JPS5816984B2 - Naraiseigiyohoushiki - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a tracing control method, and particularly to a tracing control method for correcting errors caused by friction between a stylus of a tracer head and a model.

In tracing control, the stylus of the tracer head comes into contact with the model, and the feed axis is controlled in accordance with the amount of displacement to process the workpiece.

This tracer head is usually x, y. It has a Z-axis differential transformer, and a displacement signal in each axis direction corresponding to the displacement of the stylus is obtained from each differential transformer, and the angular component in this displacement direction is calculated to distribute pulses to the feed axis.

In contour tracing, ideally the displacement of the stylus described above occurs in the direction of extension of the line segment connecting the point of contact between the stylus and the model and the center of the model, that is, in the normal direction of the model.

However, there is friction between the stylus and the model, which causes the stylus to tilt in the tangential direction as well, and the displacement direction of the stylus may lag behind the ideal state in the tracing direction, resulting in an error.

The present invention has improved the above-mentioned drawbacks, and its purpose is to perform angle indexing from the signal detected by the tracer head, and to prevent the direction of displacement of the stylus from being delayed from the ideal state with respect to the indexed angle signal. The object of this invention is to cancel the errors caused by the curvature and perform highly accurate tracing control.

Examples will be described in detail below.

FIG. 1 is a block diagram of the tracing control circuit in an actual case of the present invention, where IX, IY, and IZ are the x, y, and
Z-axis differential transformer, 2X, 2Y2z are amplifiers, 3 is a synthesis circuit, 4 is a speed component calculation circuit, 5 is an excitation circuit for the differential transformer, 6 is a switching circuit, 7 is a displacement direction indexing circuit, 71.
72 is a displacement direction indexing section having substantially the same configuration, 8 is an indexing correction circuit, 9 is a difference circuit, io, i'+ is a sum circuit, and 12 is a distribution circuit.

The differential transformers IX, IY, and IZ of the tracer head are excited by the output of the excitation circuit 5, and output a displacement signal ε + g y lε2 corresponding to the displacement of the stylus, which is amplified by amplifiers 2X, 2Y, and 2Z, respectively. and added to the combining circuit 3 and switching circuit 6.

If the displacement vector ε of the stylus of the tracer head is shown in FIG. 2, the differential transformers ix, iy.

The displacement signal IE X l' y lε2 of 1z is
If the output of the excitation circuit 5 is sinωt, then g, = a cosφcosθsin ωt −(1
)εy=εCO8φsinθs in ωt −(2)
It can be expressed as ε=εsinφsin ωt''·(3).

Note that θ is the component ε of the displacement vector ε projected on the X, Y plane.
xy is the angle from the Y axis, and φ is the angle of the displacement vector ε from the X, Y plane.

The difference circuit 9 outputs a displacement signal ε, which is a predetermined reference displacement signal ε.

The difference Δε=ε−ε. is calculated, and the deviation signal Δε is added to the speed component calculation circuit 4.

The speed component calculation circuit 4 calculates a deviation signal Δε and a reference speed signal V based on the characteristics shown in the figure.

A normal velocity signal v8 and a tangential velocity signal vT are calculated and output from the normal velocity signal v8, and are added to the distribution circuit 12.

The switching circuit 6 selectively outputs displacement signals related to the copying plane. If the copying plane is, for example, the X and Y planes, it is set so that the displacement signals ε and εa are selectively output, ε9 is added to the displacement direction indexing circuit 7.

In this case, by using the X and Y planes as the scanning planes, the displacement signal ε force becomes almost constant, and the displacement signal ε from the synthesis circuit 3 changes mainly in response to the displacement signals ε force and ε9. Become.

Further, the displacement direction indexing circuit 7 determines the displacement direction θ from the displacement signals ε and εa, outputs displacement direction component signals sin θ and eos θ, and applies them to the distribution circuit 12 .

The distribution circuit 12 outputs a normal velocity signal v8 and a tangential velocity signal vT.
and the displacement direction component signals sin θ and cos θ, v = v cos θ + V sin θ ...
...(4)XN TV, =vNsinθ+vTcosθ...
(5) Feed control signals for the X and Y axes.

W is output, and feed control in the X and Y directions is performed in the direction in which the deviation signal Δε becomes zero by outputting the feed control signal .

The above-mentioned operation is explained assuming that there is no error in the displacement direction of the stylus, and is almost the same operation as conventional tracing control.

However, in reality, as described above, due to the friction between the stylus and the model, the displacement direction of the stylus tends to be delayed in the tracing direction compared to the ideal state, thereby causing an indexing error.

For example, as shown in Fig. 3, the stylus displacement amount ε force is determined from the displacement signal ε force when the scanning plane is the X, Y plane, and the displacement direction is further determined. Then, a normal velocity signal V and a tangential velocity signal vT are obtained.

Feed control signals for the X and Y axes are output so that the feed rate V can be obtained, and as shown in Figure 4, the stylus, which should have been displaced only in the normal direction, is now displaced from the model. Displacement also occurs in the tracing direction due to friction, and the displacement vector ε
If Δθ deviates from the normal direction, Δθ with respect to the displacement direction θ
An error in determining θ will occur.

Therefore, as shown in FIG. 5, displacement signals 6. , By is a correction vector ε in the direction of 90° with respect to the displacement ε force determined from By.

By adding the vector εIJ''-, the ideal state of displacement direction θ can be determined.

In order to obtain this correction vector ε, an index correction circuit 8 is provided as shown in FIG.

This index correction circuit 8 has a deviation signal Δε, a displacement direction component 5 inches (θ+Δθ) including an error angle Δθ from the displacement direction indexing section 71 of the displacement direction indexing circuit 7, and cos (θ+Δ
θ) signal and displacement amount signal, and as described later, a correction vector ε.

An operation is performed to obtain the component ε in the X and Y directions.

picture. ε.

9. is output and added to sum circuits 10 and 11. Displacement signal ε regarding the X and Y planes obtained by the displacement of the stylus
Excluding engineering, εa, excitation signal component sinω, etc., ε 2εCOS (, θ + Δθ) ・・・・・・(
6) ε 2 εCOS (Δ+Δθ) ・・・・・・
(7), that is, if the displacement signal ε including the error angle Δθ with respect to the displacement direction θ in the ideal state is obtained, then the correction vector ε is obtained.

The X and Y direction components ε. picture. ε.

As a, ε = ε 5in(θ+Δθ)...
・・・(8) CX C ε 0ε cos (θ+Δθ) ・・・・・・(9
) ay c should be output from the index correction circuit 8, and the sum output ε force , εX from the sum circuit 10.11
1 is εX1−εχ+εcX = εcos (θ+Δθ)+εcsin(θ+Δθ)6
y1=6y+6Cy=ε5in(θ+Δθ)−ε cos(θ+Δθ).

Here, if Δθ=jan'(εC/ε), then the displacement direction indexing section 72 of the displacement direction indexing circuit 7 has (one,
(A signal that does not include the error angle Δθ of one trial will be added.

Therefore, from the displacement direction indexing section 72, the displacement direction θ in the ideal state is determined.
A signal of cos θ indicating the components in the X and Y directions will be output.

FIG. 6 shows a block diagram of the index correction circuit 8 for determining the correction vector ε as described above, and 21 indicates the deviation signal Δε.
A comparator circuit whose output is tt 111 when 0 or more, 2
2 is a detection circuit for making a range of 1 good without correction in the displacement direction as a dead zone, 23 is an AND circuit, and 23 is a constant voltage E.

25 is a switching circuit, 26 is a memory circuit consisting of a capacitor C9, resistor R1, amplifier AMP, etc., 27 is a multiplication circuit, and 28 is a displacement signal section. A polarity switching circuit 29.30 is a multiplication circuit that switches the polarity of the signal F according to the feed direction signal F.

Deviation signal Δε=ε−ε.

When Δε〉0, the output of the comparator circuit 21 is 1″,
Further, assuming that the dead zone is ±α, the output of the detection circuit 22 becomes tt 1 m when 1Δε1>α.

When the outputs of the comparison circuit 21 and the detection circuit 22 are both 1'', the output of the AND circuit 23 is 1'', and the switching circuit 24
is switched to the ground side, and the output of the detection circuit 22 is t.
At time t111, the switching circuit 25 is turned on.

As a result, the capacitor C of the memory circuit 26 becomes discharged.

When the deviation signal Δε is Δε<0 and 1Δε1>α, the output of the comparison circuit 21 is tQjj, the output of the detection circuit n is "1", and the output of the AND circuit 23 is ttOy
y, so the switching circuit 24 is at a constant voltage E.

Since the output of the detection circuit 22 is 1'', the switching circuit 25 is turned on and the capacitor C of the storage circuit 26 is at a constant voltage E.

It will be charged by

As a result, a signal corresponding to the magnitude obtained by integrating the deviation signal Δε is outputted via the amplifier AMP.

The multiplication circuit 27 multiplies the displacement signal ED from the polarity switching circuit 28 by a value corresponding to the charging voltage ec of the capacitor C of the storage circuit 26, that is, a signal corresponding to the magnitude of the integrated deviation signal Δε.

The capacitor C and resistor R of the storage circuit 26 and the gain of the amplifier AMP are selected in advance so that this multiplication output corresponds to the absolute value 1sol of the correction vector.

Therefore, assuming that the gain of amplifier AMP is 1, EDxeo
=1εcl.

In addition, the multiplication circuits 29 and 30 multiply the signals of the displacement direction components 5in (θ+Δθ) and cos (奸Δθ) from the displacement direction indexing unit 71 by the absolute value 1εcl of the correction vector, so that (8), ( 9) ε shown in the formula.

8. ε. Signals A and B are obtained.

When the scanning feed direction is reversed, the polarity of the feed direction signal F is reversed, so the polarity of the displacement amount signal applied from the displacement direction indexing section 71 is reversed by the polarity switching circuit 28 and applied to the multiplication circuit 2γ. become.

Also, only when the deviation signal Δε is Δε〈O, the capacitor C of the memory circuit 26 is charged, and a correction vector corresponding to the charging voltage and displacement signal is obtained, and the correction is performed accordingly. But this is Δε
Normal velocity V when <0.

indicates that the stylus is in the direction of biting into the model, and this is the direction in which the friction between the stylus and the model increases, so as mentioned above, the stylus D displacement direction lags behind the scanning feed direction and errors occur. This is to correct this error.

On the other hand, when Δε〉0, the stylus is controlled in the direction of moving away from the model's proboscis, which means that the friction between the stylus and the model is decreasing, and therefore the displacement direction of the stylus is in the tracing direction. The delay is negligible and there is no need to perform correction.

That is, it is sufficient to leave the switching circuit 24 switched to the ground side.

7a and 7b show the memory circuit 26 in the index correction circuit 8.
2 is an explanatory diagram of the charging voltage ec of the capacitor C and the deviation signal Δε, in which the horizontal axis indicates time t.

Assume that when 1=0, the approach of the stylus to the model is completed and the tracing activation is performed.

At the completion of the approach, Δε=−ε□ He c””
It becomes 0.

Then, the switching circuit 2 is activated by copying activation.
5 is turned on, and the switching circuit 24 is at a constant voltage E.

Since it is switched to the side, a constant voltage E is applied. The capacitor C is charged by the electrician ec at a constant voltage E as shown by the curve in the area D1.

, or the displacement amount ε of the stylus increases, and the deviation signal Δε becomes 1ε.

As time goes by, it decreases towards KOK.

As a result, in the region D1, the correction vector has a magnitude lε.

The output of the multiplication circuit 27 (ec-EDX) increases as the displacement ε of the stylus increases.

Then, the multiplication circuits 29 and 30 output the respective axial components and add them to the summation circuits 10 and 11, so the displacement direction indexing section 72 receives a signal gXIIεy1 with the error angle Δθ corrected.
is added to the displacement direction component signal cosθ from the displacement direction indexing section 72.

sin θ will be output.

Also, at the time of 1=11, eo2E1 and Δε=-5μ.

This is for the case where the dead zone of the detection circuit 22 is ±5μ, and in the region D2 after t=t, the terminal voltage ec of the capacitor C is kept almost constant, and Δε=
-5μ is maintained while copying control is performed.

In the region D2, the switching circuit 25 is turned off when Δε is within the dead zone, and at that time the capacitor C is slightly discharged via the amplifier AMP and the like.

In the vicinity of the dead zone, Δε changes slightly as shown by the dotted line in FIG. 7, thereby increasing the charging voltage e of the capacitor C.

As shown by the dotted line in FIG. 7a, the voltage changes slightly due to repeated charging and discharging.

However, these changes are extremely small.

Therefore, tracing control is performed with precision determined by the dead zone.

Accuracy can be improved by reducing the dead zone mentioned above, but there are limits due to the accuracy of the stylus, the resolution of the detection circuit, the response speed of the mechanical system, etc., and it is practically impossible to make the dead zone extremely small. Undesirable.

In addition, Fig. 7 shows a case where Δε slightly changes around -5μ, but depending on the shape of the model, it may also change around +5μ, and tracing control can be performed with an accuracy of ±5μ. will be carried out.

1 FIG. 8 shows a block diagram of the main parts of the displacement direction indexing section 71 or 72, in which 31 is a phase shifter, 32 is a rectifier circuit, 33.
34 is a division circuit, and 35 is a sum circuit.

In the case of the displacement direction indexing section 71, the displacement signal ε ε is (1
), as expressed by equation (2), the excitation signal sin
Since it includes ωL and the aforementioned error angle Δθ, εX:εCO8φcos (θ+Δθ) sin ωt
−(14εy=εcosφsin (θ+Δθ)s
inωt...1→.

The displacement signal εA is phase-shifted by 90° by the phase shifter 31, so that ερ0εeO8φ5in(θ+Δθ)CO8ω
L ...... becomes αQ, and in the sum circuit 35 (-
+ε() is added, so εCOSφ5in(
A signal of θ+Δθ+ωt) is obtained.

Since the rectifier circuit 32 removes the alternating current component 5 inches (θ + Δθ + ωt) by rectification, the dividing circuit 33.34
εe08φ is added as one input.

Since is added to the division circuit 33 as the other input, the division result is cos (θ+Δθ)sinωL,
Since ε is added to the division circuit 34 as the other input, the division result is sin (θ+Δθ) sinωt.

By adding the outputs of the divider circuits 33 and 34 to a rectifier circuit (not shown) similar to the rectifier circuit 32,
By removing , signals of displacement direction components cos (θ+Δθ) and sin (θ+Δθ) are obtained, which are added to the index correction circuit 8 described above.

In the displacement direction indexing section 72, in the above explanation, the sum circuit 1
0.11, the operation corresponding to the case where Δθ is set to zero is performed, and the X and Y components e08θ and sinθ of the displacement direction θ are
The signal will be obtained.

As mentioned above, the stylus of the tracer head needs to contact the model with a predetermined contact pressure to perform tracing, so the amount of displacement obtained by this, that is, the amount of displacement of the stylus's original position during tracing. The reference displacement amount ε.

If the displacement amount ε corresponding to the actual position of the stylus is the reference displacement amount ε.

The copying process is controlled so that ε-ε is equal to ε-ε.

Regarding the deviation signal of =Δε, as described above (when Δε〈0,
By charging the capacitor C of the memory circuit 26, a signal corresponding to a value obtained by integrating the deviation signal Δε generated in the direction in which the stylus bites into the model is output.

Constant voltage E in this case.

The integration time constant determined by the resistor R and the capacitor C is selected in consideration of the contact friction between the stylus and the model, the feed rate, and the like.

The terminal voltage of the capacitor C is then applied to the multiplication circuit 27 as an integral output.

The output of the multiplication circuit 27 is the absolute value 1ε of the correction vector as described above.

This corresponds to l, and its size changes depending on the tracing state.

That is, if Δε<0, the integral value gradually increases as described above, and if Δε=0, the integral value at that point is output as is, and if the signal ED from the displacement direction indexing circuit 7 is constant. , the absolute value 1εcl of the correction vector is proportional to the above-mentioned integral value.

This correction vector has X and Y direction components εcx l
ε.

The scanning accuracy is improved by adding the displacement signals ε and ε to the displacement signals ε and ε to perform control.

As explained above, the present invention uses an index correction circuit to prevent the displacement of the stylus from being delayed in the scanning direction due to friction between the stylus and the model, resulting in an error in the index angle. By creating and correcting a correction signal corresponding to the amount, it is possible to improve the accuracy of copying.

For example, the conventional error of approximately ±20μ can be improved to approximately ±5μ, which corresponds to the set dead zone, according to the present invention.

There is also an advantage that the scanning speed matches the set speed.

Note that a correction vector ε is used for the displacement vector detected by the tracer head.

Although it is easier to perform calculations when the angle is set to a right angle direction, the present invention is not limited to this, and various additions and changes can be made in addition to the embodiments.

[Brief explanation of drawings]

Fig. 1 is a block diagram of an embodiment of the present invention, Figs. 2 and 3 are illustrations of stylus displacement vectors, Fig. 4 is an illustration of error occurrence due to stylus delay, and Fig. 5 is an illustration of correction. An explanatory diagram, FIG. 6 is a block diagram of an example of the indexing correction circuit,
FIGS. 7a and 7b are explanatory diagrams of the capacitor terminal voltage Δε, and FIG. 8 is a block diagram of an example of the displacement direction indexing circuit. IX, IY, IZ are tracer head differential transformers, 3
4 is a speed component calculation circuit, 7 is a displacement direction indexing circuit, and 8 is an indexing correction circuit.

Claims (1)

[Claims] 1. A stylus of a tracer head is brought into contact with the model, and in addition to a displacement direction indexing circuit which is a displacement signal regarding the scanning plane from the tracer head, a displacement direction signal is outputted from the displacement direction indexing circuit. , and adds the displacement signal from the tracer head to a synthesis circuit, outputs a displacement signal ε according to the displacement of the stylus from the synthesis circuit, and outputs the displacement signal ε and a reference displacement signal ε. A deviation signal Δ( of the difference (ε−ε.) between and the displacement direction signal to the distribution circuit, the distribution circuit outputs a feed control signal for each axis, and the feed control for each axis is performed in the direction in which the deviation signal Δε becomes zero. In the copying control system, a displacement direction signal including an error angle from the displacement direction indexing circuit, a displacement signal obtained from a displacement signal regarding the copying plane, the deviation signal Δε, and a feed direction signal are used. An index correction circuit is provided, and the index correction circuit calculates, with respect to the deviation signal Δε generated in the direction in which the stylus bites into the model, a signal corresponding to a magnitude obtained by integrating the deviation signal Δε. The absolute value of the correction vector is obtained by multiplying the displacement amount signal, the absolute value is multiplied by the displacement direction signal including the error angle, each axial direction component of the correction vector is outputted, and each axial direction component is multiplied by the displacement direction signal including the error angle. and a displacement signal related to the scanning plane to correct an error angle, and add the addition result to the displacement direction indexing circuit to output a displacement direction signal with the error angle corrected. method.