angle detector 5 outputs an angle pulse P(Δθ) at every predetermined rotation angle Δθ. Also AD converter 7
continues to output the ever-changing tightening torque T as a digital signal.

The arithmetic device 8 reads the threshold value Tth set by the input/output device 10, and reads the digitized tightening torque T for each angle pulse P (Δθ) and compares the two (step 200). New tightening torques T are sequentially read and this comparison operation is repeated until the tightening torque T exceeds the threshold value Tth. Note that the threshold value Tth is set as a torque value slightly smaller than the final target tightening torque Ts within the elastic region A shown in FIG. On the other hand, the arithmetic unit 8 has a built-in clock pulse generator and integrates the required time t after the start of tightening. When the time t required to satisfy the condition in step 200 exceeds the set time t1, this is determined in step ₂₀₂ , and a warning is issued as the bolt is rotating together with the nut (step 204), and tightening is stopped. do. Further, if the time t required for the condition to be met in step 100 is less than or equal to the predetermined time _t2 , this means that the condition is met in step 20.
6, the bolt or nut is locked due to deterioration of the thread, or the bolt or nut has already been tightened, and an alarm is issued (step 204), and the tightening is stopped. That is, steps 202, 204, and 206 detect an abnormality and generate an alarm because the time T required for the tightening torque T to reach the threshold value Tth is outside the range of t ₂ < t < t ₁ . , the tightening stop signal S is sent to the phase control circuit 1
1, the semiconductor switch 3 is opened, and the electric motor 4 is turned on.
This is an abnormality detection step that cuts off the power supply.

In this embodiment, in order to accurately manage the tightening amount, the additional tightening amount due to inertia after the power is cut off is corrected. For this reason, the principle of this inertial correction will first be explained here. Now, if we omit friction, the equation of motion during tightening becomes as follows.

Jd ² θ/dt ² = T-Eθ ……(1) Here, J is the inertia factor as seen from the output shaft of the motor, θ
is the rotation angle after the start of tightening of the bolt, E is the spring constant of the bolt, E=dT/dθ, and an approximate value may be set in advance or a value measured during tightening may be used. T indicates the torque of the electric motor, and t indicates time.
Since T becomes zero after the power is cut off, at this time, Jd ² θ/dt ² =−Eθ (2) holds true. This equation (2) is expressed as θ=θ ₀ , d at t=0
If we solve under the initial condition of θ/dt=ω ₀ , we get becomes. Here β ² = E/J, φ = tan ^-1 (θ ₀
β/ω ₀ ). From this equation (3), the maximum value θm of the rotation angle θ after the power is cut off is becomes.

On the other hand, the maximum tightening torque Tm of the bolt is the rotation angle θ
Since it is at the maximum value θm of , it is as follows.

Tm=Eθm=√ ² ₀ ² + ₀ ² ...(5) Here, if the speed fluctuation rate of the motor just before the power is cut off is small, the tightening torque just before the power is cut off T ₀
Since T ₀ = Eθ ₀ , equation (5) becomes as follows.

T _n =√ ₀ ² + ₀ ² ...(6) As is clear from this equation (6), if the tightening torque T ₀ just before the power is cut off and the angular velocity ω ₀ at that time are known, the maximum tightening torque It is possible to predict T _n , that is, the final tightening torque Te in consideration of the amount of tightening due to inertia. This Te is the final predicted tightening torque, and the power is cut off when this tightening torque Te, which is immediately calculated from this tightening torque Te, reaches the final target tightening torque Ts, which is preset according to the type of bolt. Then, the final target tightening torque will always be achieved.
It becomes possible to tighten accurately with Ts. Also (4)
The rotation angle θ can be similarly predicted from the formula.

When the tightening torque T exceeds the threshold value Tth (step 200), the arithmetic unit 8 temporarily stores the tightening torque T at that time as the initial value _T0 in the above equation (6), and also uses the rotation angle detector at that time. The angular velocity ω ₀ is calculated from the time interval of the 5 angular pulses P (Δθ) and is temporarily stored as the initial value ω ₀ of the angular velocity ω (step 208). In other words, the angular velocity ω is inversely proportional to the time interval of the angular pulses P (△θ), and this time interval is proportional to the number N of clock pulses accumulated within this time interval, so the angular velocity ω is inversely proportional to the number N of clock pulses. It is calculated as The computing device 8 uses these initial values T ₀ and ω ₀ to compute the equation (6) above to calculate the final predicted tightening torque Te (prediction step 210). Then, a new tightening torque T ₀ is read every angular pulse P (△θ) until this final predicted tightening torque Te exceeds the final target tightening torque Ts, and at the same time the angular velocity ω ₀
is calculated and the operations in steps 208 and 210 are repeated (stop determination step 212). The time t required until the condition of step 212 is satisfied is the same as that of steps 214 and 2 as in steps 202 and 206.
At step 16, the set values t ₁ ' and t ₂ ' are compared to determine whether there is an abnormality. And these steps 21
If an abnormality is detected in step 4, 216, a warning is issued (step 204), and if there is no abnormality, tightening is stopped without issuing an alarm.

In this embodiment, if the alarms in steps 202 and 214 for detecting co-rotation and the alarms in steps 206 and 216 for detecting thread abnormalities are provided separately, the type of abnormality can be immediately known. .

Also, in this embodiment, steps 208 and 210
The final predicted tightening torque Te is calculated by sequentially reading the new tightening torque T ₀ and the angular velocity ω ₀ until the condition is satisfied in step 212, so the final predicted tightening torque Te is calculated very accurately. It may be configured as follows. That is, in the embodiment shown in FIG. 5, after the tightening torque T exceeds the threshold value Tth (step 2
00) torque increase amount (T-T ₀ ) is (Ts-Te)
(stop determination step 218), the tightening is stopped, and the time t required for this condition to be met becomes the set value.
When t ₁ ' is exceeded, it is determined whether co-rotation of the step 220 has occurred or whether the bolt has entered the plastic region.

According to this embodiment, step 210, which takes a relatively long calculation time, only needs to be executed once, so it is particularly suitable for a device that performs fastening at high speed. Note that in FIG. 5, steps that are the same as those in FIG. 3 are given the same reference numerals, so the description thereof will not be repeated.

<Rotation angle method> FIG. 6 is a flowchart of the rotation angle method, and FIG. 7 is a diagram of its tightening characteristics. In the configuration shown in FIG. 1, the tightening rotation angle θ is detected by calculating the angle pulse P (Δθ) output by the angle detector 5.

The arithmetic unit 8 first determines that the tightening has progressed normally and the bolt has entered the elastic range A (Fig. 6,
Step 300). In other words, since the rotation angle θ from the starting point to the elastic region A is almost determined by the bolt and the tightened object, the set value θ ₁ that enters the elastic region A is read from the input/output device 10, and the rotation angle θ is set to this set value. This comparison operation is repeated until the rotation angle θ reaches _{the set value θ 1 .} Step 302,
304, 306 and 324, 326 are abnormality detection steps similar to those in FIGS. 3 and 5 above.

The arithmetic unit 8 then determines in a threshold value determination step 308 that the rotation angle θ exceeds the threshold value Tth. This threshold value Tth is preset as an angle smaller than the final target tightening angle θs, and is input to the input/output device 10.

When the rotation angle θ reaches the threshold value θth, the rotation angle θ at that time is read as the initial value θ ₀ in the above equation (4), and the time of the angle pulse P(θ) of the rotation angle detector 5 at that time is read. The angular velocity ω ₀ is calculated from the interval (step 314), that is, the angular velocity ω
₀ is inversely proportional to the time interval of the angular pulse P(θ),
Since this time interval is proportional to the number N of clock pulses accumulated within this time interval, the angular velocity ω ₀
is calculated as being inversely proportional to the number N of clock pulses. This angular velocity ω ₀ is compared with a preset angular velocity ω ₁ (step 316), and ω ₀
> ω ₁ , it is determined that the rotation is co-rotating, and an alarm is issued (step 318), and then a tightening stop signal S is output to stop tightening. That is, when co-rotation occurs, the rotational speed increases. If co-rotation has not occurred, the final predicted tightening angle θf is calculated in prediction step 320, and the calculations in steps 314, 316, and 320 are sequentially repeated until this final predicted tightening angle θf exceeds the final target tightening angle θs. I'll give it back. (Stop determination step 322).

Furthermore, in this embodiment, the final target tightening angle θf is calculated by sequentially rereading the initial values ω ₀ and θ ₀ of the angular velocity and rotational angle until the conditions at step 322 are satisfied in steps 314 to 322. The present invention may be constructed as shown in FIG. 8, although this will be very accurate. That is, in the embodiment shown in FIG. 8, after the rotation angle θ exceeds the threshold value θth (step 30
When the additional tightening amount (θ - θ ₀ ) in 8) exceeds (θs - θf) (stop determination step 330), tightening is stopped, and the time t required until this condition is satisfied is set to the set value.
If t ₁ '' is exceeded, it is determined in step 332 that there is co-rotation.

According to the embodiment shown in FIG. 8, step 320, which takes a relatively long calculation time, can be executed only once, so it is particularly suitable for a device that performs fastening at high speed.

<Torque/rotation angle method> Figure 9 is a flowchart of the torque/rotation angle method.
No. 10 is a diagram showing its tightening characteristics. Point A in FIG. 10 indicates the beginning of the elastic region, and point B indicates the end of tightening.
First, the input/output device 10 determines the set torque value Ta corresponding to point A and the rotation angle θ from point A to point B.
Various values such as s=θb−θa are input. The change in the tightening torque T with respect to the rotation angle θ changes due to the frictional resistance of the joint surface between the nut and the object to be fastened, but since the contact pressure is small before entering the elastic range, the difference in frictional resistance causes the tightening. The change in torque T is small. Therefore, if the torque that starts to increase rapidly immediately after entering the elastic range is the set torque Ta, then
The starting point of the elastic region can be detected almost accurately. Further, the rotation angle θs is determined by the type of bolt, and since the axial force of the bolt is proportional to the elongation of the bolt, and this elongation is proportional to the rotation angle θ, it is determined in accordance with the target axial force.

The calculation device 8 reads the set torque value Ta set by the input/output device 10, and reads the digitized tightening torque T for each angle pulse P (Δθ) and compares the two (torque determination step 40).
0), steps 402, 404, and 40 in which new tightening torques T are sequentially read and this comparison operation is repeated until the tightening torque T exceeds the set torque value Ta.
6 is an abnormality detection step similar to that shown in FIG. 3 and the like.

The calculation device 8 calculates the rotation angle θ=ΣΔθ based on the angle pulse P(Δθ) output by the tightening angle detector 5. The arithmetic unit 8 calculates the rotation angle θ when the condition of step 400 is satisfied as A in FIG.
The rotation angle θa of the point is stored (step 40).
8) Using this point A as a starting point, the rotation angle θx is sequentially calculated and temporarily stored (rotation angle calculation step 41).
0). That is, this rotation angle θx is θx=θ−θa
It is calculated as The arithmetic device 8 then compares this rotation angle θx with a preset rotation angle θs, and when θx<θs, it calculates the rotation angle θs again using the new rotation angle θ in step 410, and calculates the rotation angle θs using the new rotation angle θ. ＞
When θs is reached, a tightening stop signal S is output (stop determination step 412).

In this embodiment, steps 402 for abnormality detection,
Although steps 404 and 406 are provided in the torque determination step 400, similar abnormality detection steps may be provided in other steps, such as the stop determination step 412, or in either of these steps 400 and 412.

Furthermore, in this embodiment, the read torque T and rotational angle θx are directly compared with the set values Ta and θs, but in reality, the rotational speed differs due to differences in frictional resistance, so the influence of inertia is considered as described above. If the equation (4) is used for correction, even more accurate control becomes possible.

FIG. 11 is a flowchart of an embodiment in which such inertia correction is performed. In step 414, the rotation angle θ ₀ immediately before the power is cut off is stored, and the number of clock pulses accumulated between the angle pulses P (Δθ) is stored. The angular velocity ω ₀ is calculated from the reciprocal of N and stored. Then, in step 416, the above equation (4) is calculated to calculate the final rotation angle θe.

In the embodiments shown in Figs. 9 to 11 above, point A is detected because the torque T exceeds the set torque value Ta (torque determination step 400), but this point A is determined by the increase rate △T of the torque T. The configuration may be such that the determination is made based on the information. For example, point A can be determined based on the fact that this increase rate △T exceeds a pre-stored set value, or the change rate of this increase rate △T, that is, (△Tn−△T _o-1 ), can be determined in advance. If the configuration is such that point A is determined based on the fact that the value exceeds a set value, errors due to differences in frictional resistance can be eliminated, and the initial stage of the elastic region can always be accurately determined.

FIG. 12 is a flowchart showing an embodiment in the latter case, in which the arithmetic unit 8 calculates each torque Tn, T _o-1 read by successive angle pulses P (Δθ).
The difference △Tn, △Tn=Tn-T _o-1 is calculated and temporarily stored (step 418), and the difference between successive differences △Tn, △T _o-1 , that is, the rate of change △ of the increase rate △T Calculate Tn−△T _o-1 and set value α
(torque determination step 420), and since this rate of change is greater than or equal to α, point A is determined.

<Axial Force Management Method> Several principles of the axial force management method can be considered, but here two principles and examples based on them will be explained.

First, the first principle will be explained. FIG. 13 is a diagram showing the tightening characteristics, and in this figure, the horizontal axis indicates the tightening angle θ, the upward axis of the vertical axis indicates the tightening torque T, and the downward axis of the vertical axis indicates the axial force N. The axial force N of the bolt is proportional to the elongation of the bolt within the elastic range, and this elongation is proportional to the tightening angle θ. Assume that the point where this proportional relationship is extended and the axial force N becomes 0 is θ=0. If this hypothetical point θ=0 is taken as the origin, the axial force N and the tightening angle θ
As shown in the figure, the relationship is a straight line passing through the origin, and if the spring constant is the same, that is, if the bolts are of the same type, the axial force N and the tightening angle θ will always change on this straight line. Therefore, if a predetermined final target axial force _N3 is applied by the bolt, the final target tightening angle θs is also uniquely determined. However, it is difficult to measure the axial force N during tightening, so the above-mentioned θ
It is also difficult to determine the origin of =0.

On the other hand, the relationship between the tightening angle θ and the tightening torque T changes depending on the friction coefficient of the contact portion between the bolt/nut and the fastened material. In FIG. 13, characteristic A indicates a case where the coefficient of friction is small, and characteristic B indicates a case where the coefficient of friction is large. Therefore, even if the final target tightening angle θs is uniquely determined, the final target tightening torque Ts for this tightening angle θs is different between characteristics A and B. However, within the elastic range where characteristics A and B are straight lines, these straight lines pass through the origin, so T=ΔT/Δθ×θ. Therefore, the final target tightening torque Ts is: For characteristic A, Ts(A) = (△T/△θ) _A ×θs For characteristic B, Ts(B) = (△T/△θ) _B It can be calculated as ×θs.

As mentioned above, by multiplying the amount of change in tightening torque T with respect to tightening angle △θ=θ _{o -θ o-1 by} the final target tightening angle θs, the final If the target tightening torque Ts is calculated and the tightening is stopped when the tightening torque T reaches this torque Ts, it is possible to accurately tighten with the final target axial force Ns. Note that θ
Since s/Δθ is a constant, it may be calculated as Ts=K(T _o −T _o-1 ), K=θs/Δθ (7).

14 is a flowchart of an embodiment based on the first principle. The calculation device 8 reads the threshold value Tth set by the input/output device 10, while reading the tightening torque T for each angle pulse P (△θ), compares the two, and sequentially applies new tightening torque until T>Tth. The applied torque T is read and this comparison operation is repeated (step 500). Steps 502, 504, 50
Reference numeral 6 denotes an abnormality detection step similar to each of the embodiments described above.

Based on the angle pulse P(△θ) output by the tightening angle detector 5 for each set tightening angle △θ, the calculation device 8 calculates
The tightening torques Tn and T _{o -1} for a certain tightening angle θ _o and the previous tightening angle θ o-1 are temporarily stored while being sequentially rewritten. Then, when the condition in step 500 is satisfied, the above equation (7) is calculated (step 50).
8), Final target tightening torque corresponding to target axial force Ns
Calculate Ts and store this tightening torque Ts. If the electric motor 4 is stopped when the tightening torque Tn during tightening, which increases as the electric motor 4 rotates, matches the final target tightening torque Ts calculated in step 508, tightening can be performed at the target axial force Ns. Become. However, in this embodiment, the final tightening torque Te is predicted when the rotation has completely stopped, taking into account the amount of additional tightening due to inertia immediately after the motor power is cut off, and this final tightening torque Te is equal to the final target tightening torque Ts. When the limit is exceeded, the power is cut off. That is, in step 512, the final tightening torque Te is predicted from equation (6), taking into account the influence of inertia when the power is cut off at time θn (prediction step). In step 514, it is determined whether or not the predicted final tightening torque Te exceeds the final target tightening torque Ts, and the angle pulse P(△
θ), the angular velocity _ω0 is newly determined from the tightening torque Tn and rotational speed Nn at that point (step 51).
0), these are set as new initial values in step 51.
2,514 calculations are repeated (stop determination step). The time t required until the condition in step 514 is met is compared with set values t ₁ ' and t ₂ ' in steps 516 and 518, similarly to steps 502 and 506, to determine whether or not there is an abnormality. If an abnormality is detected in steps 516 and 518, an alarm is issued (step 504), and if there is no abnormality, tightening is stopped. The constants such as the inertia factor J and the spring constant E required for the calculation in the prediction step 512 are as follows.
It is input in advance by the input/output device 10 or stored in the memory 9.

Next, the second principle of the axial force management method will be explained. 1st
FIG. 5 is a tightening characteristic diagram, in which the horizontal axis indicates the tightening angle θ, the upward axis of the vertical axis indicates the tightening torque T, and the downward axis of the vertical axis indicates the axial force N. The axial force N of the bolt is proportional to the elongation of the bolt within the elastic range, and this elongation is proportional to the tightening angle θ, so by extending this proportional relationship, the axial force N
Assume that the point where is 0 is θ=0. If this hypothetical point θ=0 is set as the origin, the axial force N changes on a straight line passing through the origin as shown in the figure with respect to the tightening angle θ. Therefore, for bolts of the same type, if a predetermined axial force, that is, a final target axial force Ns is applied, then the final target tightening angle θs will also be uniquely determined. However, it is difficult to measure the axial force N during tightening, and therefore it is also difficult to determine the origin of θ=0 described above.

On the other hand, the tightening torque T changes depending on the magnitude of the friction coefficient of the contact portion between the bolt/nut and the fastened material. In FIG. 15, characteristic A shows a case where the friction coefficient is small, characteristic B shows a case where it is medium, and characteristic C shows a case where it is large. Therefore, the tightening torque at the final target tightening angle θs, that is, the final target tightening torque Ts changes as Ts(A), Ts(B), and Ts(C) depending on the characteristics A, B, and C.

Now, let us assume that the tightening angle θ at a certain tightening torque in the elastic range, that is, the threshold value Tth, is the threshold tightening angle θt, and for example, this threshold tightening angle for characteristic A is θ.
t(A), it is clear from FIG. 15 that θs/θt(A)=Ts(A)/Tth holds true. The tightening angle from this threshold tightening angle θt(A) to the final target tightening angle θs,
In other words, in the case of characteristic A, the additional tightening angle θx is θx(A) = θs - θt(A), so T _s (A) = θ _s / θ _s - θ _x (A)・Tth holds true. do. By generalizing this equation, the following equation can be obtained.

T _s = θ _s / θs - θ _x・Tth ...(8) Figure 16 is a diagram showing this equation (8), and as is clear from this figure, tightening is performed starting from the threshold tightening angle θt. As the tightening progresses, the additional tightening angle θ _x = θ−
θ _t increases, and the target tightening torque T _s in equation (8) for this additional tightening angle θ _x during tightening also increases. When this target tightening torque T _s becomes equal to the actual tightening torque T, that is, the curve of equation (8) changes to the characteristic A,
If tightening is stopped when tightening is stopped when B and C intersect with straight lines A', B', and C' which are parallelly moved by threshold tightening angles θ _t (A), θ _t (B), and θ _t (C), the final target tightening can be achieved. It can be tightened with an angle θ _s, that is, with an axial force N _s .

FIG. 17 is a flowchart of an embodiment based on this second principle. In this figure, step 508 in FIG. 14 has been changed only to step 520 for calculating the additional tightening angle _θ Therefore, the explanation will be omitted.

<Yield Point Method> Next, the arithmetic program PRG(Y) of the stress point method will be explained. FIG. 18 is a flowchart of the load-bearing point method, and FIG. 19 is a diagram showing its tightening characteristics. This embodiment is suitable for tightening steel bolts made of mild steel or medium carbon steel. In other words, in the case of steel bolts, as shown in Fig. 19, a yield point Y clearly appears on the tightening characteristic curve representing the tightening torque T against the tightening rotation angle θ, and around this point the slope of this curve, that is, the torque Predetermined rotation angle △θ of T
The rate of increase △T becomes rapid. This example is
A sudden decrease in this rate of increase ΔT is detected and the tightening is stopped.

When the selection switch 14 selects the load-bearing point method, the arithmetic unit 8 executes the program PRG(Y).
At the beginning of , a tightening operation start signal is sent to the phase control circuit 11, and the gate pulse generation circuit 12 generates a phase-controlled gate pulse G. For this reason, the semiconductor switch 3 is fired in phase synchronized with the trigger pulse G, and a drive current begins to flow into the motor 4, so that the motor 4 begins to rotate. As the bolt is tightened, the rotation angle detector 5 outputs an angle pulse P(Δθ) at every predetermined rotation angle Δθ. Further, the AD converter 7 continues to output the ever-changing tightening torque T as a digital signal.

The calculation device 8 reads the threshold value Tth set by the input/output device 10, and also reads the digitized tightening torque T for each angle pulse P (△θ) and compares the two (step 600) to determine the tightening torque. New tightening torques T are sequentially read and this comparison operation is repeated until the tightening torque T exceeds the threshold value Tth. Note that the threshold value Tth is set as a torque value slightly smaller than the final target tightening torque _Ts within the elastic region a shown in FIG. Note that step 6
02, 604, and 606 are abnormality detection steps.

The calculation device 8 calculates a certain rotation angle θ _o-1 and a predetermined rotation angle △
The respective tightening torques T _o-1 and _{T o at} the rotation angle θ o after θ are stored in the built-in memory (RAM), and these tightening torques T _o-1 and T _o are sequentially applied as the rotation progresses. It's being rewritten. The arithmetic unit 8 calculates the tightening torques T _o-1 and T _o for each of these predetermined rotation angles Δθ (step 608), and continues the tightening torques T o -1 and T o for each of these predetermined rotation angles Δθ (step 608), until this difference ΔT _o becomes equal to or less than the predetermined value ΔT ₀ stored in advance. This operation is repeated (step 610). The time t required until the condition of step 610 is satisfied is the same as that of step 61
At 2,614, the set values t ₁ ′ and t ₂ ′ are compared to determine whether or not there is a tightening abnormality. Note that the difference ΔT _o is also the rate of increase of the tightening torque T with respect to the rotation angle θ, and if this difference ΔT _o satisfies the conditions of step 610 and no tightening abnormality is detected in steps 612 and 614, Furthermore, it is determined whether the tightening torque T _o at this time is within the range of pre-stored set values T ₁ and T ₂ (step 616), and
If it is within T ₁ to _{T 2} , it is assumed that the tightening is progressing normally, and a tightening stop signal S is output, the power to the electric motor 4 is cut off, and the tightening is stopped. If a tightening abnormality is detected in steps 612, 614, and 616, an alarm is issued (step 604), and then a tightening stop signal S is output to stop tightening.

FIG. 20 is a flow chart of another embodiment of the stress point method, and FIG. 21 is a diagram of its tightening characteristics. This embodiment is suitable for tightening alloy bolts made of alloy steel or non-ferrous alloys. In other words, for alloy bolts, as shown in Figure 21, the rate of increase in the tightening torque characteristic curve gradually decreases from the latter half of the elastic region, and as in the case of the steel ball (Figure 19), the yield point Y is clear. does not appear in Therefore, in this embodiment, the stress point B is detected using the maximum rate of increase ΔTmax within the elastic region.

That is, in FIG. 20, the predetermined rotation angle Δθ
The difference △ _To , that is _, the rate of increase in the tightening torque between _the (step 654),
This maximum increase rate △Tmax is the newly calculated difference △
Can be exchanged with T _o . Steps 656, 658
is an abnormality detection step similar to steps 602 and 604 described above. As a result, the maximum increase rate ΔTmax corresponding to the slope of the maximum slope line M in FIG. 21 is stored in the memory (RAM) built in the arithmetic unit 8. When the condition of step 652 is satisfied and the difference △T _o starts to decrease, a new difference △T _n is found again (step 660), and this difference △T _n is the maximum value △Tmax and the constant C (however, 0<C< These steps 660 and 662 are repeated until the value becomes less than a predetermined value C.ΔTmax (stop determination step 662). Note that in FIG. 20, steps that are the same as those in FIG. 18 are given the same reference numerals, so the description thereof will not be repeated.

<Load Control Washer> Next, a tightening method using a load control washer will be explained.
Figure 22 is a diagram of the tightening characteristics according to this tightening method.
Various methods are possible for the operation up to detecting the deformation start point C (hereinafter referred to as the first half operation) and the operation for detecting the predetermined tightening amount after the deformation start point C (hereinafter referred to as the second half operation). The various operating methods before and after the deformation start point C can be combined in various ways to configure various embodiments.

First of all, the first half operation can be configured using exactly the same calculation program as the stress point method. That is, the deformation starting point C can be detected based on the flowchart of FIG. 18 or FIG. 20 in the same way as the points Y and B of FIGS. 19 and 21. Therefore, the description of the first half of the operation will not be repeated, and the second half of the operation will be explained below based on flowcharts of various embodiments shown in FIGS. 23 to 27.

FIG. 23 is a flowchart showing an embodiment of this second half operation. In the embodiment of FIG. 23, the time t at the deformation start point C is stored as ta (step 700), and the calculation is started from this time ta. When the time t=t-ta (step 702) becomes equal to or greater than the pre-stored time ts, it is regarded as the deformation end point D (stop determination step 704) and the tightening is stopped.

The embodiment shown in FIG. 23 has a very simple configuration because it can utilize the timer built into the arithmetic unit 8, but the embodiment shown in FIG.
When combined with the embodiments shown in Figures 1 to 31, detection of the deformation end point D tends to be inaccurate. However, when combined with the embodiment of FIG. 32 in which the tightening speed is controlled constant, the tightening can be stopped exactly at the deformation end point D, which is preferable.

In the second half operation example of FIG. 24, the rotation angle θ of the deformation start point C is stored as θ _a (step 750),
Rotation angle θ ₀ starting from this rotation angle θ _a
= θ - θ _a is compared with the previously stored angle θ _s (step 756), and if θ ₀ >θ _s , the tightening is stopped.

In addition, in FIGS. 23 and 24, steps 706,
708 and 710 are steps for abnormality detection similar to the previous embodiment.

In the embodiment shown in FIG. 25, the increase rate ΔT of the tightening torque T (step 8
00) has exceeded a pre-stored set value ΔT ₀ ', the deformation end point D is determined (stop determination step 802).

In the embodiment shown in FIG. 26, the minimum increase rate △ between the plastic deformation ranges C and D is obtained using a procedure similar to that shown in FIG. 20.
Memorize Tmin (steps 850, 852, 85)
4) When the increase rate △T _n (step 856) after the increase rate △T _o starts increasing after the deformation end point D reaches the predetermined value C・△Tmin (however, C > 1) (stop determination step 858) This outputs a tightening stop signal to stop tightening.

In the embodiment shown in FIG. 27, first, the angular velocity ω is calculated from the reciprocal of the number N of clock pulses within a predetermined rotation angle Δθ (step 900). If the tightening speed is not controlled constant, the angular velocity ω will be approximately constant between the plastic deformation ranges C and D, and after the deformation end point D, the angular velocity ω will decrease as the tightening torque increases. The structure is such that the deformation end point D is detected (stop determination step 902).

By appropriately combining the embodiments of the second half operation shown in Figs. 23 to 27 with the embodiments of each of the first half operations described above, an optimal tightening method can be constructed for different bolt types and tightening conditions. I can do it.

In the examples of each tightening method explained in detail above,
As shown in Figure 1, the tightening torque T was determined using a strain gauge and then AD converted, and the rotation angle θ was determined by integrating the angle pulse P (Δθ). , rotation angle θ
can be determined by various methods, and the various methods will be explained below.

FIGS. 28 and 29 are a block diagram and a flowchart showing one embodiment. In this embodiment, the tightening torque T is determined from the motor current, while the tightening angle θ is determined from the angle pulse P (Δθ) outputted by the tightening angle detector 5 at every predetermined tightening angle Δθ. It is.

In FIG. 28, 20 is a current detection resistor connected in series to the motor 4, 22 is an AD converter that detects the motor current from the voltage across this resistor 20 and converts it into a digital signal _ID , and 24 is this AD converter. A second digital arithmetic device 26 calculates the tightening torque T and the angular velocity ω at an arbitrary rotation angle θ _o based on the digital signal _ID . Reference numeral 26 is a memory that stores the arithmetic program of this second arithmetic device. In this figure, the same parts as in FIG. 1 are given the same reference numerals, so the description thereof will not be repeated.

Next, the operation of this embodiment will be explained based on FIG. 29. When the switch 2 is closed, the arithmetic unit 8 ignites the semiconductor switch 3 via the phase control circuit 11 and the gate pulse generation circuit 12, and starts the electric motor 4. The AD converter 22 quantizes the voltage across the resistor 20, which indicates the motor current, at a cycle much shorter than that of the AC power supply 1, and converts it into a digital signal I _D. The second arithmetic unit 24 integrates this digital signal _ID over a half cycle or one cycle of the AC power source 1 to calculate the effective value of the current I _M = _ΣID , while
By calculating the angular velocity ω from the reciprocal 1/N of the number of clock pulses N accumulated within the time interval of the angle pulse P(Δθ), and further adding a predetermined rotation angle Δθ for each angle pulse P(Δθ), The rotation angle θ is calculated (step 950). In addition, this second arithmetic unit 24 also generates two consecutive angular pulses P(△
The angular acceleration dω/dt is calculated from the difference in angular velocity at the rotation angles θ _{o -1} and θ _o at which the rotation angle θ o is output (step 952).

On the other hand, in the case of a DC motor, its output torque T is T= _kφIM . Here, φ is magnetic flux and k is a constant. However, the torque T that actually contributes to tightening with an electric tightening machine is affected by the inertia during speed fluctuation (J
dω/dt) and friction torque (T _f ), the following equation is obtained.

T=kφI _M −Jdω/dt−T _f (9) The second calculation device 24 calculates the actual tightening torque T by calculating the equation (9) (step 95).
4) Output together with the rotation angle θ.

FIGS. 30 and 31 are a block diagram and a flowchart of another embodiment. In this embodiment, the tightening torque is determined from the motor current, and the rotation angle θ is calculated digitally based on the motor current and voltage, and the speed characteristic equation of the motor.

In FIG. 30, 30 is an AD converter that converts the motor voltage into a digital signal V _D , 32 is a second digital arithmetic device, and 34 is a memory that stores an arithmetic program for this arithmetic device 32. In this figure, the same parts as those in FIGS. 1 and 28 are given the same reference numerals, so the description thereof will not be repeated.

In this embodiment, when a current flows through the motor 4 due to the closing of the switch 2, the motor current and voltage are converted into digital signals I _D and V _D which are quantized by AD converters 22 and 30, respectively, at a cycle much shorter than the AC power cycle. is converted to The second arithmetic unit 32 sequentially reads these digital signals I _D and V _D and integrates them over a half cycle or one cycle of the AC power supply to calculate the effective values I _M =ΣI _D and V _M =ΣV _D ( (Step 960 of FIG. 31).

Generally, the angular velocity ω of a series-wound commutator motor is determined by the following speed characteristic equation.

ω=k·V _M −R·I _M /φ (10) where R is armature resistance, φ is magnetic flux, and k is a constant. Although the magnetic flux φ is generally a function of the current I _M , it is assumed that its changing characteristics are stored in the memory 34 in advance.

The second arithmetic device 32 includes the AD converter 22,
The current I _M output by 30 is read in sequence and the calculation of equation (10) is performed to calculate the angular velocity ω (step 9).
62), calculate the tightening torque T that actually contributes to tightening by calculating the above equation (9) (step 96).
4). The second arithmetic unit 32 outputs these angular velocity ω and tightening torque T every half cycle or every cycle of the power supply.

The arithmetic device 8 sequentially reads the angular velocity ω _o and the torque T _o output from the second arithmetic device 32 at predetermined time intervals, and calculates the tightening angle θ from the relationship θ=ω _o t (step 9).
66), output these θ and T together with the tightening torque T at each predetermined tightening angle Δθ.

In the embodiments shown in FIGS. 1 to 31 above, the angular velocity ω changes as the tightening progresses, but the present invention can also be applied to those in which the angular velocity ω is controlled to be constant. For example, there is a device that takes advantage of the fact that the armature back electromotive force is proportional to the angular velocity ω, and controls the angular velocity ω to be constant by controlling the conduction angle of the semiconductor switch 3 according to changes in this back electromotive force.

FIG. 32 is a flowchart of an embodiment in which the angular velocity ω is kept constant in this manner. In this embodiment, the motor voltage is converted into a digital signal V _D by an AD converter, and then the effective value V _M of the voltage is obtained by integrating this signal over a half cycle or one cycle of the AC power supply (step 970). The motor current I _M is calculated using the speed characteristic equation (10) (step 972), and the tightening torque T is calculated using the above equation (9) using these voltage V _M and current I _M (step 974). .
Further, in step 976, the rotation angle θ is calculated.

Furthermore, in some of the embodiments described above, when determining the tightening torque T from the motor current, the reduction in torque due to inertia and friction is taken into consideration, which makes it possible to accurately detect the torque T. It is clear that the invention can achieve its initial purpose even without these amendments.

As described above, in the first invention of the present application, it is possible to select the torque method, the rotation angle method, the torque/rotation angle method, the axial force management method, the force-bearing point method, and the tightening method using a load control washer. Therefore, from among these tightening methods, it is possible to freely select the optimum tightening method corresponding to the difference in tightening conditions of the object to be tightened, etc., and the required tightening accuracy. In other words, if you want to tighten to a target torque value, use the torque method, if you want to tighten to a target rotation angle, use the rotation angle method, and if there is a difference in the friction coefficient between the tightened body and the nut due to washers etc. If you want to tighten at a predetermined rotation angle, use the torque/rotation angle method. If you want to tighten different objects to be tightened with the same target axial force, use the axial force control method. If you want to tighten, you can choose the load-bearing point method, and if you want to tighten using a load control washer, you can choose the tightening method using a load control washer.Moreover, one device is sufficient, so it depends on the tightening method. There is no need to prepare different tightening devices. In addition, all processing up to the time the power is turned off is performed by digital signal processing, so compared to analog signals, there is no phase delay during signal processing or errors caused by the signal holding means, and tightening accuracy is significantly improved. .

Furthermore, if this device is constructed from analog circuits, separate circuits will be required for each tightening method, resulting in a significant increase in the number of parts. For this reason, productivity is poor, and adjustment of various parts, especially temperature compensation, is extremely difficult and reliability is reduced. However, according to the present invention, since it is constituted by a digital circuit, it is possible to easily adapt to various tightening methods using a program. This reduces the number of parts, making it possible to improve the productivity and reliability of the device.

In addition, if an abnormality detection step is provided in the torque determination step or stop determination step to detect tightening abnormalities from a predetermined time period until these determination conditions are satisfied, it is possible to reliably detect abnormalities such as bolt co-rotation and thread thread abnormalities. can.

Furthermore, since this invention is constructed using a digital arithmetic unit, an abnormality detection step can be provided by simply changing the program, and it is also easy to issue different warnings depending on the above types using the program. It is also simple to configure so that the type of abnormality can be easily confirmed. Further, by changing various setting values and programs, it is possible to flexibly deal with changes in the signal conditions of the bolt types, and it has the effect of being highly versatile.

[Brief explanation of the drawing]

Figure 1 is a block diagram of one embodiment of the invention; Figure 2 is a block diagram of one embodiment of the invention;
The figure is a flowchart of the overall operation, Figures 3 and 4 are flowcharts and characteristic diagrams of the torque method, and Figure 5 is a flowchart of the overall operation.
The figure is a flowchart of another example of the torque method, Figures 6 and 7 are flowcharts and characteristic diagrams of the rotation angle method, Figure 8 is a flowchart of another example of the rotation angle method, and Figure 9. and Fig. 10 are flowcharts of the torque/rotation angle method and their characteristic diagrams, Figs. 11 and 12 are flowcharts of other embodiments, and Figs. 13 and 14.
The figure shows a characteristic diagram based on the first principle of the axial force management method and a flowchart of its embodiment, Figures 15 and 16 are characteristic diagrams based on the second principle, and Figure 17 shows a flowchart of its embodiment. Figures 18, 20 and 1st
Figures 9 and 21 are flowcharts and characteristic diagrams of the load-bearing point method, Figure 22 is a characteristic diagram when a load control washer is used, Figures 23 to 27 are flowcharts of the latter half of the operation when a load control washer is used, and Figure 28 is a flowchart of the second half of the operation when a load control washer is used. 32 are block diagrams and flowcharts showing various embodiments of detecting torque and rotation angle. 4...Electric motor, 8...Digital calculation device, 9...
...Memory, 14...Selection switch, T...Tightening torque, θ...Tightening rotation angle, S...Tightening stop signal.

Claims (1)

[Scope of Claims] 1. For detecting the tightening torque and tightening rotation angle of an electric motor as digital signals, the torque method, rotation angle method, torque, rotation angle method, axial force management method,
A memory that stores calculation programs for the load-bearing point method and the tightening method using load control washers, a selection switch that selects one of the calculation programs stored in this memory, and a tightening method selected by the selection switch. A bolt tightening device with selectable tightening method, characterized in that it is equipped with a digital calculation device that performs calculations according to a calculation program of the method and outputs a tightening stop signal, and is characterized in that the tightening method can be selected. 2. Claims in which the tightening torque is detected by a strain gauge that measures the tightening reaction force of the electric motor, and the tightening rotation angle is detected by calculating the angle pulse output by a rotation angle detector for each set rotation angle. The tightening method selection type bolt tightening device according to item 1. 3. The tightening method selectable bolt according to claim 1, in which the tightening torque is determined from the motor current, and the tightening rotation angle is determined by calculating the angle pulse output by a rotation angle detector for each set rotation angle. Tightening device. 4. The tightening method selection formula according to claim 1, in which the tightening torque is determined from the motor current, and the tightening rotation angle is calculated digitally using the motor current and motor voltage using the speed characteristic equation of the motor. Bolt tightening device. 5 The motor is phase-controlled so that the angular velocity is constant, the tightening torque is calculated digitally from the motor voltage and speed characteristic equation, and the tightening rotation angle is calculated from the product of the angular velocity and time. The tightening method selection type bolt tightening device according to item 1.