Time to get REAL with RH Probes

One of the hotly debated topics with RH measurements is: How much moisture in total volume is REALLY being measured when these are set within concrete?

The Basics

RH Probes measure water vapor. These do not and cannot measure water in any other form.

ANY water present in liquid form remains unmeasurable with ANY type of humidity measurement device or technique.

Water vapor/humidity exists only in open air space. Moisture volume in an open air space can ONLY be calculated when the amount of air space is known. For example: If the humidity is at 75% and a probe is stuck into a small quart container and then stuck inside a 55 gallon steel drum, the humidity measurement simply says 75% and CANNOT differentiate size or area being tested. According to the RH sensor, both the quart container and the 55 gallon drum contain the same volume of water.

WE would know better because we can see it (the volume difference between the quart container and 55 gallon drum)....what we CAN'T see and CAN'T know is how much space is being measured within the concrete.

What Effects Humidity Measurements?

The easiest to understand and well documented is temperature. For example, if the air temperature is 85oF (29.44oC) and the RH is 65%, but the temperature decreases to 65oF (18.33oC), the RH at 65oF will be 100% with NO change in moisture volume.

If the concrete temperature is at or lower than 71.9oF, the concrete is at atmospheric dew point.

NOTE: I used 65oF-85oF since that is the temperature range of installation acceptance by many manufacturers

Depending upon the surface permeability, alkalinity, etc., the concrete surface could be near saturation or completely saturated.

When concrete enters the fray, the considerations become increasingly complex.

In the presence of hygroscopic materials, these materials; aggregate (which vary in hygroscopicity) unhydrated cement, secondary cementitious materials, plasticizers, cement grinding aids; these will absorb and/or adsorb moisture. The more moisture these materials absorb/adsorb, the higher the moisture content can be, even if the measurable RH remains constant.

More dramatically is the presence of alkaline salts. Alkaline salts have a variety of critical humidity thresholds, where these influences do NOT exist in an open or non-hygroscopic system where no alkalinity exists.

The critical humidity thresholds of alkaline salts are extremely low, meaning these WILL absorb moisture from the air, even in arid to semi arid conditions.

Where alkaline salts really get tricky is that each has a saturated humidity threshold as well....which means a saturated alkaline solution will lower measurable humidity, irrespective of the saturated liquid volume.

This is how calibration solutions were discovered and how these are used. Certain salts at full saturation will have a predictable RH value, which allows the accurate calibration of moisture testing devices and equipment.

Industry Standard Calibration Solutions

The most common industry calibration solution is sodium chloride and water. This is the most common due to a combination of safety, ease of use, availability and RH stability across a wide range of temperatures.

It is common to see undissolved sodium chloride salts within a calibration solution container. The reason for this is, excess sodium chloride will not effect the RH stability, but dilution of the solution can effect the RH stability.

These are all the stable RH of the different calibration salts based on 70-75oF temperature. Each of these is a saturated solution:

Lithium chloride: 11% RH

Magnesium Chloride: 33% RH

Sodium Chloride 75% RH

Potassium Sulfate: 97% RH

To evaluate and calibrate devices, each of these solutions can be used for cross-referencing and establishing the stability of the testing device/sensor across a wide spectrum of RH levels.

RH, Air Space in Concrete

The graphic at the beginning of this article allows for calculation of water volume as weighted against the weight or air.

In an earlier research, it was noted that even in the measurable air space, RH/water vapor made up what should be considered an insignificant volume of moisture.

In an air space of one cubic meter, at 75oF, it would take slightly more than 264 gallons of liquid water to fill that air space.

Conversely, it would take slightly less than 2 ounces (0.06 litres) of water in a cubic meter of space to produce a 100% RH.

0.06 litres of water....2 ounces...how in the world did entire industries (flooring, coating and concrete) get duped into believing such a minuscule amount of moisture could cause a coating or flooring failure????

Now, think of the tiny pores in concrete...heck, even large pores....how many thousands of square feet of concrete would be needed to produce a cubic meter of air space...and then over that immense volume of material...2 ounces of water could produce 100% RH.

Perspectives in Moisture Testing and Practical Demonstrations

RH isn't the only "fact" that feels embarrassing when practical applications are applied.

When the calcium chloride testing was the predominant methodology, I had been an advocate of these tests, believing the experts I learned from, until I found there were no correlations between the vapor emission levels and flooring/coating failures.

Back then, 5 lbs. was considered the conservative upper moisture tolerance of floor covering materials.

To place this in perspective, the calculation of 5 lbs. was weighted over a thousand square feet over a 24 hour period.

After discovering and then proving just how ridiculous this was....as part of a class demonstration...I would take a group out to a concrete surface, together we would mark off 1,000 s.f.

I would then take exactly one gallon of water, which would be the vapor emission equivalent of 8.34 lbs over a 24 hour period in accordance with ASTM F 1869 and pour it into a small hand-held pump-up sprayer. NOTE: Vapor emission level is based on water weight, NOT pressure as was so often misrepresented....

I would challenge anyone in the class to take the sprayer with the gallon of water and try to dampen the entire 1,000 s.f. of concrete surface before running out of water.....no one was ever successful...usually the water ran out about halfway or sometimes three quarters of the way through. NOTE: in a later demonstration, I would take exactly 10 pounds of water and do the same demonstration....where the 1,000 s.f. STILL couldn't be dampened before running out of water...I used the 10 pounds to emphasize that was TWO and even THREE times the upper limits of the stated moisture tolerance of a successful flooring installation.

I pointed out that this volume of water wasn't sufficient to even dampen the concrete surface within a 15 minute time table...yet we are to believe this amount of water was somehow causing floor and coating failures as this emitted through the surface within 24 hours, when it wasn't possible to even dampen that same amount of surface area within 15 minutes? NOTE: It is a fallacy to believe even that level of moisture will emit from concrete.

Calcium chloride, used in this test method (ASTM F 1869) attempts to reduce the RH under the test dome down to 18% RH and if left will continue to absorb moisture until that equilibrium is reached or the salts became saturated. The alkaline salts native to concrete have a lower critical humidity threshold than does calcium chloride, meaning there is moisture the calcium chloride CANNOT draw out of the concrete. As the alkaline salts in the concrete become increasingly concentrated, this lowers the RH and restricts the ability of the calcium chloride to draw moisture from the concrete.

As concrete becomes more alkaline...BOTH RH and calcium chloride methods become less effective in measuring moisture content in both vapor and liquid form. NOTE: This does NOT mean these devices and/or methods are defective in any way...this simply points out the data has been misinterpreted and incorrectly represented. If conducting a forensic study, in my experience, such results can produce essential data not discoverable using any other method(s).

Back to the Basics

Back in 2009, I began a quest to try and find SOMETHING that could at least give a reasonable correlation between moisture levels and floor/coating failures...

Mission accomplished!