Diffuse on axis lighting Rig: my current iteration

Started by Levantiner, February 28, 2016, 03:31:59 PM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.


{this has been posted in another forum} 

I recently saw another post in this forum on the same subject I thought I would post my approach here. Its certainly not perfect but it has taken me three years to develop this part of my macro rig for coins

MY title of this thread may be confusing, and that may be because I have a confused understanding.   I am referring to what is most often called Coaxial lighting, or alternately Axial lighting.     

In 2013 I was taught at the ANA summer seminar that axial lighting, the name they used, needed a beam-splitter and a spot light.  The spotlight and beam-splitter was used to direct the light onto the coin so that the apparent source of the light was the lens. I tried that and hated it.   I know others love, and even insist on using spot lights, but its not for me.

So I did a bit of thinking and research.  I found that not every one used spot lights and some just used a large diffuse light source. 

The reason I don't like spot lights  is related to my processes when taking coin photographs.  It is also probably related to my collecting focus.  Although a large part of my collection consists of coins  around 40mm in diameter, I also have coins and medals ranging from a few mm in diameter through to nearly 70mm in diameter. A single spot light,from my experience, needs readjusting  when one changes coin size. {This is usually due to changing the magnification so to ensure that the coins image fills  a large part of the Cameras sensor}

The spot light mount just added to the clutter of my work area ( as did Janscos).  When I first looked at improving my axial lighting I contemplated adapting a Leica Visoflex. The plan was to replace the mirror in the visoflex with an Edmunds beam-splitter. The light source would be projected thorough the view finders pentaprism  behind which a  polarizing filter can be installed.  A second polarizing filter could be set after the beam splitter to cancel out the transmitted light.  I never tried constructing it as there were just far too many variables I needed to take into account( plus expense).

Here is an example of a Visoflex( I am not recommending any one buy it)


I then looked at how the problem is addressed in Machine vision.  I looked at DOAL ( Diffuse ON AXIS Lighting) Units.  For example(Again I am not recommending any one buy this item):


I then developed my own DOAL unit using the following Edmunds products:

The beam splitter( the price is considerably higher than when I purchased mine):


Self adhesive black out material( camera flock):


I find the self adhesive flock so useful and cheap that I  always keep a spare sheet

A long time ago I posted the first iteration of this set up...it has evolved many time since then.   A month ago I modified my DOAL set up further. After having made a panel light that attaches to my camera lenses I had some extra LED strips left over:

I think from the following photos it is fairly obvious what I did.  in the process of building this latest iteration  I finally found a use for one of my dust gathering jansco lights. The power supply for the jansco was perfect for the LED strips!:

The beam-splitter part of the rig wit integral lighting:

This is how it looks when the light is on:

The insides of the beam-splitter housing are covered with the Edmunds blackout material.  Likewise the work area of my macro stand is also covered with the same flock. When there is no coin in position the cameras histogram records only black.   There is one disadvantage of covering the work area in that flock: it is a dust magnet and it really shows the dust. I use a 3M lint roller to remove the dust from the flock. That Lint remover is as useful, for me, as an 11% grey card. Instead of dusting ( which just puts dust in the air I use the lint remover( except for coins of course)

Crucial to the whole rig is a shroud that absorbs the light that has passed through the beam splitter. It is also lined with the Edmunds Back out material. Without it I have huge problems with reflections from the top surface of the beam splitter. The Shroud:

The out side of the shroud is covered with black contact paper.

Heres how it all goes together. I  used the camera to take these two shots...you can see the swiss arca standard clamp that normally holds the camera above the  DIffuse on axis lighting rig

with the shroud in place:

Here are some recent examples of photographs:

The following photo is almost straight out of the camera. the only editing was to convert it from RAW to a JPEG:

IN my experience this lighting source is brilliant for gold whether proof, proof-like or circulated.

It is not so good for low relief coins this following shot is of a proof-like strike, but it is a very shallow design and when edited the coin actually looks cartoonish. I will probably try a honeycomb diffuser with normal lighting for a better shot.   Like the previous photo  this photo is unedited. I don't worry too much about composition before editing(before any one makes comment about the orientation of the coins in the last two photographs)

After editing in Photoshop Elements

the rig as it stands is good but isn't perfect:   

Some faults( these are not the only ones but they are the most annoying):

1:  The LED strips I used are not dim-able.  The light is either on or off.  I had thought the LED strips were dim-able but I was wrong( I found out after assembling the lighting unit). Certainly  the rig would be much more versatile with that function built in ( may be next year I will rebuild it)

2.  The space to get coins into the lighting area is too small. I should have built the beam-splitter stand higher to give more space. The same applies to the shroud I had made it so that I could change coins while it was in place but the gap just isn't high enough. To change photography subjects I remove the shroud, position the coin and then replace the shroud.

3.  I am not sure I have oriented the beam splitter in the most optimal way.  I suspect if I had it mounted 90 degrees to the current set up I would  be happier( but I won't know until I try it)

The Camera is a Sony A7rM2, lens was Sony FE 90mm F2.8. Shot in RAW, converted to JPEG by Capture one express and final editing, if any, in photoshop elements 14


Quote from: Levantiner on February 28, 2016, 03:31:59 PM
1:  The LED strips I used are not dim-able.  The light is either on or off.  I had thought the LED strips were dim-able but I was wrong( I found out after assembling the lighting unit). Certainly  the rig would be much more versatile with that function built in ( may be next year I will rebuild it)

The LEDs in my USB microscope can be dimmed, showing that it is technically possible. Dimmer switches (essentially a hand-adjustable resistor or conductor or whatever you want to call it) are cheap and widely available...

An unidentified coin is a piece of metal. An identified coin is a piece of history.


It depends on the LED type. I  tried a dimmer switch with these and it was a failure


Just now read your description of your setup. Very instructive ! Gives some ideas i may work out in the future.
(But for now i just stick with simple)