frosted cd, london? will 600 sandpaper do?
 

hey all,

been following the agus35 thread for a bit & cannot wait to get mine built (agus, you rule)

so, i have all my parts, a f1.4 55ml nikkor lens, cd ready for breaking, wood & screws for chassis, close up lenses, just struggling on the ground glass bit

(btw, using http://www.mediachance.com/dvdlab/dof/index.htm step by step guide & my camera is a pdx10p)

the maxell set is so hard to find! i tried simply computers, went down myself as the website had the correct picture, when i got there it was the xl but had that same product number - tried dixons, maplins, currys, scan.co.uk, dabs - tons - no joy

so i picked up some p600 silicone carbide sandpaper, so the question is, for those of you that have had experience, would the 600 really compromise image quality?

and does anyone know where those darned maxells are anywhere in london?

(i tried looking for aluminium oxide, no avail)

thing is i plan on shooting a final uni project using it, so i really need to get this built asap

any help would be very very appreciated, have so much respect for every person that has pitched in

hey

i phoned simply & misco, they were crap, they had no idea, so i called maxell, they said they only had the blue xl ones in stock and the woman guessed that the frosted cd's were only available from america

soooo, do i run with the 600 silicon carbide paper,
or could any wonderful people in america send me a frosted cd (i'll pay for p&p & cdpack of course),

or, are they all wrong and is there anyone who knows of a uk souce,

or where i can get some aluminium oxide

- and finally, what cd packs in uk have the clear cd?

cheers :)

Don't take particular notice of this. The method is my own personal preference.

Sony CD-R disk packs also have the clear CDs. Unfortunately some packs have moulded spacers which are not as dimensionally accurate. They also have a cut-out on the edge which throw them out of balance. Another cut-out has to be filed in exactly 180 degrees opposite to balance them. Otherwise they are workable but may give you some interlace artifacts in your camcorder image due to slight movement of the projected image on the disk if it doesn't run true.

In other Sony packs, the clear disks are actually CD-Rs which have been taken out of production before the metal coat and top layer was applied. These disks are optically superior but have guide tracks on them. The groundglass finish must be applied to the guide track face. This is easily found by rubbing the surface with your finger which will partially erase them if they are there.

I found that sanding the disks made them too opaque. The texture of the finish was also made up of a lot of scratches.

I tried placing the disk on a flat hard surface on top of a piece of thin paper. I placed a piece of wed & dry 600 grade silicon carbide paper face down on the disk. I did not rub the disk but instead put extreme pressure on the back of the paper to stamp pits into the disk by using the round end of a metal scriber handle to rub the back of the paper itself. It is essential for the wet & dry paper not to skid on the disk otherwise scratches, not pits are made in the disk. The back of the wet & dry paper wears out fairly quickly. A better method is to get hold of an old complete ballbearing from an automotive repair shop. All that matters is that it still turns freely. Hold the centre (axle) part and roll the outer rim of the bearing on the wet & dry paper with lots of pressure. To get the loca pressure through the wet & dry, you need to tilt the bearing slightly so that the rounded corner of the outer rim runs on the wet and dry paper.

You will get lots of thin opaque tracks pressed onto the disk. You just have to keep at it until all the tracks become one complete opaque surface.

You will need to lift the wet & dry paper often and reposition it so that local densely packed areas on the paper eventually contact the entire disk. By this method you will achieve what I regard as a "beginner level" finish which will enable experiments. This surface is very vulnerable to damage from finger contact. I subsequently furthur dressed a disk I made by this method with fine aluminium oxide in a light water mix, by rubbing the disk lightly on a sheet pf flat glass, then I lightly dry-polished the disk with a soft flannel cloth to restore some transparency to improve the light transmission.

I made other disks with various grades of fine powders but no plastic disks by ther methods came
up to the performance of the pressed disk.

Whilst prototyping, I find the best method of mounting the disks onto the spindle was to apply a thin coat of water clean-up bathroom silicon sealer onto the contact face and a thin bead on top of the joint between the disk and the hub. Immediately position the motor and disk horizontally and power the motor up. Provided the motor is firmly mounted, the disk will automatically align itself and the sealer will migrate to where it needs to be. Run it for about five minutes then leave it at rest to set properly.

This stuff is good because it is robust enough to hold the disk secure but can be torn away and re-done if the disk is no good. When removing the disk, it is best to use a blunt blade out of a disposable razor to work at many points around the disk/spindle face junction as a wedging not cutting action. Watch your fingers!! The spindles are hard to get ( usually one per CD player only. )

They are quite fragile and if broken, cannot be glued back together with any hope of satisfactory alignment.

On "www.dvinfo.net/media/hart" all my test images posted there, except those identified as being through a glass GG were shot through the pressed, dressed, back-polished disk.

It's going to take you over an hour to get a decent pressed finish.

Look at the other methods which have been used by other builders which may be better than mine.

Good luck.

bob,

wow, thank you for such a comprehensive reply, i'll get on that method straight away, bit unsure about the ball-bearing bit, i'll just wander into a mechanics and see what they have to say

securing the disc to the spindle, i'm assuming that's for smoother rotation?

i'm not sure i'll be able to find any aluminium oxide, so is the flannel polishing advised? or, if i alternated between the two, could i get a smoother surface

sorry for being dim but "You will need to lift the wet & dry paper often and reposition it so that local densely packed areas on the paper eventually contact the entire disk" - i thought, better to keep it still so each part of the disc recieves the same degree of coarseness from the paper, which seems to me to be evenly coated?

oh and i've been studying those images obsessively, bookmarked, now i need to pull it off!

thanks once again for the detailed reply and i will make sure to post up test pics and build pics as soon as i have anything resembling ready

The pressed method is my own and probably not the ideal. It just gets away from lots of flying dust or water getting around.

If you can't find aluminium oxide, going straight to a backpolish with a flannel cloth will probably do. The result may be a bit patchy though. You may have to do the pressing bit then the back polishing bit a few times in turn. For back polish, light and fast not heavy and slow, otherwise any pieces of grit embedded in the CD may come out and scratch across the disk. You'll have to be scrupulously clean. Any finger grease will become permanent less opaque patches in the finish.

As to the ball bearing, you are after a complete bearing including the inner and outer races. The best ones might come out of automotive air-conditioner compressor pulleys. You could improvise with a furniture castor which would also give you a handle of sorts to hold on to.

As to lifting the wet and dry paper and re-positioning it, I thought it would be uniformly coated but there seems to be variation across the sheets in the form of ripples or bands.

The piece of thin paper under the clear side of the disk is important. This is soft enough to prevent scratching on the clear side but firm enough that the disk is not bent or cracked by the pressure you put on it from the groundglass side.

In relation to securing the disk to the spindle, if you are using the entire spindle including the pressure plate which holds the disk on in a CD player, you don't need to fasten the disk on. (The pressure plate is either held against the disk by a soft spring or a magnetic arrangement).

The purpose is to allow the disk to run true by its own gyroscopic effect when it spins. Small imperfections or wear and tear on the spindle face don't matter too much. But with the AGUS35, if we don't hold the disk onto the spindle somehow, its simply going to fall off. When we fix a disk directly to a spindle hub and not use the spring or magnetic retainer which keeps it there in the CD player, any imperfection on the spindle face will now come into play because we have eliminated that self aligning feature.

Drilling small holes through the disk and the spindle and using screws to hold them together is an option but the precision and tools required are beyond most semi-skilled home-builders. The next best method is to use a glue of some kind. It has to be material which is not hostile to plastics and something you can pull off later if you get it wrong or the disk is damaged in the AGUS.

Even with a very tiny error, the disk will not run true which will put the spinning projection face (groundglass) very slightly in and out of focus with each rotation for the front lens and for the camcorder.

I found this creates two defects in the image.

Softness because the image is alternating between soft and sharp and the camcorder only sees soft because the disk is spinning fast. This defect causes great difficulty in getting sharp focus.
The second defect is that the image also moves slightly in a circular motion in the camcorder's view. It is only a very slight movement but enough to cause interlace artifacts in the image.

Like thin horizontal lines like highlights along a shiny rail will be very flickery.

Getting a perfect alignment with screws is next to impossible at the home-handyman level. Using glue and my method of mounting is not perfect but gives a better chance. It may not self align the first time around and has to be done again. This is why I used the water clean-up bathroom sealer. It is firm enough to hold the disk on but peels off nicely from the spindle. Once it is running right, then I add a bit more to the front of the spindle hub and disk centre to make sure it doesn't come off when doing some camera moves. The gyro effect makes the disk reluctant to suddenly change position.

Sharp vertical lines with have a feathered edge even if the camera has not been moved while shooting. It's not a big deal but if you can have the best through a little more care then you should.

This stuff above only relates to my own experiences and assumptions. For optical theory look at the Brett Erskine posts on Aldu and Agus threads.

bob,

you're an angel, i've literally a few final questions

it's all about that elusive groundglass - i've now 150 spare cd-r's, (my friends are worrying), thing is, the clear cd's were used for protection & were exposed, so there are small scratches on the surface - is there any way to remove these scratches? one has many small ones, another is clearer with one large one. i heard that an even finish was most important, so can i go ahead with these disks anyway?

i think i will go the glue route as i'm sure i'd screw the drilling up - just a standard silicone sealant will do right?

apart from that, i think i'm on way, the cd motor has tons of wires, i assume its the two that come out of the circular motor bit that i need to hook up, the colours are yellow and blue, or two blacks, anyone have any idea how these correspond to the black and red wires on my battery box?

the mechanics said they threw the ball bearings away & he's no idea when they'd get another in, so i'm thinking furniture castor is the best way to go

right, i'm off to saw circles out of mdf for my wood, gotta say, i'm enjoying doing this, a large part of my work is 3d animtion & sat in front of a computer but now i get to buy jig, soldering iron etc and get my hands dirty, lovin it ;)

Yes, welcome to the addiction, Joyita :)

Just some other thoughts - I'm using Mylar Drafting film (currently 'double matte', but will try 'single matte' too) attached to my clear cd. For me, it works great. My first adapter was a static one (no spinning) made with real glass, ground with aluminum oxide. Despite my best efforts, I could never get the surface as even an consistent as the drafting film. Not sure how easy to be to find in London, but in the U.S. any architectural supply or blueprint supply shopt has this stuff. I got two 11x14 sheets for around $2 US. I stuck it to my cd with tiny squares of double-stick tape which I spaced all around the edge of the cd. A thin line of glue would also do it.

So far, I've found that minor scratches (and even slightly less-than-minor scratches) don't matter much unless they're circular enough to look like a line when spinning. My current CD is pretty scratched and it all blurs to a nice smooth even surface when moving.

cheers jonathon

i can see myself lying awake fretting about grain and scrathes (ok, i lie, i am already), an addiction it truly is

i think i will give the drafting film a go, my local art suppliers have some, so, i'll do that and the pressed wet/dry paper route, i need an immediate fix for a film deadline, but i'm also planning on taking my cam travelling in mid-july, so i've got some time to experiment - maybe when i've got some results, we could start a thread for picture comparisons of the various methods

sticking it down round the edges, does air blowing in pose any problems(air bubbles?!), or have you sealed it round the internal hole as well?

Re the CD drive motor.

Without seeing the motor I can't help you. Most are Mabuchi style simple DC 1v to 9v, which look like little tin cans. Generally there are three motors all mounted on the back via their connecting pins to a circuit board and on front to the frame.

One is the CD disk motor. It has a hub which is on a long shaft. The other motors have shorter shafts and drive the disk tray in-out or the lazer read-head transport but are otherwise the same. I actually used a tray motor for the GG disk in the AGUS35 plumbers version because it had the shorter shaft and fitted in the enclosure. The motor shaft diameter was the same and the spindle hub slipped straight on.

The motors have two flat pins sticking out the back which solder direct to circuit board. You unsolder these to remove the motor from the board which is useless in our application, then use hookup wire onto those flat pins to connect your 3v battery supply and a switch. You'll need a solder sucker to get the motors off. Without it you'll end up cooking the little plastic insulators on back of the motors the flat pins feed through.

To get the disk motor off the frame, you will have to remove the drive spindle in order to get at the screws which pass through the frame into threaded holes in the front of the motor. Take care when remounting your motor to your own mounting plate. Don't pull the screws down firm without checking as you go by turning the motor shaft. Those screws can reach through the motor case and home down onto the armature inside and ruin it. You may have to use little washers under the screws to pack them out.

The spindle/hub is not a strong part and you need to take extreme care pulling it off otherwise it will snap off just beneath the flat flange where it joins the sleeve.

Scratches are no big deal, but you are better off without them. Choose the face which has the most scratches for the groundglass side.

For the motor mount plate, I ended up using a thin style black back CD disk box (not jewel case). I cut it down so that it would fit and attached it to the enclosure with three mounting screws which were fixed to the case first, then had extra nuts on them to hold the motor plate on. Between the motor plate and case, there is a light coil spring around each screw. This provides fine adjustment for the disk to make sure it sits flat on the focal plane. Its best to have independent adjustments for front lens backfocus and camcorder to groundglass distance adjustment. You can keep it simple and use it for backfocus but this means you have to pull the cover off to do it. Unless you are looking through the camcorder at the groundglass and seeing on a large monitor, getting the focus right is difficult.

As for the case, the purest form of the AGUS35 (analagous to the DOGMA95 movement) is the plastic CD-R case. There are some drawbacks to using it. Its a bit light and the disk to case clearance is a bit close for unskilled artisans to manage. I used larger diameter sewer caps but this also has drawbacks if one later adds the porroprism (or mirror) image erector when it is sorted. The project-box versions may lend themselves better to having this added.

The images on www.dvinfo.net/media/hart titled mtatk2f1.jpg or similar give an idea of how good the pressed disk can be. There are other variables of course, the most significant being the relay lens from disk to camcorder. Brett Erskine's posts are the most helpful here.

Good luck.

You'll really know you're addicted once you get to the "dangit-I-wish-I-could-just-get-back-to-shooting-and-stop-fiddling-with-all-these-tools-and-lenses-and-stuff" point. And you won't start shooting because you *really* want to shoot with your adapter, but its not *quite* good enough yet and you know there's just that ONE other thing you could do, and...

Oh wait, we're talking about you, not me :)

On the film, I cut out the little hole in the middle with my Xacto - so I end up with a disc of film with a hole that looks just like a CD. I just pressed hard on the tape - didn't do any air drying. If gluing, I'd probably clamp it - or put it under some books instead of hot-air drying. Be sure that whatever you use for the adhesive, you only do it on the very outer edge and right around the center hole. You need to keep the main radial part of the CD clear.

Bob's right on -

On my motor, I used a piece of thin but stiff plywood - my motor (the little tin can kind) has an annoying little nipple on the bottom which prevents just setting it on a flat surface, so I drilled a hole in my plywood which this could set in, and then I used JB-Weld (amazing stuff) to cement the motor to the plywood. It's rock-solid and completely flat. I then drilled four holes and used four bolts.

The most important part of this contraption (as Bob says) is to be able to very finely adjust the distance from your SLR lens to the spinning CD. Seriously - a single millimeter makes a big difference. You need a good way to make very fine height adjustments and then be able to lock it hard once its correct. You won't need to make further adjustments once its right, but it takes some careful work to get it just right. I'm off to work but will post the system I used... not the best but it certainly worked. The focus marks on my SLR lens line up exactly.

"The Magnificent Obsession". - The quest for the perfect AGUS35.

We are all engaged in the ultimate undercapitalised risk dispersed research and development exercise. One is always looking for that little bit better here that little less awkward to use there.

I console myself with having done a one-shot live music video which I could have shot with the camcorder direct. I have entered it into a local film-fest as an experimental work.

Strive on.

As for the drafting film. Another method might be to fix it between two clear disks and fixing the two together with glue on the outer edge. This might be flawed through introducing two more reflective surfaces into the mix.

Hey Bob,

When you said :
"I tried placing the disk on a flat hard surface on top of a piece of thin paper. I placed a piece of wed & dry 600 grade silicon carbide paper face down on the disk. I did not rub the disk but instead put extreme pressure on the back of the paper to stamp pits into the disk by using the round end of a metal scriber handle to rub the back of the paper itself. It is essential for the wet & dry paper not to skid on the disk otherwise scratches, not pits are made in the disk. The back of the wet & dry paper wears out fairly quickly. A better method is to get hold of an old complete ballbearing from an automotive repair shop. All that matters is that it still turns freely. Hold the centre (axle) part and roll the outer rim of the bearing on the wet & dry paper with lots of pressure. To get the loca pressure through the wet & dry, you need to tilt the bearing slightly so that the rounded corner of the outer rim runs on the wet and dry paper."

What about using a rolling pin instead? Is this what you term, 'Pressed Glass discs"?

Did you find a better grade of oxide after doing this test for experimental-use discs?

With a rolling pin, you won't get enough localised contact pressure. The furniture castor method should work.

For experimentation using other methods like rubbing with sandpaper or rubbing the glass around on a sheet of glass in a slurry of jif abrasive cleanser, silvo cleanser or whatever, use some ordinary CD-Rs. Place them label side down on a sheet of glass in the slurry and rub them about. The white coat will come off first or printed lable, then the silver layer will come off. Once that is off, you are down to your groundglass surface.

Whilst this will be a pretty crappy groundglass, it will suffice for setting up backfocus and trying the principle until you can get something better. A CD-R disk with the silver layer removed is quite adequate and optically true. The only downside with some specimens is a slight green cast to the colour balance which may be corrected by manual white balance in the camcorder or colour correction in post.

As for clear disks, you will find one each in packs of "Lazer" brand DVD+Rs, possibly DVD-Rs as well. These are diversion from actual production as they have guide tracks on them.

You could delaminate (split and peel) early DVD+Rs
then polish off the purple layer for a gg but the latest ones seem to be made of much sterner stuff and seem to have a surface treatment to harden them. These are harder to dress in AO5 as a groundglass.

Ok, I went ahead with the rolling pin idea tonight and it was alot of work, but I got some GG out of it, but still not opaque method? Right now I am using 180 LUBE paper Diamond Grit (this is what is says on the back). IT does a pretty good job but it hard going. I have to give credit to the people that have come up with decent GG's - they are hard work!

Thor Labs Diffusers
 

Has anyone tried these? They look like they'd be perfect, and they're pretty cheap:

http://www.thorlabs.com/NewGroupPage...tGroup_ID=1132

I saw the original link on some other thread here, but I've been meaning to try one out.

Might save you some elbow grease, Mandy-

Yep, those have been used :)

They provided decent results for static designs I believe. They still show grain of course.

I knew that others had tried them, but I could never figure out which grit they used. I read where some people were planning to _buy_ the 1500, but then I never heard a follow-up. Do you know who actually bought one? I'll search a bit more and see if I can dig it up...

I think (not sure) that Mandy's making a static design -- so this might be a good direction.

No, I am using the CD motor - spinning CD. Static might end up being a viable option depending on how well I can make the end product look - but I have to try this way first because I have come too far now.

Ah -- okay, my bad. You'll be much happier with the spinner (I was), and in that case the THOR products won't help you unless you move things elliptically.

What Thor products? Have you finished your design? Do you have a url for it, would love to check it or even some video/pics that you have taken with it.

THOR Labs -- they sell pre-made laser-etched fine-grained diffusors (i.e. ground glass).

Yeah - there's some not-very-finished stuff on my website:

http://www.aqua-web.com in the film section.

Jonathon, great looking site! I see you went with the project box design - can you tell me how the heck you manage to calibrate the Focal Length AND got it to stay in the project box? I have been racking my brain over this one since I start this project - aside from having a hard time get the right focal lenght for my Sigma Zoom II, attaching into the project box has become sort of a mystery on how you got it in, screwed it into place AND then calibrated everything!?!?!?

AND what is the frosted CD? I am thinking that it is homemade?

The footage looks great for a VHS-C - colours are sharp as well. Great job!

Yeah the camcorder is certainly the weak link :-( Someday I'll upgrade, and I'm sure I'll have to improve the adapter at that point. It's just a 'hobby' after all for me.

I mounted the CD motor into a square piece of plywood. This is attached by four bolts to another piece of plywood which is affixed to the back of my project box. There are 'spacer' bolts in between the two pieces of plywood... I used a focus chart at a precisely measured distance at the lowest end of my SLR lens and focused the lens to this. Then... I adjusted those four spacer nuts and moved the plywood holding the cd motor until I had a focused image, according to my view of the GG through a loupe.

Theoretically this would allow precise focusing and also allow for correction of slight off-axis 'warble' to the disc. This is a big issue which is hard to pick up on... if your disc is not spinning completely flat, the GG is actually moving in and out slightly -- which causes softness in the focus.

In reality, it was a pain in the ass, and it ended up slightly off anyway. In my _next_ adapter :) I'm going to make the focal flange adjustment and lockdown something that can easily be done from outside the box. My idea worked, but it's not great. I'd advise you to think of other alternatives.

Do realize that you have a small margin of error here... and what you trade off for that error is one end or the other of your focal length and accuracy of the marks. In other words, if you're off by a bit in your flange distance, it will likely still work, but when you're lens says you're in focus at 5m, you may actually be in focus at 7m.

On my GG -- I used a piece of single-ply mylar architectural film glued to the cd. With my low-res cam, it worked fine, and saved tons of time. A sheet of this stuff is like $2 at any architectural/blueprint supply shop. It will need improvement once I have a higher-res camcorder, but for now it works great... plus, once you move the GG - your problem is less one of grain, and more one of flicker and oscillation error (as mentioned above).

Hope this helps...

Another tip: look on ebay for viewfinders for older pro-cameras. I have an Ikegami viewfinder which I bought for $9. It has a beautifully sharp CRT image (albeit b/w which I prefer) and I can flip the whole thing upside down to fix the image rotation problem if need be... or maybe mount it to a helmet (!) for a steady-cam-like project I'm working on. I had to dig a little to get a pinout to hook it up, but it's a good solution for image rotation.

Jonathon,

My own personal steady-cam is on hiatus right now. I started building it in the summer - got the gimbal and yoke finished, but stalled on the arm (the really tough part).

Ok, back to the DOF machine.

I have done it exactly the same way, except ( and a big one ) I don't have a focus chart for this lens. I got it at a charity shop, so nothing with it to help me out - just a Sigma Zoom II, 35-105mm (done the search on Google and that was no help). So I have had to adjust it by eyeing it - and it is an absolute chore because of the lack-luster macro lens I set up in series with a magnifying lens. What I tried to do was set the Mark to Inifinity and then the lens at 105mm so it was as close as it could go - then I made the adjustments - but getting it right is so darn hard!

Also, figuring out where to drill the hole in the plastic so it is lined up with the GG and the SLR is lined up as well is a nightmare! The Project boxes are only $10.00, but I don't like to waste money! How did you get over this hurdle?


I use the magnet trick to fix the upside down image, so that doesn't really bother me.

I love the footage you had - the colours were soooo bright and sharp, but having said this, I did notice the wobbling in the upper left hand corner. But it did not detract from the nice image at all.

BTW, what does "according to my view of the GG through a loupe" mean?

A loupe is a little magnifier. large-format photographers use them on a ground glass for those big cameras to check focus. I used (and got) mine while I was a typesetter. I magnified printed materials to check the halftoning, etc. Anyways...

If possible, take the macro out while setting the focal flange length. It just complicates things. In other words, take the camcorder off, and the macro out, and try to look directly at the ground glass with nothing else but your naked eye (or a loupe if you have one). You don't really need a focus chart, but do find something with lots of fine detail... you can open up a book with small type -- that works well. A focus chart is just easier to use (there's plenty for download/print online... such as: http://www.creativevideo.co.uk/refra...back_focus.htm) The idea behind the star is as you focus, the center will blur and sharpen. You focus until the sharpness extends the farthest into the center.

It's _much_ easier to focus to the short end than to inifinity. I don't know much about your Sigma lens... but crank it down to the nearest focal setting and place the book at exactly that distance (measure it with a tape measure) from the front of the SLR lens. Lock all of this down if you can :) Then look directly at the ground glass from behind (as if you were the camcorder) and adjust the focal flange distance until the words on the page of the book are as sharp as they can get. I found it helped to put plenty of light on the book, but to have the rest of the room (particular back where I was) much darker.

The way I handled the hole drilling: I figured it out on paper -- and used Adobe Illustrator to make and print myself out an exactly measured template, with the hole exactly where I wanted it. I taped this right onto my project box and drilled right through the paper template. It worked well - got it in the right spot on the first try.

Ideally, (another TODO for my next adapter), you'd have x-y translation on your GG as well, so you could make slight left right up down adjustments after everything was in there...

One other comment -- my notes earlier about the 'leeway' you have on your focal flange depth don't apply if you're using a zoom. For a prime, you can usually find a spot which is in focus but may not line up with the marks. But for a zoom - if your focal flange distance is wrong, than you'll find that the lens will be in focus on one end of the zoom and out of focus on the other. If your FFL distance is correct, this won't happen - and you're best approach to focusing will be to zoom in, focus, then zoom back out. Worthless unless your focal flange distance is exactly right. :)

Jonathon,

Thanks for the information - that will help out alot. I am going to try to do that adjustment tonight. An X-Y adjustment would be a great idea as well - because even though it is a small area, once you drill those holes in the housing, there is no going back - so adjusting the GG up or down would be boon!

The loupe, could this be substituted with a magnifying glass?

Do you think your CD wobbles because it is not optically true or do you think it is the motor or the casing or a combination? I don't seem to have the same problem, but then again, I don't get the nice colours you get as well.

As far as focusing down on the closest point - I am used to working in a TV studio, and we would always zoom in and focus to infinity, and then zoom out to make sure that everything was in focus - so that is new to me.

I am going to try the book/fine print thing tonight - to infinity and see what happens and then I will try the close (as you suggested) and see what happens.

Nope - my 'wobble' comes from my mounting being slightly out of whack... such is the crude mechanism I have for adjustment. The sure sign... with the 'lid' off, I turn on my CD and as it spins if I edge a piece of paper up to it ever so slowly, rather than a steady HIIISSSSSSS, I get PHHT PHHHT PHHT PPHHHT as the 'high point' of the CD (from my mount) makes contact. Not good. The next one'll be better (and the one after that, and the...)

And yes, a magnifying glass would be fine for a loupe -- better probably, because a loupe is really supposed to rest against the surface it magnifies. Not good for a spinning disc!

With the magnifying glass, it is still very difficult to make out anything clearly. With the DV (I know, I know - I will try without it tomorrow during the day when I have light) I found this out, interestingly enough...

When I pull my zoom out to 105mm, on infinity - the 'STAR' (pdf from the link) picture was out of focus, even though at 35mm at infinity, it looked good. If I turn the rubberized barrel to 12 (# in beige on the barrel) inbetween the 3 and 4 (#'s in green), it comes into focus. So does that mean :

1) that is needs to be recalibrated to make sure it focuses on infinity instead of on 12 (in between the 3 & 4 on the rubberized barrel)
2) That the target is not far enough away (it is about 3 meters at present).

I think it might be the former (1) because when I moved the camera back, the STAR still did not come into focus, but I could be wrong.

I wonder how your mounting went out of whack?

Forgive me if some of this comes off as brain-dead stupid and it sounds like I'm patronizing or talking below you... I'm not, I just want to explain myself clearly, and perhaps this will help. You probably already know _all_ of this, but...

Yes, using any kind of magnifier on the regular LCD screen of your camcorder won't help much - you'll just see very big pixels. This is the real advantage of removing the camcorder and looking directly at the ground glass with your eye... your eye has much higher 'resolution' than your LCD monitor :) and once you can magnify that ground glass image with a loupe or even a plain old magnifying glass, you'll really be able to see whether the image is in focus or not.

If it were me, I would take the following approach:

Zoom your SLR lens to 105mm (it's longest setting?) and leave it there.

Base your focus adjustments on the marks on the lens. Here's where I'm most fuzzy... I also have a zoom lens for my SLR - it's a cheap 80-200, but it may shed some light. I have two different rotating rings:

Ring 1 is the aperture, with numbers 3.9, 5.6, 8, 11, 16, 22 - which represent the F-Stops for the size of the aperture opening. It's the ring closes to the mount.

2) is a big barrel combination focus/zoom. The entire barrel slides up and down the lens and can rotate as well. Sliding it up and back changes the zoom, (which is marked down on the barrel which gets exposed as I shorten the zoom). Rotating it adjusts focus. I have _two_ sets of numbers on the ring itself, as it sounds like you do too. In my case, the two sets of numbers (one in white, one in red) represent the focal distances in meters and feet, respectively. I wouldn't be surprised if yours were similar. There will be an infinity sign at one end. Turn the lens to the farthest point opposite inifinity and find the number. This represents the nearest point your lens can focus. In my case, this is 1.3 meters.

3) Put your star chart on some kind of stand, perpendicular to the camera _EXACTLY_ this distance away. For my zoom lens, this distance would be 1.3 meters.

Now... Leave everything exactly the way it is, and move _only_ the adjustment on your focal flange distance (in my case, this is the nuts between the two pieces of plywood) until the star is _most_ in focus like this:

Theoretically, you won't be able to resolve all the way to the exact center of the star. This means that at some point outside of the center of the star, there is a blurred circle. As you adjust your focal flange depth, this circle will grow and shrink. The goal is to make this center blurred circle as small as possible. Once you're reached that point, then you should be pretty close to having your focal flange depth correct.

In reality, this measurement is measured in tenths of millimeters (For my Canon FD mount, the spec'd focal flange depth is 42.1 mm) (!). It's awfully hard to adjust 20 thread-per-inch nuts at a tenth of a millimeter precision. So just get as close as you can.

Now -- if you get all that right, then theoretically, you should be able to simply turn the lens to a number, say 5 meters... put the target at exactly 5 meters and it should be in perfect focus. If you're really good, then you should be able to get in focus, and then pull your zoom to any point (without twisting it) and it will stay in focus.

Much of the focusing that happens in larger movie productions is done by tape measure... not by eye. (Eye is used to check of course, but generally, it's the numbers on the AC's/Focus Puller's tape measure that determine the position of the dial). This is why the focus marks matter.

See my new post on infinity - but I think you have solved it.

Thanks Jonathon, I am going to try that next. No, it did not sound patronizing and I appreciate the time and effort your put into it. What you say makes sense 1.3 meters - focus, should focus on everything. My barrel and zoom lens sounds exactly the same as yours and works in the same way - thanks for clarifying what the marks on the ring mean.
I am going to have to try and lock down my GG for this and get a decent tape measure because I do want to get as exact as possible.

Also, I just finished watching your 'rain' video - and I have to say it looks soooo much like film that I cannot believe you are disappointed with your device!! The colours are great and the movement, DOF and grain look amazing. A great job.

Kudos, Jonathon!
 

Wow, first time I saw your site - amazing stuff!

Great stuff on your site too about avisynth and virtualdub - I'm using other means for this part (Magic Bullet and deinterlacing in Premiere Pro) - wow, your depth of field stuff is unreal...

Whoa!

Okay, okay, I am definitely building this thing.

Mandy/Jonathon - thanks for sharing your findings! Amazing!

Thanks for the nice words --

Mandy, I'm not at all unhappy with the current adapter... particularly for my VHS-C cam, I just have certain things that I wish were better. It's the curse of these things... there's always something that would be 'a bit better if I only...'

Mark, be careful! Once you get hooked, you'll find yourself perusing weird optical/hardware web sites at 3am the night before a big 8am meeting at work... neglecting your significant others... breaking your piggy bank to buy some aluminum angle iron... and noticing things that other people are throwing out as garbage, thinking, "Hey... I could use that for..."... you know -- all the classic symptoms of an addiction (like Crack, just more constructive). Regardless, it's certainly fun!!

How do you like Magic Bullet? I've never actually used it, but their demo videos certainly look nice. I've always been happy 'enough' with my avisynth tricks to not shell out any $$ for a more commercial solution, but I always worry that there's some magic in the commercial approaches that I can't replicate... (way off topic now... this thread is about the frosted ground glass, is it not) :)

I will have to jump in with the addiction thing. Hey, this thing gets out of hand on the quest to make the perfect DOF machine or whatever. That is why I ask question after question, even if there are silly simple stupid questions!

Working on the Focal length today and getting tracing paper and maybe, if I can find it - architectural paper as well!

(OT - I have used Magic Bullet in the past - waaaaay too slow for rendering times). Most NLE's have enough plugin that come with the program that you don't need it - colour curves, colour correction, grain etc)

Jon and Mandy - we're addicts
 

Okay, you two have seduced me to cross over to the "dark side." Finally got my SLR off my kid sister because it has a nice 28-105mm zoom lens I'm going to test with... tomorrow, it's the macro/c.u. (close-up) adaptor - I'm gonna test with the DVX100 I own but want to really build this for the new Sony HDV cam I've rented - the FX1/Z1... Amazing low light so with this system, it should be a real hit!

Guys, what macro size is the best? +2?

Also have to buy some cheapo CD player at Walmart. Jon, great pics on your site - the architectural paper looks like a hit.

Mandy's right - rendering slow on Magic Bullet. Your stuff is tremendous - really, I'm an expert at watching all this stuff on the web and I can tell - you've nailed the filmlook stuff perfectly with your avisynth solution.

VHS-C? Amazing!

So I'll be starting tomorrow, folks... Will keep you posted!

Mark

Just to be different - GLAD press'n'seal
 

Guys, because I have it here at home and it's almost self-adhesive, I'm gonna test out with GLAD "Press'n'Seal" wax-paper like film to make my clear CD into a "GG."

Just thought you might want to try this - there is logo etc. on roll of stuff but you can get an area large enough without logo to place over CD...

will know more once I actually get it spinning - but it's diffuse and just looking through lens at it in dark room, upside down reverse image appears nicely on the CD....

More later.

Cheers.

Mark keep me posted on the Press and Seal stuff. I went out tonight and hit all the computers stores looking for packing CD that were clear without the green tint. Came up with a few. I will check into the Glad Press and Seal material if it looks good for you!