Homemade 35mm Adapter

[Bob Hart]

I am a little confused.

Those are not any images I have posted anywhere. I may have also confused others with my post above.

In terms of sharpness, I should have used the words, "seemed as apparently as sharp as". This is because the footage was transcoded to MiniDV and letterboxed from the camera playback and dubbed straight to a DVD recorder.

The disk from this was played into a video projector so it is not a path which would show differences in sharpness at HDV resolution.

At its best it seems apparently as sharp. The flashing effect of thin bright highlights against dark backgrounds which happen when the camera is handheld on a fixed subject mostly goes away with relayed imaging.

[Bill Porter]

Bob, we're talking about Agus' pics.

[Bob Hart]

Slightly off-topic but references a few posts back in this thread is a low-no budget feature project titled "The Cage" which I have lent a bit of a helping hand to by holding the odd boom pole now and then and shooting some background extras footage for a "Special Features" for their intended DVD-Video release.

For those who may be interested, they have just posted a trailer which was prepared for the wrap party and this apparently is located at the following web address.

http://www.youtube.com/watch?v=hSPQnV2vH8I

I don't know if this here is a clickable link. I don't have the knowing of all that haitch ref stuff which makes the web addresses switch through.

[Bob Hart]

I am enquiring with Ohara in Japan to see if they are still making the cut blanks of CD-R sized optical glass disks with 15mm center hole, 1.3mm unfinished thickness.

Is there anyone else making their own from unfinished blanks?

[Wayne Kinney]

Bob,

Yes, im having 1.2mm thick toughened glass blanks made. 6mm centre hole, 86mm outer diameter.

[Bob Hart]

Wayne.

If it is not a trade secret best kept, can you advise per unit cost of your raw disks? Would your source be able to make to standard CD-R disk size.

Have you had any problem with the groundglass finish causing outer circumference stress risers in the glass. My disks came with chamfered edges which are apparently a necessary part of the process. I found I had to maintain the chamfers as the disk wore down to 0.9mm in figuring and dressing.

The disks also ring like a bell once they are dressed.

The Ohara disks are S-BSL7, outer and inner circumferences concentric, (they make glass substrates for hard disk drives) capable of a fine GG finish but were AU$50 per unit to Australia in 2004. I am awaiting advice on availablility.

My original enquiry to Ohara was for finished hard drive substrates which would have been ideal and by coincidence, closer to your disk size but for one small issue, the glass is not transparent.

In the end I decided to go with the original CD-R size to keep things simple and to allow higher shutter speeds relative to disk rpm, but Ohara could have supplied smaller had I requested then in 2004.

[Wayne Kinney]

Bob,

If you can email me at info@sgpro.co.uk, maybe we can sort something out.

[Bob Hart]

If anyone is using the tumbler method of making their own groundglass disks, there is a reasonable fix for the uneven finish problem. It seems the glass disk on which the groundglass disk is ground against, tends to pick up and rotate which messes up the continuiuty of the cycle.

So far with the first AO300 stage where raw disks are figured into shape and smoothness, a piece of automotive tyre inner tube, cut to the diameter of the tumbler barrel, placed between the grinding glass and the bottom of the tumbler barrel seems to cause enough grip to stop the problem.

If it recurs with the finer grades which pick up with more force, I shall try a piece of a non-slip bathmat, you know the ones with the multitude of suckers on them.

Repeat of info.

For tumbler barrel 160mm internal diameter.
Disk diameter 120mm approx.

For Grind - barrel is best at 60rpm.

For Polish on felt, - barrel is best at approx 90rpm. A little faster is okay as long as the platen does not start to swing or bounce inside.

Angle of inclination aprox 45 degrees.

If you make drive belts out of automotive inner tube, put one internal twist in them, (one end inside out not simply crossed over). they run longer and better.

[Bob Hart]

Furthur on the tumbler method.

Best results after stabilising the glass grinding disk seem to come from tilting the barrel up to about 30 degrees off horizontal and putting more slurry in which seems to take a little weight off the platen but it doesn't jump about.

Results so far with the AO5 micron grind seem to be that the groundglass finish is denser than previously, flicker problem is gone but resolution is not as good.

Aperture of f8 can be used on the ultrawide lenses in low contrast lighting conditions without flicker or groundglass artifacts. The disk figuring from raw disks is also better.

The glass grinding disks turn out to cause another problem. The grind now goes out uniformly to the edge of the optical disk whereas previously with the shallower tilt, most of the grind seemed to be happening in the centre. This now causes a distinct ridge to build on the edge of the grinding suface where the optical disk being dressed does not touch it.

When separating the two glass disks at completion of a stage, care has to be taken not to allow the ridge to brush across the ground finish as it scratches or sweeps a wider area clean of groundglass texture.

[Alex Chong]

Hi Bob,

Great to read that you are still working on GG preparation using the tumbler system. But I am very curious, have you looked into using focusing screen as a GG as others have done? Or you are onto something here with the tumbler and grinding system. I felt its just too much work to prepare a GG from rough cut round glass to polish and then to GG. I have been meaning to ask you this a long time ago but felt I kept it long enough and my curiosity just wouldn't keep me quiet. Sorry in advance for being so nosy.

I understand you have done a mini35 of your own, any chance of seeing some pics and video?

[Bob Hart]

Alex.

The home grinding of disks happened because I could not get CD-R sized disks which were glass, thin enough, with a correctly centred 15mm hole, already finished.

At the time, the Ohara disks from Japan were the only source I could find, though I have since been advised Knight Optical in the UK can do them complete.

As I have learned things, most of the glass disks have had several re-works done on them.

Because I have used full CD-R sized disks and my design is built around them, I have not explored the focussing screens.

There is I think a .jpg of the non-erecting and erecting versions of my sewer pipe based device as www.dvinfo.net/media/hart

This folder is in bad need of housekeeping. The files there are a non-systematic accummulation from when I first started experimenting with the idea. I'll have a look there and see if I can find the .jpg name and edit this post.

[Bob Hart]

Alex.

Dvinfo edit timed out and I could not correct the post above.

I couldn't find a .jpg of the erecting "flip" version. There is an image of the earlier version at this address -

http://www.dvinfo.net/media/hart/Aguscam.jpg


Don't attempt to build to any of my .pdf designs. They should be regarded as conceptual only as things have moved on since they were posted.

[Alex Chong]

Bob,

thanks for the direction. Checked out the files. I liked the picture with the oranges on the branch. The color is nice. Not too sharp nor soft. But on other pictures, they are a bit soft and have some hazy feel probably due to light reflecting off inside the Agus35. Not sure how to avoid this though. But generally, the image is nice. Saw the tumbler. Wow, thats alot of work creating the thing. And it really spin very fast too.

Personally, I have at this moment given up the idea of using spinning disk as I couldn't get the right materials to even start. So now I am working on something along the line of Letus35. Design is done and just need to send off to the CNC shop for prototype fabrication. Hopefully can showcase it here in a couple of months time. Using focusing screen, flipped and has rail supports. Won't say anymore and I will wait until its in my hand and I have tested it with some video sample. If nothing great comes out of this, I will shelf the whole idea and buy one thats in the market now. Fingers crossed.

[Bob Hart]

Alex.

The oranges images are probably closest to where my current efforts are at. That particular groundglass disk was the very best I ever did and I have not been able to replicate it since although the last one I have just done comes close. The original got broken at an airshow over a year ago.

Some of the "smoky" images I think were the ones shot through a stack of Hoya close-up lenses 1+, 2+, & 4+ , which yields an inferior image due to bad chomatic distortion. These lenses were never intended to be abused in such a manner.

If your design is close to Quyens in physical size and groundglass image size, my guess and it is only a wild guess, is that you would need something like a +5 achromatic dioptre on front of a camcorder zoom lens setting of between 45mm and 55mm.

Going by the size of his mirrors, the focal point of a +7 might fall short of the distance between the GG and the front of camcorder to get a sharp focus on the GG.

Quyen's implementation is a valid one and the design inherently dampens against noise and vibration. If you have access to CAD design and machining, then you are fortunate.

Interestingly, Quyen's flip design uses larger mirrors than my experiments. It seems he has done this to enable the mirror paths to be oriented at 45 degrees to preserve the prime lens and relay optical centres on a vertical centerline, whereas mine is offset low and to left.

The groundglass is much smaller than a CD-R disk sized groundglass and represents good economic use of materials.

My personal preference would be for the groundglass excursion to be more like 3mm to de-resolve any dust specks which get onto the GG but that would also shake the crap out of any adaptor or cam attached.

The excursion must be orbital, not linear. If linear there will be periodic frames where the groundglass texture becomes visible due to the frames syncing up with the halt at reversal of direction.

While it is very inconvenient to design in, you need to make sure you have facility to accurately set the correct backfocus for the SLR lens on front and that the groundglass is perfectly aligned at right angle across the optical centerline or can be adjusted to this accuracy.

Its not so critical for 50mm or longer focal lengths. However for wides and ultra-wides, if the backfocus or alignment is off, these lenses will not perform at wide apertures or have a soft area somewhere in the image which will not sharpen with moving the lens focus ring.

Some of the modern lenses which are full film frame but also optimised for digital SLRs seem to have better performance related to light distribution across the GG.

The Sigma 14mm f2.8 seems to stop tighter to f8 - f11 without showing the grain in certain lighting conditions. But this might also be my imagination.


Good luck with your endeavours. If ever you encounter a legal practitioner in Brunei by name of Colin Clarke, please pass on my regards.

[Bob Hart]

Another little best practice issue has arisen with using platens in a tumbler and wax to mount the raw disk to them.

There is enough flex in the glass disks, especially after they have been partially processed and have become thinner, they will distort between two mountings and no longer present an entire surface contact with the grinding surface.

If the disk is unmounted then remounted before the next stage of finer grit is done, there will be caused a variable density of texture around the disk. It is enough to cause a flicker problem.

Best practice therefore is to keep the disk mounted on its platen and change the grade of the grinding material, rather than swapping several disks across a single grade. I have only one platen. The obvious solution is to make a few more.

Another handy trick for reducing wax contamination of the dressed surface after boiling the disk off the platen and lifting the disk off, is to fill a sink with hot water and place the boiled pot into the sink and allow the hot water to overflow into the pot.

Slowly run cold water into the sink. Presently, the wax on the surface will set in thin sheets and can be skimmed off. As the water cools, the disk can then be brought out without wax particles setting on the ground surface. There will still be some but solvent based stain remover like Preen will lift most of it off.

I tried making another wax composite disk, this time by placing two ground disks face to face with wax in between and relying on the depth of the grind itself to determine the thickness of the wax. Again, the image is by far and away the sharpest but the variable density problem is still there.

The groundglass grain remains, but the evident contrast of the grain seems to be diminished. It seems to take another f stop of setting on the SLR lens to make it as evident but there was no science involved in this observation.