Octagon35: spinner w/flipper, relay and bellows
 

I've been working on this for many months, and I finally have something that can produce images. The ground glass I'm using right now is only temprorary (I'll be replacing it with something thinner and that has finer grain).

I have very little time and probably won't get to post a screen grab for a while...

Still to do:
-I still have to prepare a little area for the batteries (2 9v) within the adapter.
-As you can see part of the GG is sticking out the top, so I just need to cover that up.
-It doesn't have an achromat (yet)
-I need a different potentiometer, as the one I'm using (1K) is far too sensitive.

http://www.particleproducing.com/images/p1010073c.jpg
http://www.particleproducing.com/images/p1010070c.jpg

An explanation of the design:

The most notable feature is perhaps the backwards Mamiya bellows on the front. This is so I have adjustable flange focal distance, giving me a the choice of just about any 35mm lens out there to use with this (I only have to make the appropriate mount). I'll eventually have a cine lens on there of some kind, but for now that's a 50mm FD shown in the pic. The relay lens is a 28mm FD attached to an FD/XL adapter I made. The main lens mount attaching the front lens to the bellows is also home made.

The spinner is actually a recent design change. The original plan was to have this be an oscillator, with the ground glass pertruding into the bellows. I wound up deciding at the last minute not to spend another x months getting my oscillator to work right, so I turned it into a spinner. The oscillator design I had is actually pretty promising, I think -- it incorporates a very wide (4mm) oscillation radius, with counterweights and a cog belt/pulleys to make sure the spindles don't get out of sync. I decided it was too much work for the advantages I was going for.

As I said earlier, the ground glass on it now is only temporary (I originally had a thinner piece with finer grain, but the glass cutter couldn't cut it big enough out of the piece I had, so the glass in it now is actually meant for a window!). The new glass will be also be easily interchangeable and very stable. The one on it now has a very slight wobble, but that wil be solved once I have time to work on the new GG.

The black areas are actually 1/4" foam that pop out so I can access the mirrors/motor etc.

Comment welcome!

Wow, that is a great design. I would love to see some stills from it.

So... this ties in nicely with a thread of mine. Does a 28mm FD lens on a mount work as a relay? How does it focus close enough Without an Achromat?

damn son...

nicely done...big old looking piece of hardware.

Wow, some adapters make cameras look like leaf blowers. This one looks like Camera Obscura mounted on a XL2.

Thanks for the comments... the next step is to do that major upgrade on the GG, and than I'll provide some stills. I shot some "video" (at this point, I hate calling it that ;)) with it before and it shows a lot of promise. The vignetting is significant and it's a little fuzzy, but after the new GG, fuzzy should be the last thing that comes to mind, and I'll be trying different achromat combinations until it's perfect.

Matthew, the 28mm lens is around 20cm from the GG (after making it through the mirrors) and about 2.4cm in front of the camera body... the FD-XL mount I made has that much space built into it, with screws on the inside for fine tuning. With those measurements, nothing but the lens is needed to make it work -- the spacing between the camera and relay is all that's needed. From your picture, it looks like you have something that works just fine with a little tweaking... you just need to mount the lens to the camera, right?

It is a big and bulky looking thing, and not the prettiest, but I realized that any adapter using a 28mm lens as a relay would have to have similar dimensions front-to-back.

Justin.

What is the aperture of the SLR on front of your adaptor, (not the lens you are using for relay to the camcorder.)

If it is not a f1.8 lens or aperture set to no tighter than about f2.8 then this might cause your vignette.

If it is a f1.8 lens then :-

I would be inclined to shorten the 24mm spacer distance between the 28mm lens and the XL camera body to about half what it is now.

This will mean, the GG may have to be furthur forward if the focal range of your 28mm relay lens is no longer enough.

This should hopefully give you a larger image onto the camcorder CCD and eliminate the vignette but I don't know what this then does to your mirror path.

Otherwise, what you have there seems to be a viable solution.

Thanks for the advice Bob!

I was playing with it some more today and experimenting a little with different distances between the GG, lens and camera, but to crop the frame enough to eliminate the vignetting I'd have to make more significant design changes than I'd like to. EDIT: I meant to include that if I wind up deciding I need to crop it, I'm just going to find a slightly longer focal length lens, so I don't have to change my setup at all. It was a lot of machining... the next one will have far less metal parts and build much faster. I want to make the design such that someone with ordinary tools could make one.

BUT, I aligned the mirrors better and that helped quite a bit (the image was slightly offset, so more of the darkness was in the frame than should have been). I was experimenting with a PCX and BCX behind the GG, and the BCX really helped. So I'm hoping the combination of a new GG and the BCX will just about eliminate the vignetting, though I think I can get even better results with different lens combinations to condense the light into the relay.

The front lens is f1.8, 50mm, full open, so that's not the problem. I'm looking at old Angenieux zoom cine lenses on ebay, so that might change the equation, too.

Once it's all working perfectly, I think I'm going to make a complete, very throurough "how to" guide w/ technical drawings, so people have a clear cut way to make one of these.

WOW, thats an impressive piece of work. From what you wrote, I understand that your adapter has an image errection system too, right?

Yep! Good thing too... if it weren't for that, the relay lens setup would have to be a lot more complicated.

So the relay lens is farther away from the chip than it would be using something like the EF adapter from Canon? Reason I ask is when I messed with a 24mm and the canon EF adapter as a relay lens the GG had to be way to far away and it made the entire adapter longer than it would have been using the stock XL lens.

Ahh, yes... that's why the presence of the image flipper is so important (not just for image flipping). By the time the light reflects through all those mirrors, the optical distance is far enough to resolve the ground glass, but the physical length of the device is much smaller. The exact conversion (as Quyen Le pointed out in a previous thread) is magnification x focal length. The magnification (cropping, actual) of a 1/3" CCD is 7.2x, so 7.2*24mm comes to 17.3cm. With my 28mm lens the GG needs to be 20.3cm away, but the adapter is only about 9cm from the front of the relay to the GG.

When it's totally finished I'm going to make a complete construction guide for anyone who wants to built one themselves without spending eternity figuring out actually how, and than tweaking it once it works. The guide will be very complete and specific, with a parts list and sources for those parts as well (and it will be more suited for every-day tools... I'm not going to assume the builder has access to a lathe, for example). If I had the money, I would have bought one a while a go, but as affordable as these adapters are becoming, there are still those who could only really justify the cost by self-building.

My goal is to get it producing images as good as just about anything else out there. I can't wait :) :)

Justin!

Thanks for the concise explination about why a 28mm Lens wont work as a relay on the Redrock M2. I have been searching for that exact answer for months now, and yours was very concise!

Now... can the 20.3cm wiuthout an image erection system be shortened by using an achromat lens or finding a lens that can focus incredibly close up?

I only ask cause you seem incredibly knowledgable in this!

You're welcom! ;)

To be honest, I just don't know what can be done with an achromat, but it should be possible to shorten that distance some way. It depends on how short a distance you want, but I don't see why a shorter focal length macro lens wouldn't work... but what kind of focal length, and whether or not we're talking significant distortion, I just don't know. Various combinations of lenses between the relay and ground glass can also affect magnification (and have an affect on distance). A good way to experiment with this stuff is to get a little test lens set. You get assorted sets of different types of (glass) lenses for very cheap... like this one: http://scientificsonline.com/product.asp_Q_pn_E_3040414. Unfortunately, such lenses aren't really suitable for actual use in a project... just testing different combinations.

Also, anyone looking for a source of reasonably priced first surface mirrors, the same site has them for $5 each. These are the mirrors in my adapter and I'm completely satisfied with them: http://scientificsonline.com/product.asp_Q_pn_E_3052324

There's a book I have called Optics and Optical Instruments by B.K. Johnson (Dover Books on Physics). It's thorough, and to be honest I haven't really gotten into it too much. I'll eventually have to take a class on optics anyway (physics major).

I also found this:http://www.osdoptics.com/

It's a graphical optical system design program, free for 90 days (good thing, because it's $1000 after that, so anyone who tries it should make good use of it in that time). The only limiting factor in the demo besides time is that the system is limited to 12 optical surfaces, but that shouldn't be an issue for our applications.

I never really got into that either (intense - steep learning curve), but maybe someone else could make use of it.

Ok, hope that stuff helps.

Also, some corrections:

I'd lost my ruler and just replaced it today...

The spacing between the relay lens and XL2 outer surface is 18mm. The physical length between the relay and GG is 8cm.

I've looked around and can't seem to find any alternatives to this program tha will run under OSX (Mac or X11 based). I'd like to get my adaptor going as I've completed building my rail system and the adapter...just have to get the inbetweeny bits working.

I'm stricktly a Windows user, so I don't have any suggestions... but by no means do you need such a program. I just pointed to it in case someone anyone wanted to give it a try.

First screens.
 

Well, it's not quite finished yet, but I just couldn't resist posting some preliminary screenshots. (I'll start a new posting when the "real" screen shots are up).

I'm still experimenting with different condensors and like I said they are preliminary shots... still tweaking, calibrating, etc. The GG (Thor Labs 1500 grit) is a vast improvement over the other one I was using, but it's a little bit too transparent... in some of the footage I have the highlights tend to glow (notice the paper on the third shot). The first two shots look great on one computer, and marginal on the other... they were shot late today with just a little available light.

http://www.particleproducing.com/images/o35-2c.jpg
http://www.particleproducing.com/images/o35-3c.jpg
http://www.particleproducing.com/images/o35-5.jpg

EDIT 8/3: Just shot these and thought I'd add 'em:
http://www.particleproducing.com/images/o35-7.jpg
http://www.particleproducing.com/images/o35-8.jpg

I have some new condensors on order, and the mounting bracket's already in place, so it's almost ready to shoot with. I think it's just a matter of evening out the lighting some more.

EDIT 8/3: I just ordered a Satin Snow GG... after experimenting with different glasses, I feel like the Satin Snow will be just what I'm after.

Complaints with the Thor Labs glass:

-Highlights "glow"
-Vignetting is intense and difficult to correct
-Bokeh sucks. I know this is normally a quality of the lens, but I definately realize that it matters what the lens is projecting onto. (See the fourth pic as an example).

Justin, I just checked out your thread here...very nice work! You have a world of experimentation ahead of you on the GG side :-) I've yet to see a spinning design (including M2) that did not show halation on specular highlights (particulary during focus pulls) so you may just have to live with it. I believe it is an interaction between the GG's clear side and the diffusor surface.

Hmm, that's not what I wanted to hear! But I do think adapters like the M2 at least have less halation, no? The vignetting I can fix... the fuzziness I can work on... but the only way I know to improve the glowing is with another GG. Do you think I'm on a wild goose chase, or is it possible to get it any better?

Obviously I'll have much more control over the highlights during an actual production (lighting), but it's still an issue.

Please, any advice would be appreciated! Anyone! Anyone?

An anti-reflective coating on the front surface as long as the shiny side is to the front would help. It is a long established method with lens optics, however it is also high science beyond my resources.

I am using optical glass disks.

I had limited but highly unpredictable results with imperfect polishing of the front surface which left a fairly dense scatter of smooth microscopic pits in the polished surface. It does not seem to affect the resolution but seems to affect light transmission.

The contrast seems to be better than with a disk with a totally smooth front surface. However my efforts at a home made level between individual finishing stages of disks and final groundglass textures are probably too variant and combined for this to be a valid suggestion.

The pitted polish also seems to aggravate flicker effect as I was unable to achieve a consistent density of pits around the entire disk because some areas would polish out entirely.

Optical coatings as for lenses would be the only way to go as I see it.

Oh wait... you have your shiny side in the front? I have the ground side facing the front (I thought that was standard practice to reduce light loss from reflections off the front). Is there an advantage to having the ground side face the camera? On one hand it makes sense to me to have the ground surface/image plane facing the relay lens, but I assumed that it didn't matter because everyone (I thought) faces it forward to eliminate front reflections.

In any case, I give you credit for trying to apply any kind of process to the glass... and if I understand right, you're grinding them yourself, too?

Shiny side to the front is the way I have done it and I think many builders have gone that way. I don't know what the theoretical advantage is.

An overbrightness point on a groundglass rear face will spread forward then some will reflect back as a spill around the bright point. Hopefully most goes out through the shiny side.

An overbrightness point on a front groundglass face will spread rearward. Some will be reflected forward as a spill projection around the bright point.

I don't know if there is any particular advantage.

I think any advantage would rest in that any reflection from defects, dust, fingerprints, etc., on the shiny surface in front of the groundglass will go forward and disappear into the non-reflective black internal finish of the adaptor and may be seen only as a shadow by the camera.

Whereas if the groundglass surface is on front, some defects or dust stuck to the shiny side might reflect light back onto the groundglass surface and then become surrounded by a small coronal light spill visible enough to spoil contrast.

This is purely theoretical gibberish on my part and not backed up by any research whatever, so don't take too much notice.

"This is purely theoretical gibberish on my part and not backed up by any research whatever, so don't take too much notice."

Bob Hart, the most helpful man in the world, and still the most modest also.