Atago PAL-22S Digital Refractometer - dialog with the rep.
I had a few questions about the PAL-22S and contacted the rep. Here is the dialog - I figure if I have had these questions then someone else has. I cut and pasted it from my email. WARNING - ITS A LONG POST.
> [ Inquiry ] Technical question - PAL 22s
> I am noticing some variation in test results based upon how long I leave
> the honey
> sample on the testing stage. I saw in the literature that came with the
> product that it is
> accurate to .2%. The literature said that refractometer. calibration
> water. and sample all
> be at the same temperature - but did not say how long it should take for
> the sample to
> warm up on the testing stage.
> Some results: Yesterday I calibrated the meter in a room where the water
> had sat
> overnight with the meter. I then fetched a sample from a different room
> when I applied
> the sample I got the following readings from the same sample.
> 18.2 % at 26.5 degrees - it sat for one minute before test.
> 17.8% at 26.6 degrees - it had sat for 5 minutes.
> 17.7 % at 26.9 degrees - it had sat for 14 minutes.
> I have the following questions I would like answered.
> 1) Are these results appropriate and consistent with the theory behind the
> 2) How long do you recommend the sample dwells on the stage before the
> accurate reading is obtained?
> 3) Do you have any other suggestions on sampling technique.
> 3) How can I be assured that the daily "zeroing" of the PAL-22S
> is enough to recalibrate
> it accurately enough that I don't mistakenly sell honey based on an
> erroneous test
> 4) Are there other factors that could produce results that are erroneously
> high or low
> that I should know about?
> [ How did you find our website? ] I searched it out
Good morning. I hope you are well. I received your e-mail regarding the PAL-22S and I will be happy to help you with your questions and offer any advice I can on the best ways to measure with the PAL-22S.
There are a few things that you can do to ensure you get the most accurate readings as quickly as possible. First of all, I wanted to ask you a few questions regarding the readings that you did receive.
1. Before the PAL-22S, how were you measuring Honey Moisture?
2. Would it be possible to do a side-by-side comparison of the two methods?
3. You received readings of 18.2, 17.8, and 17.7. Are any of these in the realm of what you are looking for from your honey?
4. What was the approximate temperature of the room where the honey was stored?
Now, I do want to give you some tips regarding how to measure honey on the PAL-22S. Many people do not use enough sample. With an analog refractometer (one you look through) you only need a very small amount of sample. On the PAL-22S, we recommend you fill the sample stage up to the top of the metal. This ensures that the honey will have enough weight to push the sample against the prism surface.
Since honey is so thick, we also recommend waiting a few seconds for the sample to spread evenly over the prism surface. This also allows the sample to start acclimating to the prism/room temperature. Since this unit has Automatic Temperature Compensation, the prism and sample have to be at a stable temperature to avoid fluctuating readings.
One other thing that I would like you to try is to put the sample on the prism, wait for 1 minutes, and then keep pressing the start button every 10 seconds or so until the readings become stable. With samples like this and other very viscous samples, we recommend taking multiple readings until the readings become stable. Once the readings are stable, that means that the prism and the sample have the same temperature and the sample is evenly spread.
To answer your question regarding calibration, water is sufficient to zero set the unit every day. As long as the prism is thoroughly cleaned and dried before the zero setting, and you do not receive any error messages (Like “AAA” or similar) then the unit has been zero set. You could use a standard liquid on your PAL-22S (a solution that has a known honey moisture value), but that will just confirm that your unit is working well.
The only other factors that could result in erroneous readings would be low batteries or a dirty prism. The prism should be cleaned periodically with a cotton swab and isopropyl (or ethyl) alcohol. This ensures that there is no residue on the prism surface. Once this cleaning is done, please rinse and dry the unit completely before zero setting again and testing your samples.
I hope this information is helpful. If you need anything else at all, please do not hesitate to ask. I will be happy to help.
Wes LeMay Jr. (Mr.)
Technical Customer Service
ATAGO USA, Inc.
Wes, thanks for getting back to me. I'll answer your questions before a little discourse and a few more questions:
1) Before I bought the PAL-22S I was measuring moisture with an analog refractometer I bought off Ebay (new) that I thought was hopeless. I could not read it very well, and it was adjustable by a screw. I used it for a season and hated it because repeat tests were not consistent. That is why I bought the PAL-22S.
2) It might be possible if I can find where I put it, although I think I am out of the calibration fluid.
3) Yes these numbers are in the right realm. Less than 18.6% moisture is classed as USDA Grade A honey. However, I prefer a wide margin of error and am happier with honey that is 17.8% moisture or less. What concerned me was the variability in these measurements, and thus my inquiry. If the range was from 18.2 to 17.7 then how would I know which reading was most accurate?
4) The room where I stored the honey (my kitchen) was about 75 degrees F.
I very much appreciate your advice it makes sense to me. I have the meter, the "zeroing" water, and the honey in a box room with a dehumidifier running.
Tonight I tested a sample, filling the sample stage as you said and then let it sit a minute to even out these are the results every 10 seconds after 1 minute. when I zeroed the PAL it read 29.1 degrees.
18.2 @28.6 (degrees C)
After 5 minutes I repeated the test every 10 seconds for 10 more tests and the results were all 18.3% with the temp varying between 28.7 and 28.8.
Achieving these results gives me a little more confidence. I have a few further questions.
A) Is there a standard product available from the grocery store - like Karo Syrup for example - that you would recommend as a control?
B) Is the leveling out of the meniscus of the honey in the stage a factor in the variation of results?
C) When you said that the prism and the sample have to be at a stable temperature for the Automatic Temperature Compensation (ATC) to work at first it confused me because I thought that the ATC system meant a sample of any temperature would be automatically compensated for. Now if my new understanding is correct, my former understanding was wrong, what ATC actually provides is Automatic Temperature Compensation for tests which take place in a range of temperatures, but the sample and the stage need to be at the same temperature. Do I have that right?
D) The zeroing done with water and meter kept in the same place is good for 24 hours. Suppose I zero at home at 70 degrees and then go somewhere where the ambient temperature is 80 degrees and then perform some tests - in the proper manner allowing equalization of prism and sample for a few minutes - Will the zero I performed at home be valid? If I am correct in my new understanding in "C" above I think there should be no reason to rezero at 80 degrees and that is the whole point of ATC.
E) I'm trying to understand errors that are possible by not testing properly - is there some consistency/pattern to the types of errors or do random results occur? In other words will a sample tested that is cooler than the prism always produce an erroneous result that is higher or lower than a sample in which the prism stage and sample are at the same temp.? The same question if the sample is warmer than the prism?
F) Lastly, it seems to me from the data I have obtained that in order for a sample and the prism to have equalized temperatures it is best to wait at least a minute, and for supreme confidence 5 minutes should be enough. Would you concur?
What was shaking my confidence before your response was that I didn't know whether to trust the first reading or the last reading. Wes thanks for your time.
It is my pleasure. I appreciate all of the information you provided.
It seems that the filling of the sample stage and waiting 1 minute helped to stabilize the readings within accuracy.
Regarding the temperature, I believe the way you originally understood ATC was correct. I think my explanation might have been confusing.
ATC compensates for the temperature of the sample and calculates the reading as if the sample were at 20°C. The PAL-22S takes a measurement at the PRISM level, though. So, if the honey is a different temperature than the prism, they need time to come into equilibrium. For example, if the sample is 80°F and the unit itself is at 72°F, as soon as you put the sample on the prism, the sample starts to cool and the prism starts to heat up. After a few seconds, the sample and the prism are the same temperature.
This is why if you put a hot sample on a refractometer that is stored at room temperature, you may see fluctuating readings for the first few seconds. The prism and the sample are trying to be equal in temperature, so one is cooling while the other is heating. During this time period, if the START button is pressed, the unit takes the temperature of the prism at that time and calculates the reading. But if the sample is not done heating or cooling (based on the sample temperature), subsequent readings will change (because the temperature is different the second test).
I hope that helps clear things up. In answer to your other inquiries from below:
A) We do not really have a standard to use as a control (except for liquid that we carry). However, if you find a sample that consistently measures on the PAL-22S a specific honey moisture, this can be used. As long as the Honey Moisture content is known (or investigated by testing), it can be used.
B) The leveling of the meniscus should have no effect on the readings. Only the sample adhering to the prism is important (so be sure there are no bubbles)
C) Explained above
D) Any time that you bring the PAL-22S into a different temperature, it is recommended to zero set it once it has had a chance to acclimate to the ambient temperature. This will ensure the most accurate readings. The readings may only be slightly off if the temp only changes by 5 or 7 degrees Fahrenheit. But if the temperature changes drastically, I would recommend re-zero setting the unit.
E) The most common results you will see when the sample and unit are not the same temperature are fluctuating readings. They may be higher or lower, but the readings will fluctuate and not be stable.
F) Unless your sample is VERY hot, 1 minute is generally enough to let the sample and prism come into equilibrium. It might even happen after 30 seconds or so, if the temps are very close. Taking multiple readings until the readings stabilize is a good rule. When I test, if I have the same measurement (within accuracy) 3-5 times in a row, that is the number that I trust.
I hope this information is helpful. Please let me know if you have any more questions. I hope the techniques work for you!
Wes LeMay Jr. (Mr.)
Technical Customer Service
ATAGO USA, Inc.
Wes. Here are what I predict are likely to be my last two questions. First, I notice that readings don't seem to vary with ambient light in other words a dimly or well lit room doesn't seem to make a difference. Why is that?
Lastly, I am active on several Beekeeping web forums; Would it be OK with you if I copy and post our correspondance for the benefit of other beekeepers with your email address and phone numbers removed? I think it would be helpful to beekeepers who own this Refractometer or are contemplating getting one. I have been a member of Beesource since 2008 and have not seen anyone talk about refractometers who works for a manufacturer.
Thanks again. Adrian.
The PAL-22S has an internal LED light source and that is all the light that you need for measurement. You could measure in a completely dark room and still get a reading (though the display would be hard to see).
In fact, if there is too much light, it could overwhelm the sensor and the unit could give a warning message of “nnn” on the display. This can happen in high fluorescent light and possibly sunlight as well. When this happens, you simply need to shade the prism surface with your hand and take a reading again. Your hand blocks out any of the unnecessary light, which allows the unit to read correctly.
I would not mind the posting of our correspondence, especially if it helps other beekeepers to use the PAL-22S with confidence. If anyone has a question or needs advice on honey refractometers (we also have an analog unit called the HHR-2N), I will be happy to help them, too.
Please do not hesitate to contact me with any questions that you may have. I hope you have a wonderful day.
Wes LeMay Jr. (Mr.)
Technical Customer Service
ATAGO USA, Inc.