The refractometer is an instrument to determine the identity of a gemstone. Unfortunately, the price is an obstacle for the (beginning) gemmologist. An "official" device costs around € 600 including the lighting. If you take a look at our Chinese friends, it turns out that it can also be done for under € 90, and you also get the refractometer liquid for free.
But does it work? A refractometer in itself is a 'simple' device. Hardly any moving parts, passive optics that does not wear out, so it can certainly be mass-produced. The design has evolved for a century, so most problems should be solved by now. Enough reasons to take the gamble, so I ordered one. I compare the Ali model in the picture below with the refractometer of the Gem-A on various aspects.
Mechanical / Material
The metal exterior of both models is fine, I can't detect any difference in quality. It is in line with other experiences with Ali, metalworking is no problem in China.
Polarizing filter
Both copies have a separate polarizing filter. And it just works for both.
Lens
The lens of the new generation Gem-A refractometers is somewhat larger, and therefore easier to use. The eyepiece has a screw thread for focusing. The Ali model has a smaller lens, and focusing is done with a sliding system. In itself it is slightly inferior, but it hardly matters for use.
Lighting
The Gem-A refractometer has an external light source, which you have to buy for € 80. This offers advantages: the intensity of the light source is adjustable, and in the event of a defect you do not have to replace the entire refractometer. The Ali refractometer is equipped with an internal light source (yellow LED), an on/off button and three button cells (standard LR 44). I personally find an internal light source considerably more pleasant than an extra box, which is sometimes pushed aside. Being able to set the intensity can sometimes be tricky, but in practice I have no problems with it.
The Ali works well, but unfortunately the ergonomics are a bit clumsy. The on/off push button is mechanically a bit weak, and if you put the refractometer in its box, chances are the light will stay on because the button has been touched. After a while the button cells are empty. Furthermore, the three button cells are difficult to remove. The original cells are wrapped in a shrink tube, but very thin. The standard heat shrink tubing that I use is just a tenth of a millimeter thicker, and then it no longer fits (!).
This is something that happens with more Ali designs. The ergonomics are clearly less than with products that have been better thought out.
There are also Ali refractometers with an external light source. You then have to mount a flashlight to the refractometer. I've seen that model, and it becomes a very clumsy device. As far as the lighting is concerned, the ideal situation is that it is built-in and adjustable. Unfortunately, both refractometers do not achieve this goal.
Refractometer liquid
Good refractometer contact fluid is expensive. With the Ali specimen it was claimed that nD = 1.81 contact liquid would be included (nD is the refractive index of the contact liquid itself. It must be higher than that of the gemstone to be tested). Remarkable in itself, because with the GIA this liquid costs just as much as the entire Ali refractometer put together. Indeed, there was liquid. I wasn't that skilled yet, and later I found out that it was (probably) methylene iodide without additives. This has nD ~ 1.74, and therefore not usable in practice (all corundums are out of range). I then ordered from Ali through another supplier nD = 1.81, but it turned out to have nD = 1.76. So no good either. Finally I bought in the Netherlands at a local dealer nD = 1.78 – 1.79. That was right! Cheap is sometimes expensive.
Absolute Accuracy
The observed refractive index differs between the Gem-A and the Ali refractometer. And the higher the refractive index, the greater the deviation. It goes from 0.002 by 1.4 to 0.006+ by 1.8. That's a remarkable difference. I borrowed a number of calibration glasses, and compared both meters together with other materials of known refractive index (such as distilled water and synthetic spinel). As was to be expected, the Gem-A was right, the Ali was not. It's also rather a lot so you have to adjust for it. I have made a correction chart, which I must use for all measurements. The graph shows the measurement error that you always have with a refractometer as a comparison, it is not better than 0.0015.
The good news is that the measurements on the Ali refractometer are repeatable. It is not the case that these also differ from each other when you return to the starting position after rotating the stone. The birefringence can be determined just as accurately as on the Gem-A. The Ali model is therefore clearly less accurately build. The mechanical parts are correct, but are not properly “aligned”. In other words: quality costs time costs money.
Conclusion
If you know the limitations, and have made a correction chart, the Ali is useful. I liked making such a graph, but it can also be a bridge too far. It is clear that without separate calibration the Ali is not sufficiently useful for good measurements. I don't think it's desirable for business use.
