How to measure free or surface aggregate moisture

The moisture meter you own or buy will define how you measure your aggregate moisture levels. While most sensors measure bulk moisture, some measure only surface moisture. In either case, though, they can be calibrated to read surface or free moisture. This is the moisture that reacts with the cement in the mix and, therefore, the only moisture you need to worry about.

The free water is the water not bound inside the granules of aggregates. The “bound” or internal water is absorbed into the aggregate granules. When this water is at its maximum but there is no free water, we have the condition called Saturated, Surface Dry (SSD). This describes the condition where the granules of aggregate are filled to their capacity. Your concrete mix design is always based on aggregates being in the SSD condition.

A colleague asked me why his aggregate moisture meter reads a (low) moisture reading but when he adds the correct amount of water to the mix, according to the mix design, it turns out too dry. And when he adds more water to give the right slump, it exceeds the maximum water/cement (w/c) ratio for the mix design. If you have experienced this effect, read on.

The solution

You should always calibrate your moisture sensors to read 0 percent moisture when the aggregate is at SSD condition. If the aggregate is bone dry (no internal moisture), the granules will absorb water from the mix, which then becomes bound in the aggregate. That is why my colleague above had to add more water than the mix design called for. This, in turn, will give a false w/c ratio when your batch controller totalizes the water absorbed by the aggregates plus the added water. The reason is that the water absorbed into the aggregate is not “free water” for the purposes of the mix design. The w/c ratio is probably correct, but it comes out incorrect on the batch report.

The only solution to this situation is to spray the coarse aggregate piles to prevent them drying and allow them to absorb water. This, in consequence, creates the SSD condition. Most companies in hot areas do this as standard practice. You should not spray your sand pile, however, since it normally has some free water already, from occasional rainfall. If spraying the aggregates gives too much free water for a low w/c ratio mix, you should drain each load before transferring into the plant bin. Usually this is not necessary, because you will need to add more water in the mixer.

We have investigated the absorption effect by testing moisture sensors in tropical countries. The hot sun evaporates moisture on the surface of the pile and eventually reduces the average aggregate moisture. This, in turn, makes the coarse aggregate “thirsty” for water because it is no longer SSD. We have also found that for sand with moisture lower than 4 percent, the internal moisture (usually assumed to be 1 percent, but this varies with the type of sand) drops to less than half this value when surface moisture drops to between 2 percent and 3 percent. The same applies for coarse aggregates but the absorbed (internal) moisture in this case is usually much less than 1 percent when saturated. The actual value depends on how much water the particular stone material can absorb.

Moisture sensor types

The resistance-based sensor type is rarely seen now, because it can be very inaccurate. Adequate when sand is freshly washed, it shows up to 10 times the true moisture when it encounters material from the top of the pile that has dried out and been wetted again. This is due to the internal mineral salts that leach out during the drying process (the “surface crust” that forms in these conditions).

The capacitive types are somewhat better but are still affected to some extent. The near-infrared type of moisture sensor (Teconer, sold by Polarmoist) reads only surface moisture, since it measures the light reflected by the free water around the aggregate granules. The TDR type (Sono-Vario, sold by Mesa) sensors and all regular microwave sensors, such as the Hydronix models and Scale-Tron’s AquaSense 2280D, measure deeper into the material stream. The Sono-Vario measures deepest and is also less density sensitive. These sensors all measure total moisture, including internal moisture. However, the calibration process, if done correctly, eliminates the internal moisture. The Sono-Vario sensor claims to measure aggregates up to 1-1/4 in. to within 0.2 percent moisture level, which is impossible with other sensors. For “regular” microwave sensors, the maximum aggregate size for any real accuracy is 3/8 in. and even 1/4-in. aggregates can give errors of up to 0.5 percent.


Weighing the aggregate sample before and after the oven-dry test. PHOTO: National Precast Concrete Association

When calibrating moisture sensors, it is important that you saturate the aggregates first, to bring their internal moisture up to SSD conditions. Some producers have trouble doing this because they don’t want to spray and waterlog their aggregates, sand especially. The other common mistake is to bake the aggregate to bone dry before weighing. This drives off the internal moisture as well as the surface moisture. Both approaches are wrong. During tests at a concrete plant we found an expert in oven-dry testing who had an ingenious way to check the progress of the bake-out. He used a piece of 1/2-in. thick plate glass held over the pan for a few seconds. If it misted, it indicated there was still surface or free water. When no misting occurred, he shut off the heater. As a result, he got a very precise SSD condition for weighing the sample. Using his method, we could perform oven-dry tests to an accuracy of 0.2 percent or better.

AquaSense calibration table

The moisture sensor calibration process automatically takes care of the absorbed (internal) water. You should take at least two samples, and preferably more, over a range of high to low moistures. Because of weather, this may take several weeks. Use any convenient scoop and for each sample, from a single batch, keep scooping into a bucket until the flow stops. Take at least five scoopfuls and have someone simultaneously read the moisture from the sensor. For our calibration example, we have used the AquaSense 2280D moisture sensor. The oven-dry test results, plus the moisture sensor readings, are entered into the sensor calibration table on the connected computer. The software will then find the best-fit straight line and perform the calibration. If you do the oven-dry tests correctly, halting at the SSD condition, the internal moisture will not appear in your readings after the calibration is complete. Subsequent readings will be the “SSD readings” or the values over and above the SSD condition.

Calibration failure

If your calibration fails, you are probably not selecting a big enough range of moisture during your tests. If the range from driest to wettest is only 2 percent and your oven-dry tests have a ½ percent error, the slope of the calculated line can be off by a wide margin. Consequently you will get huge errors when the material moisture is outside the range of your samples.

Robin Shepherdson is Owner/President of Scale-Tron Inc. and has also developed most of the company’s products. An engineering graduate from the 1960s, he has been a professional engineer all his life and still visits jobsites regularly to help and train dealers and customers. His blogs on this and similar subjects are posted on the company website, Recent blogs include a shorter version of this article entitled “Surface moisture or free moisture” and “Find the best moisture sensor for your application.”


After the sampling and oven-dry tests are complete, the results and AquaSense sensor readings taken during sampling are entered into the calibration software.