First Stack

[bugbear]

Using my CHDK script to traverse focus in my Canon A630, and CombineZM under Wine/Linux I was able to make a stacked image.

The (calculated) intervals are 0.1mm, although the EXIF focus distances are rather different from this (there's an interesting relationship between where the camera thinks it's focusing, and where it's actually focusing with a reversed lens on the front...)



(all other things as per my first thread)

BugBear

[rjlittlefield]

Sorry, I missed this one earlier.

It seems to have come out well, except for color balance.

--Rik

[DaveW]

The only change I would have made was to use a different coloured and more neutral background. You want to enhance the subject being photographed, so try not to have anything competitive with it or drawing the eye away from the principle subject.

The background should do just that, fade away into the background not attract your eye. If you don't notice the background but only the subject, then that is an ideal background.

DaveW

[bugbear]

Thank you both for your comments. Since the (only) goal of this shot was to check my technical set up (ability to capture and stack) I can only regard as positive the fact that both your comments are on artistic aspects of the image.

The background was not "designed" or "chosen". It's just the wooden table I've got everything set up on!

(I'm still struggling with the lighting of my knife edge BTW - the bevel itself is as highly polished as my skills can make it, which is great for cutting vegetables, and hell on earth for photography)

BugBear

[DaveW]

As said before with your knife, skimmed lighting from a low angle from one side is the best way to bring out texture (grinding marks in your case) since if you think of these as little canyons it produces shadows on one side of their walls and highlights the other. You can then reduce the density of these shadows by a second much weaker front fill in light.

Set your side light up first and keep dropping it down to produce the modeling in the grinding ridges you want, even if you finish with the light almost level with the blade and skimming along it from point to handle.

If the secondary fill light is too strong just keep moving it further back until you achieve the effect you want in a couple of trial shots. You may need to assess the lighting by trial shots since things usually look different to the eye than they reproduce on the sensor.

http://pcin.net/update/2006/12/06/show- ... -the-week/

DaveW

[bugbear]

In this shot (stack of 10, f6.5, 6 seconds), the light was as low as could get it and still have light. The grazing angle was extremely low, although since the light is a normal 60watt bulb, it's not exactly a point source.

The ground "face" of the knife is nicely modelled (actually a bit glittery, although that may be due to over-sharpening in CombineZ).

But the (carefully mirror polished...) bevel remains nearly featureless.



I think I need more magnification, as well as better light.

Can anyone comment on the (likely?) feasability of exploiting a f2.8 28mm Pentax SMC lens as a reverse? That should pretty much double my magnification from me current 45mm f2.8

These seem to be available quite readily on eBay. And I should be able to get one "for cheap" if I wait for one with faulty focusing or other controls (which is fine for my purpose, of course)

BugBear

[rjlittlefield]

>> except for color balance
> artistic aspects

The color balance I was talking about is a technical aspect. I've never seen a match with anything close to that orange color of head. The stick looks too yellow also. Throw in that the background looks very yellow also, and I'm thinking that maybe you shot with the camera set on "daylight" while the illumination was incandescent.

If the image matches the real colors, or if you deliberately made them different, then color balance becomes an artistic aspect.

> But the (carefully mirror polished...) bevel remains nearly featureless.

Featureless is good in what should be a mirror finish. But judging from the tiny serrations in the knife edge, I doubt you've achieved quite that level yet.

Here are several possibilities to consider:

1. Could be you've achieved grazing over the broad face at lower left, while leaving the narrower bevel substantially in shadow except for diffuse illumination coming from somewhere else. On the other hand, the very bright spots of debris on the bevel argue against this possibility.

2. Could be that the bevel is slanted away from the camera, so that the edges of scratches are mostly reflecting light in directions that the camera doesn't see.

3. Could be that you've mirror-polished the flats between scratches, so there's nothing sticking up to catch the light even though there are still significant depressions where the scratches are.

If this were my problem, I would try three things:

1. Orient the knife so that the bevel surface is what's perpendicular to the optical axis.

2. Place illumination so that the light comes in perpendicular to the scratches. I see that you have scratches in two directions on the bevel, so this might take two lights. If you're using "Pyramoid maximum contrast" in CombineZ, I suspect you can get by with one light in two different positions, shoot two stacks, and stack 'em all together as one.

3. Raise and lower the light to test a variety of grazing angles. If there's nothing sticking up, then you'll have to get the light high enough to illuminate part of the lip of each depressed scratch.

Hope this helps. Mirrorlike surfaces are a ferociously hard problem.

> f2.8 28mm Pentax SMC lens
Should work OK at image center. Edges may go soft or colored, or vignette. I'd give it a shot if you can get a good price on the lens. But I'm far from convinced that it's going to give you more resolvable detail. Does your current image look sharp at say 50% of actual pixels? If not, then adding more magnification this way probably won't help much.

--Rik

[elf]

But the (carefully mirror polished...) bevel remains nearly featureless.

I can't help on the lighting, but it does look like the bevel is rounded. Here's a shot of a lathe tool bit that shows the bevel as a plane: http://img.photobucket.com/albums/v649/ ... ftView.jpg

[bugbear]

But the (carefully mirror polished...) bevel remains nearly featureless.

I can't help on the lighting, but it does look like the bevel is rounded. Here's a shot of a lathe tool bit that shows the bevel as a plane: http://img.photobucket.com/albums/v649/ ... ftView.jpg

Yeah, I know! The reason for the macrophotography is to (over)analyse my sharpening technique, and improve it...

Knives are rather hard to arrange jigs for - so freehand sharpening is (pretty much) normal, which (in beginners like me) results in convex bevels.

BugBear

[DaveW]

I thought hollow ground was the current trend for edge tools? Perhaps a smaller diameter grinding wheel would help?

http://www.ehow.com/how_4597079_hollow-grind-knife.html

http://www.britishblades.com/forums/sho ... w-grinding

http://www.britishblades.com/forums/sho ... inding-Jig

http://www.ragweedforge.com/grind.html

http://www.madehow.com/Volume-2/Hunting-Knife.html

Surprising what is on the WEB. I just Googled "hollow grinding" and I am sure "grinding knives" would find more.

Know nothing about knives, but being a carpenter worked chisels all my life.

DaveW

[bugbear]

I thought hollow ground was the current trend for edge tools?

As with all activities, once you look into it, it's more complex than you thought possible, more varied than you thought possible, and even the experts don't all agree.

Since I'm working on a kitchen knife, I'm getting LOTS of good information here:

http://www.knifeforums.com/forums/showforum.php?fid/26/

It's a little less testosterone-rich than many knife sites.

BugBear

[PaulFurman]

Interesting thread. Photographing metal is complicated... lots of special equipment & techniques for industrial inspection & measurement to research. The stacking system is working.

See if your camera can set a custom white balance with a reference shot. Lay a white tissue over the subject before the shoot to set WB.

Clean the blade before shooting. This is always an issue with macro and I'm not that great at it but it really helps to figure out a way to clean the subject. I usually end up doing a lot of tedious cloning to clean them up after not thinking of it before doing a big stack.

I used to work with a guy who could grind chisels on a big wheel very sharp... man those things were sharp, just one big concave swoop. I've got a scar to prove it

[DaveW]

Being a retired joiner I know the grinding and sharpening angles are matters of debate. A low grinding angle may produce a very thin sharp edge but it does not last long on chisels, particularly if not used for paring but chopping when used with a mallet. A blunter grinding angle may not produce quite such a sharp edge but the edge being reinforced lasts longer when chopping.

Therefore it is "horses for courses" depending on what the edge is needed for. The edge of a surgeons scalpel would not be subjected to the same forces as say an axe head used for chopping.

http://en.wikipedia.org/wiki/Knife_sharpening

At microscopic levels though there is no such thing as a smooth polished edge, so I would have thought skimmed lighting would pick out some texture from the grinding marks?

However I think this shows even with a microscope how hard it is to get a decently lit picture of an edge.

http://www.hirdyman.co.uk/2010/02/17/ra ... pe-razors/

You could do with one of these BugBear.

http://www.catra.org/pages/products/kni ... l1/bet.htm

You may find the photo's of grinding marks and honing method here of interest, also to microscopist's too.

http://www.microscopy-uk.org.uk/index.h ... roj3b.html

For the start of the article for microscope fans.

http://www.microscopy-uk.org.uk/larry/p ... proj3.html

DaveW