Thursday, July 31, 2008

Interview With Frances Baines. Part IV

For previous session (part III) click here.

10. I am not familiar at all with Metal Halides. Can you elaborate more about this type of lamp. Do you perhaps have a Metal Halide Spectograph that you can share with us? -Q contributed by Kevin Zamp-

Metal halide lamps are basically mercury vapor lamps with added compounds of various metals and halogens inside the arc tube. There's a great description of these here: http://en.wikipedia.org/wiki/Metal_halide_lamp
The exact combination determines the color of the light produced, and it is much brighter than an ordinary mercury vapor lamp. A good quality 70W flood will produce maybe 75,000 lux at 12 inches... that's illuminance at full sunlight levels. Obviously you don't
need to have the lamp so close... placed at a greater distance, the light will cover a much wider area but still provide intensely bright illumination.
The other impressive thing about metal halides is that ones with "daylight" color balance (around 5,000 - 6500K) have so many "spikes" of light at different wavelengths, produced by all the different chemicals, that the spectrum is almost continuous. This makes for extremely good color rendering - the spectrum is more like sunlight than any other lamp.
This chart shows the spectra of sunlight, one of the best metal halides, the Iwasaki/Eye Lighting Color Arc PAR36 (sold for human use) and a mercury vapor lamp for compariso
n.


Metal halides for human use (lighting showrooms, factories, airports and sports stadia, etc) are sold as double or single-ended unshielded bulbs to go inside lumieres, or as self-contained "regular" bulb shapes, eg. PAR38 or PAR30 lamps with Edison screw fittings. The unshielded bulbs can be very dangerous; some emit vast amounts of UVC. Those behind lumiere glass, and those sold as "regular" shaped bulbs are rendered safe for humans because the outer glass blocks UVC and UVB. Those being developed for reptiles, however, use the same glass as used for reptile mercury vapor lamps - which lets through useful UVB, but blocks harmful UVC.
There are two companies producing "reptile" halides at present; the Reptile UV MegaRay Halide 70W lamps (made in Canada) and the Lucky Reptile 50W, 70W and 150W Bright Sun series (made in China). The spectra from these are all fairly similar, they produce a very good white light and variable amounts of UVB. This chart shows the spectra of the ReptileUV lamp and the 70W Lucky Reptile Bright Sun Desert lamp.
However, if you look carefully at the spectrum of the Iwasaki/Eye Lighting Color Arc lamp in the previous chart, you might be able to see that this lamp, when new, also delivers very nice UVB and UVA, but this fades rapidly. In fact Andy and Janette use these for "artificial sunlight" over most of their cages, monitoring their UVB output and keeping the lamp, but adding a UVB tube, when the UV output has decayed too much.
But before we all rush out to buy a metal halide, there are a few drawbacks to these brilliant lamps.
Firstly, a very good electrical system is needed to operate them.
They are all externally ballasted, because they need really precise control of their operation. And just any old ballast won't do. The quality and color of the light is extremely sensitive to the current flowing through the arc tube. High quality ballasts are needed to avoid flicker, frequent cutting-out and re-starting, and loss of the full color spectrum.
They are ignited by a 4KV initial pulse from this ballast. This requires excellent electrical connections and fixtures designed to handle that sort of voltage, as well as the heat. Anything less - moisture or dirt in the terminals, for example, or too long a cable between ballast and lamp - and the lamp may fizz and sputter, but it won't ignite properly. (They take a few minutes to ignite and reach full brightness anyway, often with assorted flashes and flickering).
Secondly, my initial tests seem to suggest that they decay rather faster than mercury vapor lamps. This may be in part due to the quality of the ballasts used. However, as far as I know, the arc tubes in use in these lamps are quartz glass, not ceramic and, as the Wikipedia article explains, these do have a limited lifespan owing to a problem known as "ion creep" which also causes color changes during the life of the lamp. Presumably this also affects the UV output. The lamps I have tested have varied widely in their decay rates, but long-term testing of several of the ReptileUV and Lucky Reptile lamps has shown decay of between 34% and 65% in the first 1,000 hours (3 months at 10 - 12hrs a day). If these rates are typical, few lamps will have a useful UVB lifespan of more than 6 months.
Because of the variability with these lamps, I think regular monitoring with a UVB meter is essential.

11. What equipment you recommend for us to have to better measure the amount of light necessary for our reptiles? (Is owning a uvb meter 6.2 enough? Or would you recommend us to have a UV Index meter as well?)

The Solarmeter 6.2 is the most versatile and useful broadband meter available; I wouldn't be without one. But like all broadband meters, it can only measure what it "sees"; and it sees the whole spectrum from UVC to a short distance into the UVA. This means it can't distinguish between a lamp that produces a lot of dangerous very short wavelength UVB from one which produces a lot of harmless long wavelength UVB. In other words, you can use it to monitor a lamp's decay, but you can't use it to decide how similar a lamp's output is to sunlight, or to compare one brand of lamp with another.
A year ago, I would have suggested that anyone who could afford to spend a little more on their reptile equipment would find a Solarmeter 6.5 UV Index Meter a worthwhile investment. These "see" only a much narrower band of wavelengths; the Silicon Carbide sensor in combination with a special interference filter is calibrated to give a good estimate of the amount of UV below about 300nm (UVB and UVC). Dr Gary Ferguson and his team have demonstrated that this meter (and its "sister" meter, the Solarmeter 6.4) can predict quite accurately the vitamin-D3-synthesizing potential of any lamp, and the sun.
The 6.5 meter gives the readout in terms of the easily understood UV Index. If we know the basking habits of our reptiles, we can estimate the UV Index they would choose to expose themselves to, in the wild, and then decide how much UVB to offer. It's a very handy safety feature, too. The highest UV Index ever seen, at mid-day on the equator at sea level, is around 16 - 17. Reptiles would rarely venture out into that. So what would you do if you discovered your new lamp would expose your reptile to UV Index 100 or more? Sadly, this is no joke...I know some "spot" mercury vapor lamps still on sale today which emit extremely dangerous levels like that, at the manufacturer's recommended basking distances..
The only drawback with the 6.5 meter is that it's measuring such tiny amounts of light that the UVB from a fairly low-UVB fluorescent tube, at basking distances, will hardly register a flickering 0.1 or 0.2 on the meter. At the very limits of its resolution, the meter is not very accurate at such low levels.

But unfortunately, the 6.5 meter not be on every herper's shopping list, today.
It relies upon a SiC sensor. These are now considered outdated, and are being replaced by AlGaN sensors in most applications. The supply is now very limited; they are no longer being manufactured, and only one supplier, with dwindling stocks, now exists. The Solarmeter 6.5 with a SiC sensor has just unavoidably suffered a massive price increase as a result, although Steve Mackin believes he will be able to supply these very special meters for some time to come.

The AlGaN sensor is suitable for monitoring sunlight and lamps with spectra very much like sunlight. However, it does not respond well to short wavelength UVB and UVC, and as a result broadband meters fitted with AlGaN sensors do not give comparable results with many types of reptile lamp, and in fact the ones I trialled were unable to detect hazardous output from several lamps giving extremely high UV Index readings with the original Solarmeter 6.5 with the SiC sensor.
People who already own both a 6.2 and a 6.5 meter can make "conversion charts" by taking paired readings from the same lamp at different distances, which can be used to plot a linear regression and obtain a formula for calculating the UV Index equivalent to any Solarmeter 6.2 reading from that lamp. The formula will work, in theory, for any other lamp of that specific brand; but each brand will have its own unique formula, because they all have different spectral power distributions. I hope to publish a full set of conversion formula for all the most popular lamps, on UV Guide, eventually.

12. Let's say that we allowed sunlight into our reptile room. Even though it passes glass, does it still keep a lot of the spectrum below the UVB level? -Q contributed by Kevin Zamp-

No. Ordinary glass (as opposed to quartz glass) acts like a UV filter, progressively blocking shorter wavelengths of UV. Window glass blocks 99 - 100% of the UVB. It allows some UVA through (about 65% at 350nm) and almost all visible light (over 90%).
If you want to glaze a window to let in some UVB, though, special high-transmission glass (which is very low in iron) is available which will let some UVB through. In the USA, one such glass is PPG Industries "Starphire" Uncoated Ultra-Clear Float Glass. Two brands which can be purchased in the UK are Asahi Glass Company's Planibel Clearvision (28% transmission of UVB at 305nm) and Pilkingtons' OptiWhite (25% transmission at 305nm).

13. What do you think of external vs. internal ballasted lamps for MV and MH (if it comes that way) -Q contributed by Kevin Zamp-?

I don't think self-ballasted metal halides (using a simple intermal tungsten filament) are going to happen; these really need electronic ballasts because of the fine control needed.
From the point of view of light quality, one might think self-ballasted (SB) mercury vapours would be better than externally-ballasted (EB) ones, because of the tungsten filament producing some continuous-spectrum, golden light to offset the very "spikey" spectrum and very poor colour rendering of the light from the mercury arc. They are marginally better; but the arc tube is so bright it overpowers the tungsten filament. In my opinion both SB and EB lamps should be combined with "daylight" full-spectrum lighting, ideally with a continuous spectrum, as I mentioned before.

The main reason for choosing an SB lamp would be to increase the heat at the basking spot. A 160W SB MegaRay produces a very suitable basking spot for my chuckwallas (up to 120F!) which is quite impossible with a 60W EB MegaRay. Conversely, an EB lamp would be much more suitable for use with montane species, requiring high UVB but lower temperatures.

14. So, as regards provision of the best light for reptile keepers, do you have any more tips that you can share?
I think I have probably said too much already!
If anyone asked me to summarize everything in just a few sentences, I'd probably say:
Research your species! Plan to mimic its natural light environment as far as possible.
"Select and combine" lamps to get the best spectrum you can.
Always, always check your temperatures.
Never buy anything just because it's really cheap. UVB lamps are health products, like drugs, not fashion accessories like plastic plants or food bowls. Would you buy a packet of pills if you didn't know what was in them?

15. Last, but not least, what is next for you? Any exciting researches you are going to do? Books to publish?

Right now, I just want time to update the UV Guide UK website!..... but one day, yes, I do hope to write a book on the subject. There is a lot going on at the moment, too, that I'm very excited to be involved with. For example, Peter Nunn at Alice Springs Desert Park has completed a marvellous survey of the voluntary UVB exposure of wild long-nosed dragons over a full year; we hope to write up this study with Dr. Gary Ferguson.

Thank you so much for this wonderful opportunity! I wish you many success for your future research.

----- End of the Interview ------

No comments: