Refractometers and the color of light used: important!

[Randy Holmes-Farley]

I was recently asked whether the light impacts the results one gets with a refractometer.

I'll give the rational below, but it turns out it is CRITICAL that you use the same color light for calibration as for measurement. Don't move to a different room of your house where the lights may be different.

The reason for this is that the refractive index of water (fresh or salt) changes a lot with wavelength. The standard for all visible light refractive index measurements is supposed to be yellow (specifically, the yellow doublet sodium D line, with a wavelength of exactly 589 nanometers), but reefers rarely worry about such complexities.

The link below shows how much the refractive index of pure fresh water changes with wavelength (color):

http://www.philiplaven.com/p20.html

specifically, here's a graph:

http://www.philiplaven.com/RI-Comparisons.gif

Look, for example, at the difference between deep blue (say, 480 nm) and orange (about 650 nm). Let's look at just the light blue line. That difference, from about 1.339 to 1.3325 is, remarkably, about the same as the difference between pure fresh water (RI = 1.33300) in yellow light and natural seawater (RI = 1.33940)in yellow light.


So the color effect on refractive index is as large as the entire difference between salt and fresh water.

However, we have a big saving feature at work. As long as you calibrate and measure in the same light (same color), the overall effect of the change in refractive index with light color largely cancels out between the calibration and the measurement, and you are able to sort out what portion of the effect comes from the added salts in the aquarium.

What you cannot do is calibrate under one type of light, and then measure tank water under a different set of lights (for example, in a room with fluorescents vs a room with incandescent lights or outside, or in a basement and then next to a tank with a lot of blue light.

Of course, rarely would someone have such extreme light differences as the deep blue vs orange mentioned above, but that is a 100% error in the salt concentration (ie., salinity might be really 1 ppt or sg = 1.001 when the refractometer says 35 ppt or sg = 1.026). I assume most reefers would like the salinity measurement to be far, far better than that, and so paying attention to the color of the light within the range aquarists might actually encounter is important.

So be sure and calibrate and measure in the same location if using a refractometer!

[supra400hptt]

Good point, learned something today. Sometimes I go to the window using sunlight, sometimes I stand under the dining room light. I'll check tomorrow and see if it makes any real world difference in my scenario.

[grant778]

Yes, thank you a lot for this info. I tended to change between natural light and florescent if its cloudy or the sun goes behind a cloud, but I shall stick just to florescent now I think, and I will definitely recheck the salinity of my tank and newly mixed saltwater tomorrow making sure to use one light source.

[Randy Holmes-Farley]

You can test the issue by calibrating with a 35 ppt standard in one type of light, then measure the same solution in another. :)

[Spartanman22]

Thank you snell!

[discocarp]

Thanks Randy, this is great info. I never even considered that as a variable at all for even the slightest second. It never crossed my mind. I suppose this means testing on my porch where the quality of light changes depending on time and weather is a really, really bad idea.

Just one more reason to use conductivity meters instead of either refractometers/hydrometers. I'll never go back.

[Ontheway]

In daily life, I think what we may experience is difference of kelvin between lights, or artificial light vs sunlight. Another issue may arise in future is increasing availability of led bulbs which has limited spectrum then flos. I hope one day someone may come up with a pluggable apparatus to provide constant light for calibration n measurement.

[outssider]

I knew that room temperature can play a big role but didn't realize this about light. Just another good reason to calibrate before every use !!

[bundybear1981]

Hi Randy,
On this, I usually use the tanks lights to measure. Would the different intensity at different times of the day impact readings? (ie morning reading with lights running about 20% vs. peak lighting period of 70% intensity on a ReefBreeders Photon48)
thanks,
Jason

[Buzz1329]

D'oh!

I determine salinity on my 2 tanks with 2 different light sources. Sunday morning now booked.

Thanks for heads up, Randy.

Mike

[Randy Holmes-Farley]

Hi Randy,
On this, I usually use the tanks lights to measure. Would the different intensity at different times of the day impact readings? (ie morning reading with lights running about 20% vs. peak lighting period of 70% intensity on a ReefBreeders Photon48)
thanks,
Jason

Intensity won't matter, as long as you can clearly see the line. Only color. :)

[Randy Holmes-Farley]

D'oh!

I determine salinity on my 2 tanks with 2 different light sources. Sunday morning now booked.

Thanks for heads up, Randy.

Mike

Let us know if you find a difference. :)

[ca1ore]

Zoinks, one of those things that never ever occurred to me ..... but makes perfect sense since we're talking about a device that uses light refraction. Fortunately I use mine in one room only, and have a conductivity probe as well. Excellent advice though.

[whiteshark]

I always take mine out on the deck as sunlight always gave me the most distinct line, but now I wonder if the time of day I measure might make a difference.

[discocarp]

I always take mine out on the deck as sunlight always gave me the most distinct line, but now I wonder if the time of day I measure might make a difference.

Even weather! I used to test like that all the time.

[bundybear1981]

By what Randy said to my earlier post, the intensity won't make a difference. Only the light source so using natural light will still give the same results regardless of time or weather. If you were to then go indoors and use fluorescent or incandescent lighting you would likely see some difference.

[disc1]

This is another good argument against trying to calibrate one of these things with RO water as well is it not? Looking at the graphs Randy provided it seems like the difference in refractive index between pure water and 35ppt is different in different light. So even if it does work out in one light it very well may not in another light.

[DSMpunk]

Interesting. This could explain an issue I was having when I setup my new tank. I had two refractometers, both calibrated at my old apartment. I set up my new tank and had all sorts of issues with new livestock, especially snails and shrimp.

Both refracometers showed 35 ppt on the nose, so I suspected it was something else. After replacing test kits, running purigen and exhausting everything I could think of, I decided to re-calibrate my refractometers. The cailbration fluid was way off, I believe it was showing a 4-5 ppt difference.

The difference was, I used natural lighting to calibrate my refractometers in the apartment and my sump is in my basement at my new house. I have been using artificial light to test now.

May not be related to this, but I always thought it was weird that BOTH of my refractometers were off in same exact same way.

[Turbo4toy]

That's good to know and it makes sense. I guess it's good to be consistent with as many factors as possible when testing to ensure accurate numbers to compare.

[Ontheway]

DSMpunk; Although I have no access to a calibration fluid, if its a liquid with properties of normal seawater, remember water part will evaporate as long as you keep the cap open. It may skew your calibration. What you can additionally do is, read info of the salt provided by producer, on what amount of salt needs to be dissolved for 35ppt. Prepare a small amount of mix in the measurement room then check it with refractometer. You shouldnT see any large difference to calib. fluid. I do it with 70grams of salt and a two liters of rodi water.

[Sk8r]

Interesting! Thank you!

[disc1]

DSMpunk; Although I have no access to a calibration fluid, if its a liquid with properties of normal seawater, remember water part will evaporate as long as you keep the cap open. It may skew your calibration. What you can additionally do is, read info of the salt provided by producer, on what amount of salt needs to be dissolved for 35ppt. Prepare a small amount of mix in the measurement room then check it with refractometer. You shouldnT see any large difference to calib. fluid. I do it with 70grams of salt and a two liters of rodi water.

This isn't necessarily safe. You're right that you shouldn't see a huge deviation, but you can't say that 35g of salt mix into 1L of water will give you 35ppt seawater. A significant portion of the salt mix is actually water. Many of the salts in the mix are hygroscpic and absorb water from the air. So it normally takes a bit more than 35g of salt mix per liter to get to 35ppt. Sometimes significantly more. How much more depends on the salt mix and how it has been stored and what sort of humidity it has been exposed to. In short, there's no way to know how much of your salt mix is actually water. So measuring salt mix by weight can be a good way to get close, it is never a good way to ensure it is right.

[BlueCoast]

It doesn't seem like there is a significant difference when using different light sources. In a completely unscientific test, I just grabbed my refractometer and tested using a few different lights I have laying around:

Salt water
normal light I use = 1.026
rayovac led headlamp red = 1.026
rayovac led headlamp blue = 1.026 (that one is hard to read)
rayovac led headlamp white = 1.026
some teal led light from macna = 1.0255
ATI Blueplus T5 bulb = 1.026

I got similar results with fresh water, I noticed the blue was under the 1.00 line by about .002

I did see a bit of variance, but not much, maybe .001. Possibly a couple more people can try and see if they get similar results.

[Randy Holmes-Farley]

It doesn't seem like there is a significant difference when using different light sources. In a completely unscientific test, I just grabbed my refractometer and tested using a few different lights I have laying around:

Salt water
normal light I use = 1.026
rayovac led headlamp red = 1.026
rayovac led headlamp blue = 1.026 (that one is hard to read)
rayovac led headlamp white = 1.026
some teal led light from macna = 1.0255
ATI Blueplus T5 bulb = 1.026

I got similar results with fresh water, I noticed the blue was under the 1.00 line by about .002

I did see a bit of variance, but not much, maybe .001. Possibly a couple more people can try and see if they get similar results.

You did these without recalibrating in between?

FWIW, several people have reported differences to me.

[BlueCoast]

Correct, I used the same water sample and just changed light sources.

I just tried a few different lights walking around the house, halogen desk lamp, incandescent piano lamp, etc... I could not make out any difference between them. Less than a .001 difference I can't see on my refractometer. I'll check it tomorrow under natural light to see if that makes any noticeable difference.

[Ontheway]

Since many of us do not have access to ec meter, I believe salinity of most tanks are fluctuating to a degree. On the other hand, there was a recent thread somewhere here showing how much surface water salinity differs in the ocean. So, slight variations should not be deadly to our tank residents.

[apexkeeper]

Wow I never even thought about this. After reading this I ended up wandering around the house with my bottle of calibration fluid checking it under every lightbulb in the house. :lolspin:

Luckily I really didn't see any difference. However I then went outside and checked it under natural sunlight and there was a noticeable change.

[Randy Holmes-Farley]

Thanks, folks who are experimenting! :)

[Metal Man 1221]

Thank you for this Randy. :thumbsup:

I will be doing some experimenting of my own as well.

[DSMpunk]

DSMpunk; Although I have no access to a calibration fluid, if its a liquid with properties of normal seawater, remember water part will evaporate as long as you keep the cap open. It may skew your calibration. What you can additionally do is, read info of the salt provided by producer, on what amount of salt needs to be dissolved for 35ppt. Prepare a small amount of mix in the measurement room then check it with refractometer. You shouldnT see any large difference to calib. fluid. I do it with 70grams of salt and a two liters of rodi water.

The fluid I have is one of those squeeze bottles that in theory should not evaporate much even with the cap off. That said I always keep the cap on :)

I was thinking the difference may have been the fact that I was using sunlight to read from the refractometer vs using artificial lighting in the basement.

[Metal Man 1221]

While I'm on lunch, I came home to give this a try with a Phillips Hue. I did notice some slight variation, but I do not have much time to spend working on this at the moment.

[hogfanreefer]

Last night I calibrated my refractometer out on the front porch in diffuse daylight (per my routine) and then walked around to different rooms and rechecked. My kitchen which has "daylight" florescents was spot on but our bedroom that had more blue florescents was about 0.001 off. Most of the other spots were between those two.

Thanks for sharing this Randy.

[reelredfish]

I thought this may explain my issue between mine and my salinity probe.

However i just recali rated under a light in the kitchen. Checked my tank water the. Under the light in the kitchen.

While i think Randy is 100% correct here it did not explain my issue.

My two devices are still aways from each other.

[Ontheway]

The fluid I have is one of those squeeze bottles that in theory should not evaporate much even with the cap off. That said I always keep the cap on :)

I was thinking the difference may have been the fact that I was using sunlight to read from the refractometer vs using artificial lighting in the basement.

Your issue may not be related to light alone. Can you please rethink about temperature conditions of places where the refractometers calibrated in? it may be due to failure of ATC feature.

People who are testing their ref. under different light sources, I think you should bring different lights to refractometer, instead of taking ref. unit to light source, to prevent temperature variations room-to-room. Of course, unless you are 100% percent sure its equal.

Just another point, warmer fluorescent lights (high kelvin) are warmer in reality, too. The rays from warmer lights heats up surface they hit. Therefore do not come close to light source for measurements. Sunlight calibrations are also needs to be done under indirect conditions.

Regards,

[BossHoggin]

shnikey

[ecook280]

Never would have thought of that

[dkeller_nc]

This is one reason I always suggest that beginning reefers pony up the extra $50 and buy a digital refractometer like the ones that Milwaukee makes. It has its own (yellow-orange) LED light source, automatic measurement/temperature compensation and is calibrated for seawater.

My only complaint is that it cannot be calibrated with a refractivity standard, only zeroed with purified water. That said, I precisely prepare my own refractivity standards with lab equipment, and the unit I have is always spot-on at 1.026.

Milwaukee does warn you in the instructions that measurement in really bright incident daylight will affect the readings, which makes sense given Randy's graph. Of course, preventing that is as easy as shielding the measurement cell with your hand when you press the "read" button.