Official Luthiers Forum!

As promised Simon, David Collins, and I collaborated to attempt to demystify some questions surrounding hygrometers, RH testing, accuracy of specific products and more.

Please know in advance that the results that were obtained during this testing are subject to error and the entire process is such that there is a lot of room for error... Nonetheless with averaging and being as careful as we could be it's believed that the results obtained are probably pretty close but the entire point here is that YMMV. RH testing for great accuracy requires some things that we didn't have such as an absolutely stable environment with no drafts, humans breathing in the wrong direction, etc. - it's that sensitive....

This leads me to the first point - RH testing is dynamic. You can test three times in a row with the methods that we used (wet-bulbs) and get three slightly different results... This does not mean that we were smoking the drapes but it just illustrates how sensitive this testing is to environmental variables.

Here are the instruments that we used and the hygrometers that we tested:



The hygrometers are two Cole-Palmer industrial hygrometers that were manufactured around 1989 and used for industry. These were purchased by David from the University of Michigan surplus/stores and work very well. The sling psychrometer was also used but without slinging it instead mounting it near a blower.

The hygrometers are an Abbeon that had not been calibrated for 6 months, a white older Stew-Mac style, and three Caliber III's two from me that are 3-4 years old respectively and one from Simon that is brand new.







The following is some of the data obtained:

Blower/hygrometer - wet = 53.9, dry 68.2

CP Hygrometer #1 - wet 53.6, dry 67.2

CP Hygrometer #2 - wet 12.6C (this CP hygrometer is in Celsius), dry 20

In addition to using the charts that came with the instruments we also double checked our results against a web site that factors in, with user input, barometric conditions at the time of the testing.





One of the many areas where error can occur is in reading the thermometers. Accuracy is important so we attempted to estimate to the tenth of a degree even though there were no .1 graduations on the thermometers.

Using an average of readings and results from the tools we had available to us our results at the time that we checked are as follows:

All three Caliber III's were 4 - 5 degrees low.

The Stew-Mac old style hygrometer was 6 - 7 degrees low.

My Abbeon was 3 - 5 degrees high (and was quickly physically calibrated to now be in accordance with the testing).

Additional thoughts: The Caliber III that Andy Birko brought by last week tested nearly dead accurate and we can't explain why.

I want to talk with Simon before I do more testing with the substance that he sent me.

Results could vary simply because folks were standing too close to the instruments while testing.... Again no claims are being made that this testing is all that....

Regarding if a Caliber III loses accuracy with age I think that in our sample we can dispel this since my 3 - 4 year old ones read the same as Simon's new Caliber III.

Lastly a great deal more testing that is far beyond my abilities is needed to address the idea of accuracy in specific ranges. We know from experience that at say 40% RH the accuracy may be one thing and entirely different at say 55% RH. How to design and conduct a test that addresses this needs an engineer or scientist's abilities of which I don't have.

For me I now know that my Caliber III's are reading 4-5 % low at 40ish%RH which will permit me to make a mental note and use them for quick glance reads.

Abbeons are known to require calibration as much as every 3 months so I will be continuing to calibrate my Abbeon as per the test here going forward.

Thanks for looking.

Thanks Hesh.

Did you find any de-calibration happening with your Abbeon? Mine hasn't changed a bit since I got it about 14 months ago, and I tested it 3 times so far. It was perfectly calibrated out of the box, or perhaps 3 points high, and it scores the same to day. I can't decide if it is perfect or 3 high, depends on how I read a half degree on my analog thermometer. I think it is good enough for me though

_________________
Build log

Hi Hesh,
That Caliber III I sent you was a couple years old. It was a solid 3 points lower than it should be with the Potassium Carbonate test. It is also 3 points lower than the batch of 5 new Caliber III's I received a few weeks ago. So that means your results are pretty spot on -- that is good to know. Give me a call and I can tell you how to set up the salt test.

Thanks for your effort in all this!!!

Thanks for all this effort, Hesh, Simon, David.
It's interesting stuff.
First, I'm assuming that the water wells were filled with room-temperature water, not water straight from the tap.

Let's look at the wet/dry bulb results....
Using the calculator at http://www.ringbell.co.uk/info/humid.htm
(Please double check my results as I may well have made a typo or calculating error...)

It shows that the RH results from the three wet/dry bulb instruments were:
Sling/blower RH 36.2%
ColePalmer#1 RH 39.1%
ColePalmer#2 RH 40.9%
So, how to average these results- assuming averaging is the way to go??
Average all three and we get 38.7% avg. This makes the CP units 'high' and the sling/blower unit 'low'.

Or, should we assume that the CP unit results are better since they are closer to each other, and take 40% as the 'true' avg RH? If we do that, then the sling/blower unit is 3.8 RH % units 'low'...

It seems that (using 38.7% as the 'average' or 'correct' RH value) the measured RH values from the other instruments were about (I'm guessing here):
Calibers were showing ~34% RH
Andy Birko's Caliber III ~39% RH
StewMac was showing ~32% RH
Abbeon was showing ~42% RH
??
So, if we assume the CP units are showing the 'correct' RH, then the 'errors' are
Sling/blower ~4 RH %units low
Calibers~ 6 RH %units low, Andy Birko's Caliber III 1 RH %units low
StewMac ~ 8 RH %units low
Abbeon (before 're-calibration') ~2 RH %units high

So the conclusion here would be that, if the sling/blower unit is typical of wet/dry bulb setups in most shops, that they will provide results about the same as the Caliber III units tested here.

Another conclusion would be that the 'worst' that any of the units would do would be an error of less than 10 RH %units compared to the 'true' value.
But are we really that concerned with the 'true' RH value in our shops? Or is it the 'change' in RH that's the issue? For me, it's the latter, generally- I'd prefer to work at a consistent RH level from day to day, as long as the RH is within an acceptable range, say 40-50% in my climate. (or 35-45% in Michigan?). Probably any of the units tested here would show changes in RH that would cause a problem.

So, what are we worrying about??


Hesh, Simon, and David Collins: thanks again!

Cheers
John

Just to clarify, Hesh: you wrote that the Caliber units were 4-5 degrees low. Was that a mistake? It came right after you were talking about reading thermometers, so it's confusing if you mean %RH, which I'm supposing is what you meant. How accurate were the digitals on temperature, BTW? Mine always agree within a fraction of a degree.

_________________
Howard Klepper
http://www.klepperguitars.com

When all else fails, clean the shop.

Upon further review by the officials in the video booth I think that I meant %... Howard we did notice that temp wise everyone (all of the hygrometers) were very close if not exactly the same.

John - what are we worrying about? Nothing But I am out of my element describing this kind of testing so please excuse my errors and poor use of terminology. My hope is that the data can be properly interpreted by folks like you so thanks for the analysis, John!

Simon that is good to know too and thanks for your efforts as well. Would you please call me at your convenience to discuss next steps and the procedure for this? Many thanks.

Alexandru my friend that is just the thing now isn't it - when is it good enough for us even though we are not Gibson... My Abbeon does drift and needs to be checked every 3 months. I know a guy who was tasked with calibrating all of the Abbeons at Elderly Instruments (One of the coolest guitar stores in the US). He did this every 3 months and found drift of as much as 4 degrees F in just 3 months time.

I guess I got lucky with mine then (for once). Or it might have been drifting up and down

_________________
Build log

I will be very interested to see the results when you do it with the Potassium Carbonate.
At the moment you still do not know for certain which unit is reading correctly if any.
It could be one of the wet /dry intruments, it could be none of them.

I would tend to take the view that true RH rather than variation IS what we are concerned with.
If you allow 40-50 in your shop, but your meter reads 10 low you could be bracing at 60%.
The guitar then leaves the shop and goes to a guy in a dry climate whose rh meter reads 40%so thinks it is fine but is really 35% . You then have the guitar living in a climate 25 below what it was braced at.

I do not yet have a climate controlled shop so I have to pick and chose when I will brace.

Jeff-
That could happen, I suppose.
From an online article in Entrepreneur magazine:
http://www.entrepreneur.com/tradejournals/article/158623368_2.html
"Vicente Carrillo. Located in Casasimarro, midway between Madrid and Valencia, this small workshop has been producing guitars continuously since 1744. Other than a band saw, all work is still executed with hand tools, and production averages 15 instruments per month. The company is run by Vicente Carrillo Casas, who represents the eighth generation of a family that has been making guitars since 1836. ................
Carrillo recently completed a modest expansion of his shop, and he has adopted a few elements of modern technology. For example, the shop's humidity is kept at a constant 60% (compared with the region's natural 30%) to prevent the wood from twisting and to ensure the optimal conditions for the multi-layered, ultra-thin French lacquer finish. .......
Though Vicente Carrillo's prices, retailing from $3,800 to $8,900 in the U.S.,................. he is heartened that instruments made by him and his ancestors in a town of just 2,500 have enchanted players all over the world, from serious students to the inimitable Paco de Lucia, who recorded his landmark 2004 flamenco release Cositas Buenas on a Vicente Carrillo guitar. He continues to reach out to new dealers in the U.S. and other nations at the winter NAMM show and the Frankfurt Musikmesse, ever committed to the principle of "keeping production low and quality high." "

Cheers
John

[quote="JohnAbercrombie"][quote="Jeff Highland"]

. For example, the shop's humidity is kept at a constant 60% (compared with the region's natural 30%) to prevent the wood from twisting and to ensure the optimal conditions for the multi-layered, ultra-thin French lacquer finish. .......

Seems like a recipe for disaster, if the journalist has got it right.

Cheers
Jeff

This is great Hesh... thanks.

Let me throw out a few random questions.

Do you know where you can find cheap good quality thermometers for this test. I have looked at supply houses but long thermometers are $40+ each... Ebay wasn't much good when I looked and the quality is such a crap shoot.

Potassium nitrate... where do you get that? I presume this is a salt test with a closer testing value to what we are looking for than table salt of ~75%?

That sling hygrometer in a box is very cool. Can you show more pics of that close up so we can reverse engineer that? I have some parts laying around to make one (except the thermometers). How do they keep the cloth wet?

_________________
Brock Poling
Columbus, Ohio
http://www.polingguitars.com

Great test, Hesh! Have you tried a vaccum to draw air from the CP hygrometers? Motors generate heat and these fans seem tiny.

I tend to be a fan of the Potassium Carbonate test -- the reason is that the RH is directly related to a known physical standard. This eliminates the introduction of errors with improperly calibrated testing equipment or sensitive calibration methods -- both of which apply to hygrometer calibration.

There do seem to be some possibility of error introductions with salt test but most of these errors tend to be fairly negligible and most seem to not be relatable to a one-time setup and calibration test. Even with errors, it seems that the salt test will still give the most accurate calibration. However, I am not an expert in this subject but am just going on what I've read (I do have a thorough background in chemistry).

You can buy anhydrous Potassium Carbonate at a number of different lab suppliers (just do an online search). The RH of a saturated solution of Potassium Carbonate is 43.1% -- perfect for the lutherie crowd.

Filippo yes the accuracy in specific ranges is an issue and my original post asks for ideas for designing a test or tests that might shine a light on this issue. I know that my Abbeon and Caliber III's not only don't agree with each other at say 40% RH but the difference in their readings gets greater in other RH settings.

CW I'll take a look at the Psychro-dynes that we used and see if the motor heat has a possibility of skewing the results. You are correct though that EVERY little thing matters including where the humans in the room are breathing.... We saw differences in the reading simply by not being careful where we were... Also RH is dynamic and constantly potentially variable if you have a drafty room. It's a moving target...

Brock the thermometers that I purchased for my own shop-made wet-bulb are so crappy that I can't recommend them to anyone. My biggest beef is that the mercury line is so very thin that even with my stinkin Borg Opti-visor on you can barely see the line.... I would appreciate a good source too if you find one.

I'll get some more pics of the sling psychrometer for you today so you can reverse engineer one.

Jeff as soon as I can hook-up with Simon again I plan of using the Potassium Carbonate that he sent me and continuing the testings - please stand by...

Thanks everyone.

Simon-
Thanks for the info.
No need to source anhydrous potassium carbonate if you are just making a saturated solution in water. However, that may be the most common form available.
Also, be a bit careful in dissolving it, since heat is released- add the solid to water.
And, you did mean "The RH of air in equilibrium with a saturated solution......"

Using some standard - and the potassium carbonate one seems the best- is certainly the key to deciding which (if any) of the test instruments is giving the 'right answer' for RH. 'Most common', 'most expensive', 'best looking' don't really count in science!

Now Hesh has to source an aquarium for the Phase II of the tests!

Cheers
John

A little tip (that I'm willing to share now that one's on the way ) If you search on e-bay for psychro-dyne, you'll find some nice psychrometers for a pretty good price. searching psychrometer gets you hits but not for a psychro-dyne. They're going for cheap! I just picked one up that the owner claims works for $40 shipped!

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

Hi John,
Why get anhydrous? -- because if they go to the trouble to make certain that it is completely dry then it is likely a decent grade of material. Certainly not necessary since it won't be anhydrous once you put it in solution but I think it is a nice qualifier for making sure the material was decently prepared.

And yes, I meant a 43.1% RH for the vapor phase of the saturated salt solution. In other words, you have to do this is a sealed chamber!!! Also, you can "quickly" add the water directly to the salt as it doesn't release enough heat for concern. The solution will get a little warm but nowhere near enough to knock.

Off topic:
Simon-
Good point. Anhydrous doesn't just mean 'dry' actually, in some cases. Sometimes it's a question of 'water of crystallization' as I recall. For instance the (crystalline) preparations of copper sulfate (pentahydrate?) can quite a bit purer than the anhydrous preparations (fine powder). The 'quality' is more indicated by the 'grade': Technical, Lab, Reagent, etc. Anyhow, for this application it doesn't matter much what form you get the stuff- so long as it's the right compound! Even fertilizer grade would do the trick.


re: sealed container- How long do you find it takes to get the system to equilibrium so that you are getting consistent RH readings? Do you do this in a plastic bag, or...?

John

After Andy bought his psycho - dyne... I bought one too off ebay but fear not there is another one listed that you can "buy now..... " for just a stinkin $1,000......

http://cgi.ebay.com/ETC-ENVIRONMENTAL-TECTONICS-CORP-CP-147-PSYCHROMETER-/350135229484?cmd=ViewItem&pt=LH_DefaultDomain_0&hash=item5185af9c2c#ht_2222wt_1129

This is exactly the same unit that we used in our tests and that Andy and I just purchased for under $40 delivered....

I had the sincere pleasure of speaking with Simon a few minutes ago and he filled me in on the procedure for conducting the test with potassium carbonate which I will do when I find some suitable containers. Thanks Simon!!!!

This is certainly a far cry from the days when Hesh estimated his shop RH by observing the wrinkle pattern in the roll of toilet paper