The specific gravity of normal weight concrete is typically taken as about 2.4. But for a given mix design, the density should be consistent, and significant variation should be investigated.

Actual concrete density is a function of numerous variables. When a mix design is done, it is a "theoretical recipe" that should yield what is designed. Unfortunately there are variables in the batching process that can throw this off slightly, usually be about 1% or less, which is actually fairly good! When the batch is mixed, air gets entrapped in the concrete. Even when air is not entrained in the concrete by admixtures, the mechanical manipulation of the mix results in about 1 to 1-1/2 percent entrapped air. So if your theoretical density is 150 pcf and you have a 1 percent under-yield in the mix during batching, coupled with a 1 percent entrapped air content, your actual density will be 2 percent less than anticipated. Conversely, if you have a 1 percent over-yield and a 1 percent entrapped air content, those values will cancel each other and your theoretical density will match the actual density....accidentally!

In the UK we use 26kN/m3 for wet concrete.

entrapped air should not be a factor if the measurement is done correctly with vibration or rodding to eliminate the air. suggest using ASTM C138

Entrapped air is not eliminated when using either rodding or vibration as allowed by C138. There will still be some entrapped air, typically 1 to 1-1/2 percent.

perhaps there is a unit weight tolerance in C94 or perhaps on the AASHTO ... i don't remember, but i would focus on the ingredients because unit weight is just a composite number.....Each individual ingredient in the recipe has their own tolerances in their respective ASTMs. The biggest gray area in concrete (pun intended) is the water content. If you do the (controversial) microwave test and find that the batch water content is accurate, and you are confident in your field air and unit weight tests... the next step would be to look at what's going on at the batch plant including scale calibration, aggregate stockpiling, specific gravity of aggs.....NRMCA has a guidance document on all the things you can look at on the batch plant side and everything can affect the mix, also do the math and see if the theoretical uw was off by computation. it can be a rabbit hole if you have a reason to go down it.

dsg2....well stated. I've done quite a few batch plant inspections using the NRMCA checklist. It is comprehensive, including climbing inside the transit mixers to check build-up on the blades, calibrating the site glass on the water tank, checking the revolution counter, and all the batch plant items you mentioned.

I have also rejected loads of concrete because the revolution counter wasn't working.

Remember.....good concrete is a combination of water, cement and aggregates. Bad concrete has the same constituents.

Case in point - unit weight can be very important - mass concrete dam structure. Design and specs say - 2400 kg/m3 MUST be achieved. Okay. No entrained air. 40 or 75 mm diameter max aggregate. Maximum theoretical unit weight - no air voids - just at 2400 kg/m3. Actual unit weight, in place, as measured by hundreds of cores - 2325 kg/m3 with std dev of about 30 kg/m3 and COV of 1.2%. The concrete actually made could NOT achieve the design and specified unit weight required.

I would say that in order to determine, for structures where the weight of the concrete is of importance, that designs need to be based on the actual attainable unit weight with using the aggregates available and, as Ron say, 1 to 1.5% entrapped air. In the case noted, the air content would be about 3%. Of course, you would want this to have a bit of a factor of safety as well. You cannot always go by the unit weight of the cast cylinders (or cubes if that is the case).

Great advice from Ron, above....
Remember.....good concrete is a combination of water, cement and [b]aggregates. Bad concrete has the same constituents.

Concrete with or without reinforcing steel? Type of cement and aggregate? All these "ingredients" greatly affect the unit weight. For design, I use 150 pcf reinforced and 145 pcf plain.

www.PeirceEngineering.com