Return to CD Index
The Art of Observation
by Clive Washington
This article first appeared in Stonechat, the Journal of the United Kingdom Facet Cutter's Guild

Seeing What You're Cutting

If you intend to cut stones of any real quality, or even work to competition standards, then it becomes essential to evolve a well-understood set of methods for seeing what you're doing, and knowing their limitations. Different people evolve different techniques to suit themselves, and I'm not going to claim that any of these are wrong; what follows is simply what works for me. The basic rule is: if you can't see it, you can't cut it. This sounds trite but it has two consequences; firstly that your stones can be no better than the resolution of your observing techniques, and secondly that you should be aware what faults you're looking for, and devise specific observing techniques optimized to detect them. I divide the area into three sections: vision, magnification, and lighting. If any one of these isn't right, you can't work, and you can't compensate for a deficiency in one area by messing with the others.


The first, and most important thing, is to get your eyes tested regularly. Faceting is close, critical work, and persistent use of high magnification lenses can cause eye strain. You must protect your sight because, quite simply, if it fails, that's the end of your faceting, and a lot more too. You wouldn't be able to cut any more stones and you couldn't enjoy those you'd already done.

You should determine whether you are right-eyed or left eyed. If you don't know, just use a loupe to look at something. You will have instinctively used one eye or the other. I use my left; apparently many right-handed people are left-eyed and vice versa. Knowing this, you should try to use the loupe with both eyes alternately; this will initially prove very difficult, but it's worth persevering. If you start getting headaches, take a break. Most people spend far too long looking at one thing without resting their eyes - just look at the way guidelines for VDU use have developed over the last few years. I try to take a 'distance vision' break every 20 minutes or so but it can be hard to tear yourself away at times!


Your loupe, or loupes, are absolutely critical and should be the best you can afford. Over the years you will probably look at thousands of pounds' worth of stones with them, so anything other than the best is false economy. Your failure to observe a single defect in a valuable stone can make a difference to its value that is far greater than the cost of the loupe itself. I have two loupes, an 8x and a 20x, and as I will describe, they have specific uses. They should be corrected doublet or triplet glass lenses, molded plastic lenses are useless. Headband magnifiers also just don't deliver the goods - they have nowhere near enough magnification.


Your lighting setup is very important and you will probably spend a lot of time fiddling with it before you're completely happy. The problem is that different observing techniques have different lighting requirements. The best way to illustrate this is to go into a room lit with only a single point light source and a stone and magnifier. Draw the curtains, or work at night, so that there is no daylight getting in. Hold the stone in tongs (I use the 3-claw spring tongs with the spring considerably weakened) and try to look at a particular facet. Everything will be impenetrably dark unless you manage to catch the reflection from the facet, when you will get a bright image, probably too blinding to be able to see anything useful. The glare around the facet will prevent you from accurately seeing its periphery and meets. Try to see how good the pavilion point meet is - you will never be able to see clearly because only one facet will ever be brightly lit and the rest will be in darkness. The bright one seems to 'extend' beyond its true borders, even when it's accurately focused. If you now tilt the facet slightly off-reflection, as the main reflection fades to one side, you will be able to clearly see any scratches, pits, or smears, thrown strongly into contrast. This off-reflection observing technique is invaluable and we will discuss it again later.

Now take the same stone into a well-lit white room or, better, under a bright cloudy sky. The focusing of the stone is easier and the meets can, at last, be seen accurately. However all those scratches you could see under the spotlight are now much harder to see or are even invisible. This applies to surface finish too; a stone polished with 1 micron diamond will look well-polished in diffuse daylight but under a spotlight the surface will appear hazy and poorly polished.

So to make any progress you need a combination of lights. I work in a small conservatory with a plastic roof, so even in dull conditions I have more than enough light for meet observation. For polishing I have a 40W pearl-finish spotlight clamped to a roof beam about 4 feet above the lap, pointing downwards. The light is on a gooseneck (it's a modified desk lamp) and has been adjusted so that, with my normal seating position, when I swing the stone upwards for observation it's fairly easy to catch the reflection. This light is connected to a dimmer but in practice this wasn't necessary - it's only ever used at full power. The only problem with this arrangement is that when the daylight fades, I have to do less critical work.


Faceting involves the observation of four main stone features; meets, polish, flatness, and edge roundness. Each of these needs a particular observing technique. Of course you won't need to use all the techniques all the time; you will vary your technique depending on what you're doing. But having specific checking methods will allow you to home in on the right technique quickly and efficiently, and will avoid a lot of inefficient random squinting through the glass.


As you work your way through the cutting process, meet observation is the first thing that becomes critical, usually when you've descended to about the 6 micron stage, and then through until the final polish. Meet observation has two requirements; firstly the highest possible magnification, and secondly a uniform illumination on all the relevant facets. Although the 8X glass will suffice in the early grinding stages, when you finally polish the meets into position you will need to change to the 20X glass.

Despite having a lot of magnification, it can often be hard to see if a meet is correct. The eye becomes confused when looking at the meet and it's hard to perceive what's going on, especially if there are 5,6 or more facets involved - your eye just can't take in the jumble of tiny lines at the misaligned meet. To get round this, I use what I call the 'line continuation' technique. Don't try to look at the meet itself, instead concentrate on the facet edges that join it. If the meet is correct the lines will appear to smoothly pass through the meet without a hiccup. The figures below illustrate this method in use. You can't always find edges that continue in this way, but looking at the lines, rather than the meets, will usually give you a better idea of whether or not something is out of place.

The line continuation technique. Even though the meet is obscured by the dot, it's easy to see that the left-hand meet is correct, while the lines on the right-hand meet are misaligned.

Alignment of girdle breaks on a typical brilliant pavilion. Line continuation is even more useful here. Since the tips of both girdle facets have similar sharp angles, the offset in the vertical line provides the best view of the misalignment.


We already examined the use of lighting to detect polish and scratch quality. The reflection from the overhead light allows any surface defects to be shown up clearly, with the 8X or 20X glass not making a lot of difference. It's important to get the facet accurately in focus (which can be tricky) since a scratch is a very thin feature which will go out of focus easily. The procedure I generally use for polishing is to polish the facet using the spot illumination until the surface pits are gone and there are no scratches. Then you switch to 20X magnification and diffuse illumination to push the meets into their final position. Finally check again for scratches just as you're approaching perfect meets - there shouldn't be any unless you have a contamination problem.

There is a technique for observing scratches which can be very helpful. Under virtually all circumstances scratches will be generated by the lap and so will all run in the same direction. In my setup this is pretty much vertical since the stone trails tangentially on the lap, although the scratches may be transverse if your dop arm lies radially on the lap - it all depends how you use your machine. Because of the way the scratch scatters light, it's much easier to see in some directions than others. Let's suppose your scratches run vertically and you have the facet under observation so that you've just caught the spot reflection. To bring the scratch into contrast you tilt the stone slightly off-reflection so that the facet goes dark; however the scratch scatters light and thus is visible as a bright line. Now you have two options. You can 'lose' the main reflection by swinging the stone vertically or by rotating the head (the machine's, not yours!) horizontally. But a vertical scratch scatters most of its light to the sides, and very little light up and down, so if you swing the stone vertically to lose the reflection, you probably won't see the scratch. If however you rotate the head slightly so that the bright reflection moves to the side, you will be able to see the strong sideways scattering from the scratch. The same reasoning applies if your machine produces horizontal scratches - you should swing the stone off-reflection vertically since the horizontal scratches will scatter most light, and thus be most visible, in the vertical plane.

Incidentally there is another reason why it's worth knowing which direction your scratches lie. When you wipe the stone for observation, it's often hard to tell the difference between a smear and a scratch. If your scratches are always vertical, just wipe horizontally, and you will always be able to tell the difference. Smearing is rarely a big problem for me, though, as I use water lubrication almost exclusively. I stopped trying to use oily lubricants long ago, as I could never wipe the facet clean for proper observation.

When polishing very large facets, such as the table, or mains on a large stone, it can be very difficult to see the final disappearance of a troublesome scratch. Often it's possible to see if any scratch remains simply by breathing on the stone. Watch the film of moisture evaporate through the loupe and, as it recedes over the facet, it will often linger a while at any surface defects or scratches. I suspect that the surface defects act as condensation nuclei and thus cause larger water droplets to deposit on the surface at these points. These take longer to evaporate, outlining the scratch as they do. Don't use this technique on water-soluble stones!


Really flat facets add enormously to the life and activity of a stone. There are two reasons why a facet may not be flat; there may be machine deficiencies, such as lap runout or loose bearings, but these really just need a one-off engineering solution. What we are really concerned with here is the situation where a facet is being pushed to an exact meet by applying a small height or pitch adjustment during polishing. If this isn't carried out carefully the facet can end up bent or split, rather than being completely recut in a new position. Now I'm aware that I may be in conflict with some cutters by advocating 'cheating' during polishing. Some people think that it can only lead to rounded facets and that the facet should be accurately cut before polishing. To my mind this is nearly impossible because the polishing process itself removes material and thus moves the meets around. Trying to guess where they'll end up is pointless. I long ago accepted that polishing is just a fine form of cutting and that precision meet control could only be achieved at this stage. Of course, you shouldn't try to remove a lot of material, or correct a major error by polishing - this will take ages and wear out your lap (and your patience!).

As the facet approaches completion I adjust the head so that the meets are all achieved accurately. The slight adjustment really cuts a new facet at a position very close to the old one, and if this cutting is stopped prematurely, we have two facets where there should be one. The most troublesome aspect of this is that the bend or joint in the facet becomes less visible as the magnification is increased. When I'm doing this sort of correction, I invariably use the 8X glass because a bent facet appears truly flat under the 20X glass. The bend is even more visible to the naked eye if the facet is large enough - hence the need to make sure that the repolishing in the new position is completed. I suspect that repositioning facets during polishing has got a bad reputation simply for this reason; people thought that they had got it right under 10X, then found that the previously invisible bends in the facets were offensive to the naked eye. Again it is useful to use the off-reflection technique with the spotlight, which will provide a clear contrast between the flats on the facet being pushed around.

Edge Roundness

The sharpness that you can achieve between facet edges depends on many factors; the type of lap, the material being cut, the quantity and viscosity of polishing compound, if you're not using diamond, and many others. This is a poorly explored area and hopefully will be clarified by more research. So there isn't really a need to test edge roundness during cutting, as once you've decided on you cutting technique it's beyond your control. However if you're trying to refine your technique, you need to assess the sharpness of your facet edges.

The easiest way to look for edge roundness is to use the technique used by woodworkers to test the sharpness of edge tools. If an edge is truly sharp, you should not be able to scatter light off it. If you examine the stone under a bright light, and try to move a facet reflection from one facet to another, the edge between them should not light up if it's really sharp. Normally, however, a fine line of light can be seen, showing that the facet edge has a small radius. Practice with this method will allow you to assess edge sharpness fairly well. I have recently assessed the sharpness of some of my facet edges using electron microscopy and the edge radius is of the order of half a micron, which is consistent with my use of 1-micron diamond on tin for the polishing process. Even at this sharpness, it's not difficult to see light being scattered from the illuminated edge. I know that this can be improved on, since I've seen diamonds with almost invisible edges, but I've yet to achieve it in a softer gemstone.

Return to CD Index