Hasselblad Réseau Plate

[granttes]

Hey everyone,

 

Is there a way to get a réseau plate to go on to my Hasselblad 503cx like they used in NASA for their moon shots? I know they won't have any scientific value to me, but the look is interesting.

[ben_hutcherson]

I'm kind of sort of talking out of my rear end, but I don't think it's as simple as just dropping in a plate.

 

The normal flange to focal plane distance is based on the assumption that the light is traveling through air. Air has a refractive index a tiny bit over 1(I think ~1.0003 or so) while common glasses have a refractive index of ~1.47(borosilicate glass-and I only know that off the top of my head because I was measuring some glass at work last week to see if it was boro) to around ~1.6(some types of lime glass). There might be other types outside this range, but whatever the case it's quite different from air.

 

This means that you'd need to set the film plane back a bit to compensate for it. In fact, if I'm not mistaken(I've never used mine and can't locate it at the moment) most 500 series Polaroid backs by design needed to set the film plane back a bit, and used a glass plate to keep the focus correct(my RB67, where I actually use the Polaroid back occasionally, physically attaches closer to the camera than a standard Graflok-type rollfilm back). NASA had a pretty much unlimited budget for the moon cameras, and Hasselblad was a willing and enthusiastic partner-without having examined one I would suspect that the 70mm bulk backs used on the moon were specially designed for NASA to accomodate the réseau plate.

[granttes]

I'm kind of sort of talking out of my rear end, but I don't think it's as simple as just dropping in a plate.

 

The normal flange to focal plane distance is based on the assumption that the light is traveling through air. Air has a refractive index a tiny bit over 1(I think ~1.0003 or so) while common glasses have a refractive index of ~1.47(borosilicate glass-and I only know that off the top of my head because I was measuring some glass at work last week to see if it was boro) to around ~1.6(some types of lime glass). There might be other types outside this range, but whatever the case it's quite different from air.

 

This means that you'd need to set the film plane back a bit to compensate for it. In fact, if I'm not mistaken(I've never used mine and can't locate it at the moment) most 500 series Polaroid backs by design needed to set the film plane back a bit, and used a glass plate to keep the focus correct(my RB67, where I actually use the Polaroid back occasionally, physically attaches closer to the camera than a standard Graflok-type rollfilm back). NASA had a pretty much unlimited budget for the moon cameras, and Hasselblad was a willing and enthusiastic partner-without having examined one I would suspect that the 70mm bulk backs used on the moon were specially designed for NASA to accomodate the réseau plate.

 

I remember finding this page a while ago:

 

Hasselblad MK70

 

If you look at the 4th image, you can see how the reseau plate is inserted into the body. As I look at the back of my hasselblad, it looks like I can see where they placed that plate in. I don't think you have to even remove the barn doors on the back. From the image on that link, it looks like there are 2 solder joints as well.

[orsetto]

According to reference books like Richard Nordin's Hasselblad Compendium, the plate-equipped Haselblads were quite heavily modified from the standard versions. Unlike the Polaroid back, which had a self-contained flush glass plate, the post-Apollo MK70 Réseau plate is bonded to the camera back plate and seems to project out from it (meaning it would likely crash into the dark slide of any standard magazine). So MK70 magazines don't have a dark slide: once mounted, they can't be removed mid-roll without fogging a couple frames. Earlier Apollo lunar surface Hasselblads do appear to have had standard dark slides, suggesting the Réseau plate in those was recessed more into the body cavity.

 

All Réseau models omitted the mechanics involved in reflex viewing, relying instead on detachable shoe-mounted frame finders (like the SWC). No reflex mirror, and no barn doors at the rear. Lenses were modified with shutters that only opened for exposure, remaining closed while cocked (thus eliminating any need for barn doors to protect the film). Lens optics also had slightly offset back focus to accommodate the Réseau plate refraction, as ben_hutcherson surmised. In many cases specific lenses were matched to specific bodies + Resau plates.

 

Compatibility of Réseau film magazines was limited to their respective camera series (Apollo or MK70), and none will mate properly with a non-Réseau body.

 

The Resau plate solder points (seen in the MK70 pics granttes linked to above) were part of an anti-static grounding modification to prevent spurious light sparks (from glass/film friction) registering on the film.

[rodeo_joe1]

Apart from keeping the film flat, what was the exact point of the crosshairs on the moonwalk 'blads anyway?

 

Did it help measurements? I doubt it, since you'd need to know the exact subject/focussing distance and magnification for film-plane ticks to be of any use whatsoever. (One up to conspiracy theorists!)

 

And, yes, Ben has it right. Inserting a transparent plate in front of the film would need a complete recalibration of the camera's focus.

 

Just do what any sensible person would do, and fake them on digitally in post.

[tom_chow]

Apart from keeping the film flat, what was the exact point of the crosshairs on the moonwalk 'blads anyway?

 

Did it help measurements? I doubt it, since you'd need to know the exact subject/focussing distance and magnification for film-plane ticks to be of any use whatsoever. (One up to conspiracy theorists!)

 

And, yes, Ben has it right. Inserting a transparent plate in front of the film would need a complete recalibration of the camera's focus.

 

Just do what any sensible person would do, and fake them on digitally in post.

The reseau marks tell you if the image was distorted due to printing or scanning, or film media distortion during processing.

 

The moon cameras had calibrated lenses, so the fixed 60mm lens resulted in a 10.3 deg separation of the fiducial marks.

 

Distances are measured using photogrammetry - which they did do the moon by recording the position that the photographs were taken.

 

Ofcourse this all actually happened in a studio on earth.... ;)

 

 

 

 

PS: if you install a reseau plate in your earthy Hasselblad, you just have to install an equivalent plate under you screen (without changing it's registration). Then the camera will focus past infinity, but correctly on film.

[rodeo_joe1]

The moon cameras had calibrated lenses, so the fixed 60mm lens resulted in a 10.3 deg separation of the fiducial marks.

 

Distances are measured using photogrammetry - which they did do the moon by recording the position that the photographs were taken.

 

10.3 degrees? You'd have thought somebody might have done their maths a bit better before having the plates engraved, and made it a nice round 10 degrees.

 

Also, you need to have a plane horizontal reference, or known camera angle and height for any photogrammetry calculations to be at all useful.

"Somewhere around Neil Armstrong's chest height" and "That looks to be about 8 foot away" aren't really accurate enough for anything but a guesstimate - whether you have stupid fiducial marks on the film or not.

 

Taking a decent ruler and tape measure to include in the pictures would have been far more sensible.

 

Nope. My guess is the tick marks were simply a product of the cold-war hype and propaganda machine; designed to look 'sciencey' and technical on the publicly released fuzzy snapshots.

Edited January 24, 2019 by rodeo_joe|1

[tom_chow]

10.3 degrees? You'd have thought somebody might have done their maths a bit better before having the plates engraved, and made it a nice round 10 degrees....

 

<snip>

 

Nope. My guess is the tick marks were simply a product of the cold-war hype and propaganda machine; designed to look 'sciencey' and technical on the publicly released fuzzy snapshots.

There is no scientific reason for fiducials to be an even 10.0 deg apart. It's better to use the same spacing on all you reseau plates, instead of custom ones for each lens.

 

https://www.hq.nasa.gov/alsj/a11/a11Photogrammetry.html - but I'm sure it's all fake, as were the published papers ;)

[rodeo_joe1]

So, basically the astronauts' footprints, and known dimensions of the lunar module, were of far more use in photogrammetry than all the silly cross marks on the film.

 

I thought so.

 

Don't they own any rulers to simply stick across the film in Nasa's photo analysis department?

[tom_chow]

So, basically the astronauts' footprints, and known dimensions of the lunar module, were of far more use in photogrammetry than all the silly cross marks on the film.

 

I thought so.

 

Don't they own any rulers to simply stick across the film in Nasa's photo analysis department?

The silly marks are how they get the mag factor, and hence distance to the known object, with a much higher resolution than the film gate dimension.

 

I don't quite understand your disdain for photogrammetry, they spend an obscene amount of money to send some equipment to the moon (and there is a "waste of money" debate) for a very short time, and it seems quite reasonable to use as many tools at your disposal as possible to maximize the return. However, if I went up there just to take photographs, I'd be pissed about those cross hairs in my pictures.

[rodeo_joe1]

The silly marks are how they get the mag factor, and hence distance to the known object, with a much higher resolution than the film gate dimension.

 

- Please explain exactly how that works.

 

Reseau marks can calibrate a projection system for scale and distortion, but they can't possibly calibrate the taking lens without reference to a precisely-known subject size and distance.

 

The projection lens can easily be calibrated with a scale plate in place of the film, and/or with a corresponding scale on the projection table.

 

Given that all subject measurements could only be known approximately, given lack of orthogonality and parallelism with the subject(s), the use of precise reseau marks begins to look like ridiculous false precision.

[tom_chow]

- Please explain exactly how that works....

I think you know the answer, so lets just agree to disagree on this.

 

Every second they spent on the lunar surface was scripted, where and when they took the pictures, focus settings and where to point. Nasa had dimensionally accurate copies of every object they took to the moon, so orientation and measurements were available. Much of what they did, mapped out the location of the experiments, and the surrounding hills. Could the measurements be made without the Réseau marks? Yes. To the same accuracy? No. It is always better to use a instrument that is more precise than you need. Nasa tested photogrammetry using these cameras in Nevada, Arizona and Hawaii, and a decision was made to use réseau marks for their lunar surface data cameras.

[rodeo_joe1]

I think you know the answer, so lets just agree to disagree on this.

 

- No, I don't. I can't think of any way that dividing up the frame, with no matter what degree of precision, could possibly improve the guesstimate of subject angle, distance and size.

 

So please explain how you think this works.

 

In fact, how did the film - inside a magazine - make contact with a glass plate inside the hasselblad body? Because as someone else pointed out, the plate would have to protrude beyond the plane of the darkslide.

 

I also thought that the focus was fixed on the lunar 'blads.

[tom_chow]

- No, I don't. I can't think of any way that dividing up the frame, with no matter what degree of precision, could possibly improve the guesstimate of subject angle, distance and size.

 

So please explain how you think this works.

 

In fact, how did the film - inside a magazine - make contact with a glass plate inside the hasselblad body? Because as someone else pointed out, the plate would have to protrude beyond the plane of the darkslide.

 

I also thought that the focus was fixed on the lunar 'blads.

Yes you do, you just don't believe that using the fiducials to correct for distortion is any better than using the film gate, and no amount of discussion here is likely to change that. We'd have to go make some measurements. It is what NASA decided to do, and I'm just relaying the reasons.

 

Once you have the distortion corrected image size measurement to 3 figures, the accurate lens focal length and the known object size, you have the distance to 3 figures. For mapping, they use the angle between objects, and the known positions of the camera. As you intoned, this can be done without the réseau plate if the film gate is similarly accurate. But they decided to use the réseau markers - remember that in 1969, they would measure off the paper print, réseau marks were much easier to use than the film gate when you have a large print. Réseau marks also showed nonlinear distortion.

 

The réseau plate has raised edges that act as the film rails, hence the need to ground the film from static. It sticks into the film back. You must remove the film back before you can insert the dark slide, you will lose 3~4 frames if you decide to change film backs mid roll.

 

I do remember reading that some lenses were locked at infinity (ie: the 250mm for shooting out the command module window). I don't remember about the lunar surface data cameras (and I no longer have the old "paper" paper), but if you look at the picture of the footprint, you can see the focus plane through the footprint, along with the réseau marks. So that camera was not focused at infinity. If you search for pictures of the lunar data camera, some don't appear to have a focus tab, but some do.

[rodeo_joe1]

remember that in 1969, they would measure off the paper print,

A paper print? Oh real accurate!

Do you know how much paper stretches or shrinks in processing and with humidity?

 

They would more likely have projected the film onto a machined flat white surface for measurement. Which means that the projection lens distortion needs to be taken into account. Reseau marks will do that to some extent, but a calibrated scale on the projection table will do just as well or better.

 

I still think the ticks on the film were complete overkill for the degree of accuracy that could possibly be got without accurate measurement of subject distances.

 

And if the lens was focussed, then its magnification would alter with each focus setting, rendering reseau marks useless unless the exact scale distance was recorded. No EXIF data in those days remember.

[tom_chow]

...I still think the ticks on the film were complete overkill for the degree of accuracy that could possibly be got without accurate measurement of subject distances...

I thought subject distances was one of the things they were measuring?

 

Having a réseau grid allows you to verify all the distortion (film, paper, lens). Projected film will thermally distort, and does not add much accuracy while complicating the process. Having a temperature stable print works well.

 

What did we ever to before EXIF? They recorded all the camera settings for the mapping pictures, and the hard stop on the data camera was calibrated. Some of them may have been fixed focus, but I don't have that information.

[Wobbegong]

Hello

Can anyone tell me why there are so many duplicated and distorted fiducials on Apollo images. I'll post one example but there are literally dozens of them. In this image there are four duplicated and distorted fiducials surrounding the sun. How does this happen, thanks!

 

[rodeo_joe1]

Nope. I have no idea how any reflection or ghost image could have more contrast and definition than the original engraving(?). Nor how the ghost image could have an entirely different geometry and lack of linearity. Looks as if some shakey-handed child has tried to retrace the original cross hairs with a sharp pencil. 

Other published moon images I've seen look like the crosses have been added to a print afterwards. 

And what possible use those crosses are is a mystery to me as well. Because any photogrammetry would require knowing the subject distance and lens characteristic. Something that could only be done on Earth before the camera was sent on its one-way trip. Knowing how much the film had distorted after use and processing would add hardly any useful information. 

[arthur_gottschalk]

I've also been fascinated by the look of those pictures. I think it would be possible to replicate it in the darkroom with an overlay while printing. 

[q.g._de_bakker]

The use of those crosses partly is for photogrammetry, meassuring distances and perspective in the image (and yes you would need to know the lens. Which they did). And also to correct for film distortion. Emulsions swell and shrink during processing, changing geometry of the image in it. Knowing by how much is part of photogrammetry.

 

What we see here is extreme flare in the Biogon. Reflections bouncing back and forth, between film, reseau plate and lens elements. Just like images of the source of light end up in other places due to these reflections, so do marks anywhere on any of the reflecting surfaces.

Edited October 24, 2022 by q.g._de_bakker

[arthur_gottschalk]

I think the OP is more interested  in duplicating the "look" of those pictures than any actual measurements. 

[tom_chow]

The re

18 hours ago, Wobbegong said:

Hello

Can anyone tell me why there are so many duplicated and distorted fiducials on Apollo images. I'll post one example but there are literally dozens of them. In this image there are four duplicated and distorted fiducials surrounding the sun. How does this happen, thanks!

The réseau plate is a parallel glass sheet, with 2 glass-air surfaces, which reflect. The first double reflection (off the back surface, then off the front surface and back onto the film) is a veil glare that projects the reflected crosses back onto the film, with a raypath that is centered on the sun at the rear focal point of the lens. This lens is the Biogon, which sits quite close to the film (there is no mirror on this camera, and the lens extends right back close to the film plane), and has a short rear focus. The original crosses are muted by the veil glare. The shape of the secondary crosses follow the projected raypath from the point at the center of the sun through the double reflection.

[Wobbegong]

Hello, thanks for the feedback. The reason I came here is because whenever I ask these question of Apollo fans or detractors I often get conflicting, emotional and angry replies. I used to do silver halide photography with an Olympus OM-1 back in the day and I do find the Apollo images rather curious. I'd like to know what these strange curving parallel lines are on this image from Apollo 13. They appear on 6 images in this magazine only, I doubt it's something on the window as everything is in focus and look like an artifact that's part of the reseau plate. Here's a clearer view and you'll notice there's a large, somewhat askew cross quartering the image. 

I really appreciate your feedback but if you don't want to get involved with such questions I'll leave at that if you wish.

Thanks so much!

 

[tom_chow]

I do not know what that canted cross, or the double spiral is from. They both get fuzzier as you move away from the center of the image, so that would indicate something that is not on the rear surface of the réseau plate. The large canted cross also splits into a double line at the top before it disappears into the black, so that may be an optical effect of whatever is sitting on the réseau plane.

[rodeo_joe1]

Maybe someone should have considered adding an AR coating to the pressure plate? 

And I still don't see how you can get curved reflections between two plane-parallel surfaces.

Even if the secondary images are from the rear of the lens, the curvature should then centre on the optical axis, not on the source of light.