Image of grog
Greg's panorama settings for my Olympus cameras
Greg's photography pages
Greg's photo albums
Greg's photos
Greg's photo index
Groogle
http://www.lemis.com/grog/wip/workinprogress.gif
Work in progress

This page contains details about how to adjust my panorama hardware to position the entrance pupil of the lens in a position where it can be rotated around two axes without moving from its position (the “no-parallax point”). The values are specific to my hardware, but the principles aren't. It's based on a table for my Olympus cameras at http://olypedia.de/Nodalpunkte_E_System. It's in German, but easy enough to understand, but unfortunately it's rather out of date.

The basic principle is simple: the entrance pupil of a lens is a certain distance from the sensor. For a zoom lens it will probably change with focal length. Things are somewhat complicated by the fact that you can't easily measure from the sensor, only from the tripod mount, and Olympus delights in changing the position of the tripod mount (for example, there's a difference of 12 mm between the OM-D E-M1 Mark I and the OM-D E-M1 Mark II). The German page gives offsets from sensor to tripod mount for most Olympus cameras. If your lens isn't listed, you can either measure the entrance pupil location yourself—something I've found far too inaccurate—or guess that it's roughly at the middle of the front element of the lens.

The tables below assume correct positioning of the camera (below). They're the total distance along the lens axis as shown on my positioning rail, including the distance from sensor to entrance pupil on the one hand and sensor to tripod on the other hand. Your mileage will vary.

On 7 August 2016 I discovered that my Olympus Zuiko Digital ED 8 mm f/3.5 fisheye lens, which according to the table should be set at 94 mm for the E-M1 Mark I, gave better results at 97.5 mm. So something is wrong in the data below, but I'm not sure where, and whether it affects other lenses too.

E-M1 Mark II

Focal length 7-14 8mm 9-18 11-22 12-60 14-35 14-42 14-45 14-54 25mm 35mm 50mm
7 mm 118
8 mm 115 82
9 mm 113 87
10 mm 112
11 mm 85 103
12 mm 113 102
14 mm 113 84 98 66 88 88
16 mm 87
18 mm 89 93 93 64 88 81
22 mm 88
25 mm 86 64 88 70 41.2
35 mm 76 60 95 59 39
42 mm 46
45 mm 100
50 mm 58 51
54 mm 44
60 mm 50
E-M1 Mark I

Focal length 7-14 8mm 9-18 11-22 12-60 14-35 14-42 14-45 14-54 25mm 35mm 50mm
7 mm 130
8 mm 127 94
9 mm 125 99
10 mm 124
11 mm 97 115
12 mm 125 114
14 mm 125 96 110 78 100 100
16 mm 99
18 mm 101 105 105 76 100 93
22 mm 100
25 mm 98 76 100 82 53.2
35 mm 88 72 107 71 51
42 mm 58
45 mm 112
50 mm 70 63
54 mm 56
60 mm 62

E-PM2

Focal length 7-14 8mm 9-18 11-22 12-60 14-35 14-42 14-45 14-54 25mm 35mm 50mm
7 mm 123
8 mm 120 87
9 mm 118 92
10 mm 117
11 mm 90 108
12 mm 118 107
14 mm 118 89 103 71 93 93
16 mm 92
18 mm 94 98 98 69 93 86
22 mm 93
25 mm 91 69 93 75 46.2
35 mm 81 65 100 64 44
42 mm 51
45 mm 105
50 mm 63 56
54 mm 49
60 mm 55

E-PM1

Focal length 7-14 8mm 9-18 11-22 12-60 14-35 14-42 14-45 14-54 25mm 35mm 50mm
7 mm 124
8 mm 121 88
9 mm 119 93
10 mm 118
11 mm 91 109
12 mm 119 108
14 mm 119 90 104 72 94 94
16 mm 93
18 mm 95 99 99 70 94 87
22 mm 94
25 mm 92 70 94 76 47.2
35 mm 82 66 101 65 45
42 mm 52
45 mm 106
50 mm 64 57
54 mm 50
60 mm 56

Camera positioning

To take a good panorama, you need to eliminate parallax between the individual shots. This means that the camera must be in the same position for each shot. But which part of the camera? If you turn it, most parts will be somewhere else.

The part that's important is where the light goes into the lens, technically called the entrance pupil. Very frequently people call this the nodal point, though this is incorrect.

So where is the entrance pupil? There's a simple answer and an accurate answer. The simple answer assumes that the entrance pupil is on the lens axis, so the only thing you need to know is how far along the axis it is positioned.

And the accurate answer? For some lenses the position of the entrance pupil depends on the angle of view. Not only is it not always on the lens axis, it's not even a point: it's an area. Like most people, I put this category into the “too hard” basket.

So how do you measure the entrance pupil? There are various methods, which I won't describe here. I simply rely on other people's measurements. There's a table of entrance pupils for my Olympus cameras at http://olypedia.de/Nodalpunkte_E_System. It's in German, but easy enough to understand.

Knowing the position of the entrance pupil isn't enough. The distance is measure from the sensor, not something that you can easily locate. Normally you screw the camera on a rail, so the reference point is the tripod mount, which may be offset from the sensor in all three directions. In particular, it could be offset along the optical axis. The German table gives measurements for this as well. In my case, the tripod mount of the Olympus OM-D E-M1 is 2 mm behind the sensor, while the mount of the Olympus E-PM2 is 5 mm in front. These values need to be added to or subtracted from the values in the table. I'll discuss offsets in the other two directions below.

My current hardware includes a bracket mounted on a rail, which in turn is mounted on a rotator:


https://lemis.nyc3.digitaloceanspaces.com/grog/Photos/20151108/big/Pano-rail-5.jpeg
Image title: Pano rail 5          Dimensions:          3567 x 3022, 960 kB
Make a single page with this image Hide this image
Make this image a thumbnail Make thumbnails of all images on this page
Make this image small again Display small version of all images on this page
All images taken on Sunday, 8 November 2015, thumbnails          All images taken on Sunday, 8 November 2015, small
Diary entry for Sunday, 8 November 2015 Complete exposure details

 
https://lemis.nyc3.digitaloceanspaces.com/grog/Photos/20151108/big/Pano-rail-6.jpeg
Image title: Pano rail 6          Dimensions:          4608 x 3456, 2224 kB
Make a single page with this image Hide this image
Make this image a thumbnail Make thumbnails of all images on this page
Make this image small again Display small version of all images on this page
All images taken on Sunday, 8 November 2015, thumbnails          All images taken on Sunday, 8 November 2015, small
Diary entry for Sunday, 8 November 2015 Complete exposure details

 

The camera is mounted on two rails at right angles:


https://lemis.nyc3.digitaloceanspaces.com/grog/Photos/20151108/big/Pano-rail-1.jpeg
Image title: Pano rail 1          Dimensions:          4608 x 3456, 2368 kB
Make a single page with this image Hide this image
Make this image a thumbnail Make thumbnails of all images on this page
Make this image small again Display small version of all images on this page
All images taken on Sunday, 8 November 2015, thumbnails          All images taken on Sunday, 8 November 2015, small
Diary entry for Sunday, 8 November 2015 Complete exposure details

 
https://lemis.nyc3.digitaloceanspaces.com/grog/Photos/20151108/big/Pano-rail-3.jpeg
Image title: Pano rail 3          Dimensions:          4608 x 3456, 2544 kB
Make a single page with this image Hide this image
Make this image a thumbnail Make thumbnails of all images on this page
Make this image small again Display small version of all images on this page
All images taken on Sunday, 8 November 2015, thumbnails          All images taken on Sunday, 8 November 2015, small
Diary entry for Sunday, 8 November 2015 Complete exposure details

 

The scale on the longitudinal rail shows the distance between mounting point and rotator axis. The other two rails are lateral, and in principle one is superfluous, but the bottom arm of the panorama bracket is too short, so this enables me to position the side of the bracket further from the vertical rotational axis:


https://lemis.nyc3.digitaloceanspaces.com/grog/Photos/20151108/big/Pano-rail-11.jpeg
Image title: Pano rail 11          Dimensions:          4608 x 3456, 1040 kB
Make a single page with this image Hide this image
Make this image a thumbnail Make thumbnails of all images on this page
Make this image small again Display small version of all images on this page
All images taken on Sunday, 8 November 2015, thumbnails          All images taken on Sunday, 8 November 2015, small
Diary entry for Sunday, 8 November 2015 Complete exposure details

 
https://lemis.nyc3.digitaloceanspaces.com/grog/Photos/20151108/big/Pano-rail-16.jpeg
Image title: Pano rail 16          Dimensions:          3456 x 4608, 1456 kB
Make a single page with this image Hide this image
Make this image a thumbnail Make thumbnails of all images on this page
Make this image small again Display small version of all images on this page
All images taken on Sunday, 8 November 2015, thumbnails          All images taken on Sunday, 8 November 2015, small
Diary entry for Sunday, 8 November 2015 Complete exposure details

 

The important thing to note from the front view is that the lens axis intersects both the rotational axis of the rotator and the rotational axis of the rotator at the left-hand side. These don't need to be adjusted. Along with the adjustment along the axis, this ensures that the entrance pupil stays in the same place regardless of rotation.

Older stuff

The following is from an earlier version of this page. Under some bizarre circumstances it could be of use.

First I need to mount the cameras, an Olympus OM-D E-M1 and an E-PM2 on my hardware, consisting of a cheap, badly designed panorama bracket and a couple of focusing rails to compensate for the design errors of the bracket. The camera is mounted on one of two different “Fotomate” rails (more of that below) which in vertical orientation fits into the mount of the upper rotator of the panorama head (the L-shaped frame on the left). These photos show a previous camera, an Olympus E-30.


https://lemis.nyc3.digitaloceanspaces.com/grog/Photos/20120622/big/Pano-mount-2.jpeg
Image title: Pano mount 2          Dimensions:          2112 x 2816, 544 kB
Make a single page with this image Hide this image
Make this image a thumbnail Make thumbnails of all images on this page
Make this image small again Display small version of all images on this page
All images taken on Friday, 22 June 2012, thumbnails          All images taken on Friday, 22 June 2012, small
Diary entry for Friday, 22 June 2012 Complete exposure details

 

The panorama bracket comes with a useless rotator which is not adjustable in position along the rail to which is mounted. It is wrong for anything I want to do. I haven't been able to remove it, so I have just locked it in position so that it doesn't rotate and mounted it on a 16 cm Fotomate rail, which in turn is screwed to a Sunwayfoto DDP-64M rotator below (not visible here, but directly below the plate with “FOTOMATE” written on it), which does work well. The toy rotator is the cylinder round offset 8 mm on the rail.

Horizontal centring

For this particular hardware and (vertical) orientation, the camera is centred above the rotator when the scale is set to 4.4 cm. In horizontal orientation it needs to be readjusted to 5.8 cm:


https://lemis.nyc3.digitaloceanspaces.com/grog/Photos/20120622/big/Pano-mount-1.jpeg
Image title: Pano mount 1          Dimensions:          2112 x 2816, 560 kB
Make a single page with this image Hide this image
Make this image a thumbnail Make thumbnails of all images on this page
Make this image small again Display small version of all images on this page
All images taken on Friday, 22 June 2012, thumbnails          All images taken on Friday, 22 June 2012, small
Diary entry for Friday, 22 June 2012 Complete exposure details

 
https://lemis.nyc3.digitaloceanspaces.com/grog/Photos/20120622/big/Pano-mount-4.jpeg
Image title: Pano mount 4          Dimensions:          2112 x 2816, 656 kB
Make a single page with this image Hide this image
Make this image a thumbnail Make thumbnails of all images on this page
Make this image small again Display small version of all images on this page
All images taken on Friday, 22 June 2012, thumbnails          All images taken on Friday, 22 June 2012, small
Diary entry for Friday, 22 June 2012 Complete exposure details

 

In horizontal orientation, it's also important to ensure that the axis of the lens is at the height of the upper rotator (or the toy vertically mounted compass attached to the clamp). In vertical orientation this happens automatically.

The position of the camera along the upper rail depends both on the lens and on the rail I choose. The following is an adaptation of the information at http://olypedia.de/Nodalpunkte_E_System. It reduces to readings on the scale of the 16 cm Fotomate rail:


https://lemis.nyc3.digitaloceanspaces.com/grog/Photos/20120622/big/Rail-3.jpeg
Image title: Rail 3          Dimensions:          2816 x 2112, 896 kB
Make a single page with this image Hide this image
Make this image a thumbnail Make thumbnails of all images on this page
Make this image small again Display small version of all images on this page
All images taken on Friday, 22 June 2012, thumbnails          All images taken on Friday, 22 June 2012, small
Diary entry for Friday, 22 June 2012 Complete exposure details

 

This rail has a distance of 108 mm from the tripod mounting hole when the scale is set on 0 and the camera is mounted at the far end of the adjustable slot. The scale is the wrong way round, so the distance reduces as the scale indication increases. In addition, the tripod mounting hole of the E-M1 is 2 mm behind the sensor plane, and the mounting hole of the E-PM2 is 5 mm in front of the sensor plane so this value needs to be added. The real value for the E-M1 is thus offset = 108 + 2 - scale, or scale = 108 - offset - 4.

The magic here is a little function that converts the individual values in the table, so that I don't have to do it manually:

Focal length 7-14 8mm 9-18 11-22 12-60 14-35 14-42 14-45 14-54 25mm 35mm 50mm
7 mm -22
8 mm -19 14
9 mm -17 9
10 mm -16
11 mm 11 -7
12 mm -17 -6
14 mm -17 12 -2 30 8 8
16 mm 9
18 mm 7 3 3 32 8 15
22 mm 8
25 mm 10 32 8 26 54.8
35 mm 20 36 1 37 57
42 mm 50
45 mm -4
50 mm 38 45
54 mm 52
60 mm 46

I can't set negative values (the ones marked in red), so for those values I need to use a longer rail (26 cm), again from Fotomate:


https://lemis.nyc3.digitaloceanspaces.com/grog/Photos/20120622/big/Rail-4.jpeg
Image title: Rail 4          Dimensions:          4032 x 1738, 592 kB
Make a single page with this image Hide this image
Make this image a thumbnail Make thumbnails of all images on this page
Make this image small again Display small version of all images on this page
All images taken on Friday, 22 June 2012, thumbnails          All images taken on Friday, 22 June 2012, small
Diary entry for Friday, 22 June 2012 Complete exposure details

 

This one is presumably meant for mounting flash units. As shown, it is set with the mounting hole directly below the inside end of one of the slots. The scale shows 9 mm left of the 0 point on the scale. If I mount the camera in this position, at the right end of the left slot, and point it to the right, I can move the base of the rail as far as 121 mm in that direction (not quite to the end of the scale), giving me a total range of 130 mm. In this case the calculation is more straightforward: offset = 108 + 2 - scale, or scale = offset + 2 - 9.


Greg's home page Greg's diary Greg's photos Copyright

Valid XHTML 1.0!

$Id: entrance-pupils.php,v 1.11 2017/09/05 02:33:22 grog Exp $