JPH02120819A - Method for correcting bend of scanning line - Google Patents

Abstract

PURPOSE:To easily correct a bend of a scanning line in a degree at which no influence is given to the picture quality within an extent in which the cost is not increased by fixing a specific lens after the lens is deflected in the auxiliary scanning direction. CONSTITUTION:After the lens which has a magnification in Y-Y direction perpendicular to the scanning direction X-X of a luminous flux deflected by an optical deflector 3 and has the lens surface which is the maximum in magnification among lenses 2, 51, and 52 of the 1st and 2nd image forming systems is deflected in the auxiliary direction, the lens is fixed for correcting a bend of a scanning line. For example, a line image formed on the reflecting surface 4 of the deflector 3 by means of a cylindrical lens 2 of the 1st image forming optical system forms a point image on the image surface 7 of a medium to be scanned when the line image is emitted through of lens 51 and 52 of the 2nd image forming optical system and scanned along a straight line 8 in accordance with the rotation of the deflector 3 in the direction shown by the arrow. Therefore, a bend of a scanning line can be corrected easily in a degree at which no influence is given to the picture quality within an extent in which the cost is not increased.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a method for correcting the curvature of a scanning line in a scanning optical system applied to various printers, facsimile machines, and the like.

[Conventional technology]

a first imaging optical system that forms a linear image of the light beam emitted from the light source; an optical deflector in which a deflecting reflection surface is set in the vicinity of the imaging position by the first imaging optical system; A scanned medium is scanned by a light beam deflected by a light deflector, and the light beam is disposed on the optical path of the light beam between the scanned medium and the light deflector.

The light beam is maintained in an optically conjugate relationship between the deflection reflecting surface of the light deflector and the scanned medium in a plane perpendicular to the deflection surface which is the locus plane of the light beam continuously deflected by the light deflector. A scanning optical system is known that includes a second imaging optical system that forms an image of the image on the scanned medium.

In FIG. 8 illustrating an example of the above-mentioned scanning optical system,
The light beam emitted from the light source 1 is linearly imaged by a cylindrical lens 2 serving as a first imaging optical system. on the other hand,
One of the plurality of deflection/reflection surfaces of the optical deflector 3 is set to be located in the vicinity of the image formation position in sequence according to the rotation.

By the above, (1! LAEt imaged on the reflective surface 4
is emitted through the fO lens 51 and the fO lens 52 as the second imaging optical system, and is imaged as a point image on the image plane 7 on the scanned medium, and is directed in the direction of the arrow of the optical deflector 3. A straight line 8 is scanned according to the rotation. This scanning is called main scanning, and its direction is called main scanning direction XX. Further, scanning in a direction perpendicular to the main scanning direction XX on the image plane 7 is referred to as sub-scanning, and this direction is referred to as sub-scanning direction Y-Y.

If, for example, the fO lens in a scanning optical system is decentered, the scanning line that should normally be scanned as a straight line 8 becomes a curved scanning line as shown by reference numeral 8'.

Here, eccentricity refers to the lens axis shown in FIG. 9, and if what is shown by the broken line is the lens ring in an ideal state without eccentricity, it means that the eccentricity S deviates from the state shown by the solid line.
Eccentricity, which greatly affects the curvature of the scanning line, mainly accompanies a situation where the YY Herrens plane in the sub-scanning direction is shifted.

In FIG. 9, the symbol 0-0 indicates the optical axis of the lens.

[3 Wis to be Solved by the Invention] In conventional monochrome printers, facsimile machines, etc., scanning line surface warping occurs by about 0.2 to 0.5 a+m.

This level of value was acceptable.

However, in recent years, wide printers, high-precision plate-making plotters,
Furthermore, with the development of color printers that perform writing with multiple beams, scanning line surface warping has come to affect image quality as a positional deviation in writing.

For example, in a color printer using a scanning optical system that writes in each color, red, green,
Each writing scan line 811e, 8 contains each piece of information of the blue image.
G and 8B may be curved with respect to the ideal linear scanning tjA8, resulting in so-called bending of the scanning line.

The scanning line 8Re is in the sub-scanning direction Y with respect to the reference scanning line 8.
The amount of upward bending dl' for -4! Since the scanning line 8B has a downward bend amount of dB3 in the sub-scanning direction Y-Y, the maximum difference between the two is dn+dB, and a color shift occurs according to this difference.

Further, there is an example of a wide printer as shown in FIG. 10(b). In this example, writing of AO size width is synthesized by combining the scanning lines of the left and right scanning optical systems for two A2 size widths. Since the scanning lines 81 and 82 are curved by di and d2 in different directions, a total of d is obtained at the virtual line N-N corresponding to the joint between the two scanning lines.
This causes a displacement of l+d2 at the seam, impairing the image quality.

To resolve and eliminate such color deviations and seam deviations,
The eccentricity tolerance of the lens and the tolerance of the lens mounting surface must be extremely strict.

However, if the above-mentioned tolerances are tightened, the additive cost 1- becomes high, which is a problem because it does not meet the demand for lower costs of various :2t# equipment using this main scanning optical system.

The present invention is a method for easily correcting the bending of a scanning line to the extent that it does not affect the image quality, without strictly controlling the eccentricity tolerance of the lens, the tolerance of the mounting surface, etc., and without increasing the cost. It is an object of the present invention to provide a method for correcting the curvature of a scanning line.

Note that the bending of the scanning line is caused not only by the eccentricity of the lens but also by the tilting of the lens.

In other words, when the lens is centered around an arbitrary axis in the main scanning direction,
However, the bending in this case can be easily eliminated by a combination of eccentricity in the sub-scanning direction and eccentricity in the optical axis direction (displacement of the lens surface). Furthermore, the eccentricity in the optical axis direction does not pose a problem because the kW applied to the bending of the scanning line is small.

[Means for Solving the Problems] In order to achieve the above object, in the present invention, a magnification is 1. Among the lenses of the second imaging optical system, the lens having the lens surface with the maximum magnification is decentered in the sub-scanning direction and fixed, thereby correcting the curvature of the scanning line.

[For production]

The set position of the lens that has the greatest influence on the curvature of the scanning line is adjusted after the fact.

〔Example〕

First, as a reference for comparison, the imaging relationship when there is no eccentricity is shown in FIG.

This figure shows the scanning optical system shown in FIG. 8 above.

Cross-section passing through the optical axis of the lens in the sub-scanning direction (sub-scanning cross-section)
This is shown in .

As shown in the figure, the fθ lens is 2
What's the composition like? And the first fO lens 51
has a spherical surface on one side 51-1. The other side has a cylinder surface 51-2, and the second fO lens 52 has a cylinder surface 52-2 on one side.
1. The other side is composed of a toroidal surface 52-2.

The light beam that is imaged linearly (the linear direction is the main scanning direction Y-Y and perpendicular to the plane of the paper in FIG. 3) on the deflection and reflection surface 4 is deflected, and is also reflected on the spherical surface 51-1 and the toroidal surface. The virtual image position ff1 in space is expanded by the lens function of 52-2.
Connect a virtual image to V1. Note that the chief ray passing through the lens optical axis is indicated by the symbol CR.

This light beam is focused by the loidal surface 52-2, which generally has a strong power in the sub-scanning direction Y-Y, and is imaged at a predetermined position [8F] on the image plane 7.

When the deflection/reflection surface 4 rotates in response to the rotational drive of the optical deflector, the light image 1 on the image plane 7 passes through a predetermined position 1i8F in a direction perpendicular to the plane of the paper, that is, in the main scanning direction Y-■. However, since it is assumed here that the lens is not eccentric, the scanning line does not curve, and therefore, when projected onto the paper in Figure 3, this scanning trajectory deviates from the predetermined position 8F. do not have.

Next, regarding the fO lens 52, its cylinder surface 52-1
Consider a case where there is no eccentricity and only the I-loidal surface 52-2 is eccentric by an eccentric amount S in the sub-scanning direction YY. The imaging relationship in this case is shown in FIG. 4 together with the case where there is no eccentricity. In FIG. 4, the toroidal surface 52-2 without eccentricity is indicated by a broken line, and the toroidal surface 52-2 with eccentricity is indicated by a broken line.
2' is indicated by a solid line.

As shown in Figure 4, consider the case where there is no eccentricity.

The chief ray CR that has proceeded on the lens optical axis is a toroidal surface 5
After passing 2-2, continue going straight as shown by the broken line.

- On the other hand, considering the case where there is eccentricity, the principal ray cn passes through the toroidal surface 52-2' and then passes through a point on the lens solo axis at a focal length mffl of the toroidal surface 52-2. The lens is bent so as to proceed to a position shifted in the sub-scanning direction Y-'Y from the lens yl by an amount of eccentricity S on a plane perpendicular to the optical axis of the lens.

If the light LA after being bent only in the sub-scanning direction is called the main ray C, then on the sub-scanning section, the principal ray Bcn and the principal ray C
form an angle φ.

Figure 5 shows the imaging relationship, including other optical systems, based on the situation shown in Figure 4 above, on a main scanning cross-section that passes through the lens optical axis and is orthogonal to the sub-scanning cross-section. It is.

In Figure 5, the principal ray C is imaged from the lens optical axis at a point 8C on the paper surface 7 with an image height ratio of 0, and the principal ray C is focused on the image surface 7 near the scanning end with an image height ratio of about 1. It is shown that the image is focused on point 8R. The light at that contributes to the HD of the latter point 8R is the principal ray R.

In FIG. 5, the distance from the final surface t of the fO lens 52 to the image plane 7 is defined as LC. Depending on the optical conditions, the principal rays C1 and fiR are respectively [・
The light is emitted perpendicularly to the rhoidal surface 52-2. Here, the intersection between the toroidal surface 52-2 and the principal ray tC is indicated by a contour 52-2C, and the intersection between the toroidal surface 52-2 and the principal ray R is indicated by a contour 52-2R.

Therefore, on the main scanning plane, the main light II! The angle that AR makes with respect to the principal ray L%c is the scanning angle 0, and the principal ray C is on the same plane,
If the amount of deviation in the lens optical axis direction of the position emitted from the toroidal surface 52-2 is Δ, then the principal ray R is
The distance LR from the light emitted from the light beam 2-2 to the point 8R on the image plane 7 can be expressed as Ln+(Δ+LC)/cos O.

Next, the eccentric toroidal surface 52-2 shown in FIG.
When the imaging relationship according to `` is projected onto the sub-scanning section with respect to the principal ray C9 and the principal ray R in FIG. 5, it becomes as shown in FIG. 6(a).

Furthermore, when the main part of FIG. 6(a) is enlarged, FIG.
b) In FIG. 6(a), among the eccentric toroidal surfaces 52-2', the outline of the surface that emits the chief ray BC is expressed as 52-2'C1, and the outline of the surface that emits the chief ray R is Each is designated by the reference numeral 52-2'R.

In FIGS. 6(a) and 6(b), the principal ray C is shown in exactly the same way as in FIG. 4 because it is on the sub-scanning section, and is shown as being bent by an angle φ from the lens optical axis. here,
It can be expressed as φ=jan'' (S/f) (see FIG. 4).

On the other hand, the angle that the principal ray R makes with the main scanning plane is also the angle φ, which is the same as the principal ray uAC. This is because the chief ray always exits at right angles to the toroidal surface 52-2 in FIG. 5, and the lens powers of several surfaces remain unchanged even at an angle of 0. However, when such a principal ray R is projected onto the sub-scanning section, it appears at an angle ω that is wider than the angle φ, as shown in FIG. 7(a).

This angle ω is (1) = tan' (S/ f-c
ost). In this way, the principal light fiR appears to be bent from the lens optical axis, and the outline 52 of the lens surface
-2'R is also shifted by the shift amount Δ.

When the above-described FIGS. 4 to C are collectively expressed on a sub-scanning section, the result is as shown in FIG. 7(a). Therefore, let us consider how the scanning line curves as shown in FIG. 7(a).

In FIG. 7(a), the symbol "rR (rotation)" indicates the fifth rotation centering on the center of the arc of the t-roidal surface 52-2 in FIG.
The chief ray R in FIGS. 6 and 6 is rotated so as to return toward the lens optical axis by an angle of 0 onto the sub-scanning section.

Shows the virtual ray portion that appears when projected.

When considering the rotation of the principal light vAR mentioned above, such a principal light l
The contour 52-2'R that gives rise to aR is also represented in a rotated position in a similar manner, and this contour is designated by the symbol "52-2'R (
rotation)”. This contour r52-2'll (rotation)
'' overlaps the contour 52-2''C.

When considering the above rotation, the image formation position is a virtual image plane 9 that is shifted from the image plane 7 by (the distance LR - the distance tc). Since the principal rays C1 and tsR both form an angle φ with the main scanning section, the imaging position of the principal ray C is a position L C-t, an φ offset from the main scanning section in the sub-scanning direction Y-Y. 8'
C, and the imaging position of the principal ray R is a position 8''R that is LR-tanφ apart from the main scanning section in the sub-scanning direction YY.

The distance between the above position 8''C and the lens optical axis and the above position g
The difference in distance between ia'n and the lens optical axis is the amount of curvature d of the scanning line, and can be expressed as d=(Ln-LC) tanφ.

This scanning line surface edge ff1d is specified by determining the distances ML, n, and LC in accordance with determining the scanning angle 0.

Contour 52-2'C and contour r52-2 in FIG. 7(a)
In FIG. 7(b), which shows an enlarged view of the 'n (rotation)' portion, tanφ=S/ (f+ (r= 57:
17)).

Normally, S < r, so r - q 'r O, and tanφ in the above equation becomes tanφ gon-.

Therefore, since tanφ is proportional to the amount of eccentricity S, the amount of bending d of the scanning line is expressed as d(LR-LC) . . . (1).

By the way, in FIG. 5, the main light BR1 and the main light vIcR do not necessarily emit perpendicularly to the toroidal surface 52-2 in the general case, and the case where the main light does not emit perpendicularly is also considered.

In Figure 5, the chief ray C at contour 52-2C and contour 52-21
1 times r in the direction perpendicular to the paper surface of Figure 5 along the principal ray R
When $ is respectively β5z-z (CL βB-z(R), the distances between the positions @8'C and 8'R in Fig. 7(a) and the optical axis of the lens are 5(1-β*z-), respectively. z (C)), 5(1-β32-2
(R)), and the amount of bending d of the scanning line is d=S (β,, -, (R)-β龾-z (C))
...(2) How can I express it?

- On the other hand, in Fig. 5, for the chief ray C, the magnification is βC, (1, C''-f)/f = βC, and for the chief ray R, the magnification is βR, (Lll-f)/f = β R...(3) is considered.

So, if we solve for d from the above equation (1) and the above equation (3).

d=scβR-βC) ...(・1) Howl. Applying this formula (・1) to the toroidal surface 52-2, d=S (β5□-2(R)-β32-2 (C))
...(5), and equations (2) and (5) are equal.

Therefore, the same conditional expression can be applied to the bending id of the scanning line whether it is vertical emission or not.

Therefore, with reference to FIG. 1, the amount of scanning line image sharpness will be examined by applying the above conditional expression and taking into account the influence of each lens and the like constituting the scanning optical system.

In FIG. 1, the reference numeral VI indicates each lens surface shown in FIG. fO lens 51. fO lens 5
When each optical surface of 2 has no decentering, the virtual image position created by the optical system before the cylinder surface 52-1, and the symbol VI' similarly shows the virtual image position when there is decentering. Therefore,
Light cn'' passing through the virtual image position VI' and parallel to the lens optical axis
Similarly, the principal ray CR when the optical system before the cylinder surface 52-1 is decentered is the light passing through the virtual image position [VI'.
The position where the image is formed by the toroidal surface 52-2 on the non-eccentric 1 is the image forming position 8'. That is, the imaging position 8'
y is the extent to which the light eccentric before the toroidal surface 52-2 is imaged by the F.roidal surface 52-2, which is not eccentric.

Then, the variation in the virtual image position due to the lens surface before the toroidal surface 52-2, which is the final surface of the lens, is represented by the difference between the virtual image positions vr and vr''.

The amount is expressed by a function S'(y) of the image height ratio y, and the magnification of each optical surface can also be expressed by a function β(y) of the image height ratio y.

The relationship between the function S'(y) and the function β(y) is expressed as follows.

S'(y)=β'z(y)・S2+β°, □−□(y
)・Sl, -2+β'51-z (''/) 551-z
+β'52-1(''/)'5sz->''・(6
) The subscript in each term in the above formula (6) indicates the code of the lens surface (lens surface number), specifically as follows.

In this example, β2=0, so let m be the lens surface number,
Σβ'n+(y)=1.

From S'(y) in the above equation (6), [・Roidal 52-2
Lens optical axis and cylinder surface 52- when is not eccentric
The light flux when the optical system before 1 has eccentricity is imaged by the toroidal surface 52-2 without eccentricity.
) from the lens optical axis in the scanning direction = S”(y
), then S”(y)=-3'(y)(1-β5z-z (y))
It can be expressed as +S'(y)=S'(y) ・β alignment zz-z(y).

Considering the case where there is, S''(y) = (SS2-Z S'(y)) (
1f3gz-z (31)) +S'(y)"S, z-
z (1-β52-2 (y)) +S'(y)-β5
=-2(y) Howl.

As a result, the difference between the convergence position according to the main Ic line C and the image formation position according to the principal ray R on the image plane 7, that is, the amount of bending d of the scanning line, can be calculated from the above equation (2).

d=(β'-,!-z (R)(S''(R)-3rin2-
2)-β52-2 (C) (S'(C) 552-
23 If this equation is transformed to reflect the influence of eccentricity on each surface, a linear equation is obtained.

d =:21 S2 +KS1-11Ss14 +f
(5z-2'SS1.-2+Ksz-t 5sz-1
+Ksz-z ・55z-z Mountain (8) However, in the above formula, each value of the coefficient Km is as follows.

Furthermore, since the eccentricity of the toroidal surface 52-2 is greater than or equal to ssz-z, it can be seen that the sum of the effects of each surface is the value of the curvature of the entire scanning line.

When thinking about an actual lens, the eccentricity of each lens surface is 0.
It is obvious that the bending of the scanning line will be eliminated if this is done, but since the lens has at least two surfaces, it is impossible to make the eccentricity of one surface zero even if the adjustment is made in the sub-scanning direction. This is not possible unless the eccentricities on both sides are the same.

However, each lens is adjusted by 5N by ΔS of the following formula.

If the eccentricity of Sn z +ΔS is forced to remain, the following equation (10) can be obtained by tη.

dn=o=Kn 1・(Sn x+ΔS)+Kn 2
-(Sn Z+ΔS) −(10) However, n is the lens number.

If we transform the above equation (10) and solve for ΔS, we get the first-order (
11) Obtain the formula.

By applying this equation (11), it is possible to eliminate the bending of the scanning line for each lens. Furthermore, for all lenses, the above (11)
) can be adjusted to d=0.

Next, we will consider which of the lenses that make up the scanning optical system should be adjusted most effectively.

From the above equation (8), Kn+ represents the degree of influence of scanning line surface rounding, so considering the 1 sound rate β of each surface, each surface from the surface of the cylindrical lens 2 to the cylinder surface 52-1 has a Both, 0≦β≦1 (β2=o).

Therefore, β' in the above equation (7) also becomes 0×β'×1. On the other hand, the toroidal surface 52-2 has β<-1. Also, looking at the difference between magnification β(R) and β(C),
1β(R)-β(C)I <Iβ(R)+β(C)I
/2. Moreover, the value of 1β(R)−β(c)1 is also large when the magnification βm is large for the contraction.

From the above, it is clear from equation (9) above that Km of the surface with a large magnification of 1βall is large.

Therefore, it can be seen that most of the bending of the scanning line can be eliminated by adjustment by moving the lens including the large surface of 1β1.

Next, consider the amount of shadow iWI with respect to the bending of the scanning line when the specifications of each lens are as shown in Table 1 for the scanning optical system of the embodiment.

(Table-1) However, in Table-1, regarding the principal ray C, the deflection reflection surface 4
The rotation angle of the deflection reflection surface 4 with respect to the chief ray R is 0°, and the rotation angle of the deflection reflection surface 4 is 16.6°.

Then, the magnification of each lens surface in Table 1-1 is as follows.

βz (R) = βz (C) = 0- βn-z (C)
=0.952tβsi-L(R) =0.947
βB, -z (C) = 0.887 eβ5z
-z (R) =0.871, β5z-z (C
)=0.811゜β5t-1(R)=0.772
β5z-z CC) = -5,45゜β雛2-2
(R) = -0,604 The above magnification is the above equation (7), (
From formula 9), K2 = 0.11. K514 =0.0
3'+Kgl-t=0. 52-z = 0 to lQ+
．． 35+K52-2=Q, 59.

In this example, the coefficient is −・The magnification of the largest surface 52-2, βs
Knee z (C) = 7' 5.45 and β ~ 2-2 (
R)=-6,04 is -.largest, and the fO lens 52 including that surface 52-2 should be adjusted.

In addition, in a scanning optical system having an fO lens with an anamorphine surface, it is common to arrange a toroidal surface on the final surface of the fO lens and set the magnification of this surface to β-1. The toroidal surface 52-2 is placed as the final surface, and as described above, the magnification of the several surfaces is β<-1. Therefore, it is effective to adjust the fO lens 52 having the toroidal surface to:A.

In FIG. 1, if the function S'(y) is a value S' that does not depend on the image height ratio, the toroidal surface 52
-2 in the sub-scanning direction Y-Y (the toroidal surface after g* is indicated by the symbol 52-2 re), the image formation position on the image plane 7 will be at the predetermined position which is the original image formation position. The image is formed at a distance I (separated from the image position 8''y) from the position 8F, and no bending of the scanning line occurs. This means that by moving 52, the curvature of the scanning line can be eliminated.

In addition, the imaging position is at a distance of 1 from the predetermined position 8F, as shown in the double series
However, in this type of scanning optical system, there is generally an optical path bending mirror M between the final lens surface and the imaging position.
is provided to guide the light beam to the imaging position, so by finely adjusting the angle of this mirror M, the imaging position 8°'y can be moved to the predetermined position 8F.

Due to this adjustment of the imaging position, the imaging point is shifted by a small amount T in the direction of the lens optical axis, but the distance between the image plane 7 and the mirror M is
If the distance I from 8'y to the predetermined position 8F is longer by 21 minutes, the shift of the imaging point in the optical axis direction can be ignored.

More specifically, in the above example, 52zz = 0.1 mm, 5S2-2 = -0,0
8FlIfi, the above formula (11) and the above,
2. -, =o, 3s, t<, -, ==0.59, ΔS =0.34mm is found.

Therefore, if the fO lens 52 is shifted in the sub-scanning direction by 0.3=1nrQ from the ideal position assuming no decentering, it can be set to dsZξ0.

As a result, the curve of the scanning line is reduced by the cylindrical lens 2.
, which is caused by the influence of the fO lens 51, and the amount thereof is almost negligible, so the problem of scanning line curvature can be solved in one minute just by adjusting the fO lens 52.

From the above, it can be seen that even if the eccentricity tolerance of the lens alone and the eccentricity tolerance of the mounting surface are made loose, it is possible to reduce the amount of bending of the scanning line by lH1.

Next, an example of a specific adjustment method will be introduced with reference to FIG.

In FIG. 2(a), reference numeral 520 indicates an ideal fO lens without eccentricity.

The outer diameter part from the main diameter line (generating line) of this fOl men's.

In other words, the dimension to the mounting surface is ■I. Then, with this main diameter line aligned with the ideal optical axis 0-0, the housing 4
It is assumed that the fO lens 520 is attached on the lens 00. Such a state is assumed to be a reference state.

Now, in order to carry out the present invention, a housing 1ooo
The upper surface of the housing 400 is set to be 2 degrees lower than that of the housing 400. This dimension is determined as follows.

That is, when the maximum adjustment range obtained from the maximum tolerance of the eccentricity of both surfaces of each lens constituting the scanning optical system is ±ΔSMAX, 1'f which satisfies h≧1ΔSMAX+ is an arbitrary value. In this way, any adjustments can be made within the range of the dimensions and thickness of the spacer.

Further, as shown in FIG. 2(b), a specific fO lens 52 to be set on this housing 4000
Regarding 0A, the eccentricity of both surfaces of the lens is Sa.

sb and 5 at that time! If the fixed amount is set in the upward direction as ΔS (however, ΔS<151sxAXl), the thickness of the spacer 50 can be found as 5p=h+ΔS, and by fixing the lens with a spacer of this thickness, Curved scanning lines can be corrected.

The example in FIG. 2(C) concerns the fO lens 520I'3.

The adjustment amount decreases by ΔS' (however, ΔS' decreases by 1ΔSMA)
In this case, the thickness of the pacer 50' is obtained as SP' = h - ΔS', and the bending of the scanning line can be corrected by adjusting according to the above. It can be corrected.

〔Effect of the invention〕

According to the present invention, the bending of the scanning line can be reduced to an extent that does not affect the image quality without strictly controlling the eccentricity tolerance, f-plane tolerance, etc. of the lens, and therefore without pushing up the cost l-. You can easily force it.

[Brief explanation of the drawing]

Figures 1 and 2 are diagrams for explaining the present invention, Figure 3 is a plan view of the scanning optical system, and Figure 1 explains the imaging relationship when the lenses are decentered and not. Figure 5 is a diagram in which each scanning beam is projected onto the main scanning section, Figures 6 and 7 are diagrams in which the scanning beams in the same figure are projected onto the sub-scanning section, and Figure 8 is a diagram in which the scanning beams of the same figure are projected onto the sub-scanning section. FIG. 9 is a perspective view of a scanning optical system suitable for implementation, FIG. 9 is a diagram illustrating eccentricity of a lens, and FIG. 10 is a diagram illustrating bending of a scanning line caused by a conventional scanning optical system. 51...fO lens, 52-2...toroidal surface, X-X...main scanning direction, Y-Y...sub-scanning direction. Chil (4>ru60

Claims (1)

[Claims] a first imaging optical system that forms a linear image of the light beam emitted from the light source; an optical deflector having a deflecting reflection surface set near the imaging position of the first imaging optical system; A scanned medium is scanned by a light beam deflected by a deflector, and a medium is disposed on the optical path of the light beam between the scanned medium and the light deflector, and is continuously deflected by the light deflector. The light beam is imaged on the scanned medium while maintaining an optically conjugate relationship between the deflection reflection surface of the optical deflector and the scanned medium in a plane perpendicular to the deflection plane that is the locus plane of the beam. A method for correcting bending of a scanning line due to eccentricity of a lens in the second imaging optical system, the method comprising: has a magnification in a direction (hereinafter referred to as sub-scanning direction) perpendicular to the scanning direction (hereinafter referred to as main scanning direction), and
A method for correcting the curvature of a scanning line, comprising decentering the lens in the sub-scanning direction and then fixing the lens having the lens surface with the maximum magnification among the lenses of the second imaging optical system.