Thursday, October 25, 2007

Results

Check it out!

OK - its been 3 years since I created this guide and I have finished every section but the "Results". I intend to add samples (videos) of my most recent work soon . . . promise. :)
Probably before I defend my PhD thesis . . . right.

Please let me know if you have developed your own scanner based on my information and how things work for you.

Thanks,
Kyle

12 comments:

wise said...

I bought an Epson v500 when they first came out to try telecine. I even thought of using a cd case for guides but deaths in the family put it on the back burner. I am so glad to find this blog. I envisioned setting the guides vertical to get more frames and the scanner has to go the distance anyway. If I understand correctly, you went horizontal to provide more stability. I was going to use the glass from an old scanner to weigh down the film. I do not know why you got Newton rings and wonder if I will have that problem with my glass? Other than that, the main obstacle will be finding an old hard drive with a stepper. Great work.

Kyle Brunner said...

I chose horizontal because the speed of movement of the scanner optics in the vertical direction is constant regardless of the width of the scan at 4800dpi. This means that in the time it takes to scan one frame vertically, I can scan 18 frames horizontally. I had thought of trying to scan vertically, but the speed issue was most important for me. In addition, a horizontal scan produces a 150 MB image file and it is not unreasonable to expect a vertical scan to produce an image upwards of 1 GB - far too large for my original computer to handle very easily with the available memory.

Glad you enjoyed the blog.

Kyle Brunner said...

About the Newton rings - they occur when light travels through different mediums of different densities i.e. through glass to air to glass to air to film to air to glass to air to glass. The reflective surfaces of the film in the lighter portions make this a problem (dark areas of film are not a problem). Anti-Newton ring glass distributes the interference so that there are no discernable patterns (which is actually the problem - we don't notice random color unless there is a pattern we can see).

wise said...

Arrg! I really like your setup but the transparency unit window on my v500 photo is 3.25 x 10.50 inches so I have to go vertical. Your pics show your old 4870 and unless epson made different scanners and called them the same model, I don't know what you did to stay horizontal other than using your old transparency unit on your new v500. I could build a mount to turn my transparency unit sidways but I would have to make an extention for its plugin to the scanner. Vertical is easier.

KP4M4s have turned to gold...I found one but they wanted $80 for a used one. And those old full height floppy drives are more on ebay. I have been learning a lot about stepper motors so if I can't find any at the swap meets I will have to use another kind.

I have learned alot about Newton rings also. This is going to take some time, but if you like, I will keep you posted. BTW, how have you handled the film cleaning?

Kyle Brunner said...

I have the v500 and I go horizontal. It means only capturing 18 frames (r8) at a time, but I found that with my old scanner I only wanted to get 18-24 frames at once anyway to avoid image shifting during playback, only minor losses in scanning width.

As for stepper motors, there are many on the market and none of them are cheap. I was lucky to be at BYU and have an electronics lab full of old junk that they gave me. I cannot get any more, unfortunately. I have referenced a coupleof othe DIY film scanning pages where the advancing mechanism was an old printer, or a Lego gear kit. These might be acceptable alternatives.

Keep me posted.

As forcleaning, I used to buy $30 4oz bottle of dichloroethane on eBay made by Kodak and specifically called Film Cleaner. They are scarce, expensive, smelly, potentially carcinogenic, and bad for the ozone. I like how hood a job it does of removing everything that doesn't belong, but except forthevery oldest film (or particularly dirty/damaged film) I found that cut up old whitecotton T shirts clean just as well with, or withoutthe cleaning solution. Now I clean with the dry cloth only unless needed. As for the scanner and glass, it is much the same story. I used to worry about getting the glass spotless but ut just isn't worth it. In horizontal orientation, the film only touches 3-6 inches of glass by 1/3 inch tall and that is all I bother cleaning now. I used to use Windex every time, but I found that after I got rid of the fingerprints, the same typeof clean cotton T shirt was just as good dry. I also worried about residual water and amonia being picked up by the film and causing the film to degrade faster. Now I just use a can if air to blow off any dust and if I see smudges I use a cloth. I clean the AN glass top and inside surfaces everytime, but only clean the AN glass-scanner bed surfaces when needed.

Kyle Brunner said...

From the last post I am sure you can guess that I almost always reply using my iPod Touch - hence the weird letter inversions and run together words.

wise said...

Thanks, I'm back to horizontal. I thought that since you were using a glass cover 6 inches long, that you were capturing from 4.5-5.0 inches. I did the math on my transperancy cover of 3.27 inches wide at 3.81mm per frame I can do 21 frames. :^) With an auto advance 18 makes a lot of sense.
Thanks again.

I have to end with a question,do you scan emulsion up?

Kyle Brunner said...

You can get shorter glass from the supplier I linked to, but if I remember correctly, it turns out to be the same price. I am of two minds about this:
1. shorter glass means less film-glass contact (a good thing)
2. shorter glass means more guide engineering (not too hard, but still more work at this point)

If you use Epsonscan software as I do then you will find that the software clips some of the transparency area. I max out at about 18 frames, but it doesn't bother me for the reasons stated. At first, I too was also dissapointed with the little I was about to get in 1 pass.

As for emulsion side up - I am not sure right here on the spot. I can tell you for certain that I feed the film from the right side of the scanner to the left (as in the images and movies on the blog) and that the sprockets are always at the top of the scan (also as illustrated on the blog). With this orientation, the resulting scans are in the correct orientation after rotation 90 degrees (i.e. any words captured on film are read correctly on the resulting frames). By specifying right to left feed and the location of the sprockets, the emulsion side up/down question should be answered (I just don't know what the answer is).

Egberto said...

How will I nominte Richard and Jim? Eightrithms lions? Both of you have given not only a lesson class for readers, but a hard hit on the 8mm picture industry. That's what they lose money, they don't think as people; nowadays and the future. They've given up the technology associated to analogic. The apparatus you made should be sold in any shop.
See my comment on 8mm at www.scribd.com/egfior

Thanks,

Egberto F Ribeiro
egfior@hotmail.com

Kyle Brunner said...

The only links I have for Richard and Jim are on the right side of the blog.

monks19 said...

is there any custom made software that do the sme thing like yours ? Because i'm a total dummy for programming, so... and it will save some times too. By the way, grat blog. Very instructive on the process. But will it be possible to show how exactly you made your software for authomatic the scanning process ?

Thanks to answer.

Michel Rivest
rivest_michel@yahoo.ca

monks19 said...

is there any custom made software that do the sme thing like yours ? Because i'm a total dummy for programming, so... and it will save some times too. By the way, grat blog. Very instructive on the process. But will it be possible to show how exactly you made your software for authomatic the scanning process ?

Thanks to answer.

Michel Rivest
rivest_michel@yahoo.ca