Early-age restrained creep influences the cracking properties of concrete. However, conventional creep measurements require a large number of tests to predict the restrained creep as it is influenced by the combined effects of variable temperature, creep recovery, and varying compression and tension stresses. In this work, a double feedback control method for temperature stress testing was developed to measure the early-age restrained creep of concrete. The results demonstrate that the conventional single feedback control method neglects the effect of restrained elastic deformation, thus providing a larger-than-actual creep measurement. The tests found that the double feedback control method eliminates the influence of restrained elastic deformation. The creep results from the double feedback method match well with results from the single feedback method after compensation for the effects of restrained elastic deformation is accounted for. The difference in restrained creep between the single and double feedback methods is significant for concrete with a low modulus of elasticity but can be neglected in concrete with a high modulus of elasticity. The ratio between creep and free deformation was found to be 40–60% for low, moderate, and high strength concretes alike. The double feedback control method is therefore recommended for determining the restrained creep using a temperature stress testing machine.

Cracking due to temperature and restraint in mass concrete is an important issue. A temperature stress testing machine (TSTM) is an effective test method to study the mechanism of temperature cracking. A synchronous closed loop federated control TSTM system has been developed by adopting the design concepts of a closed loop federated control, a detachable mold design, a direct measuring deformation method, and a temperature deformation compensation method. The results show that the self-developed system has the comprehensive ability of simulating different restraint degrees, multiple temperature and humidity modes, and closed-loop control of multi-TSTMs during one test period. Additionally, the direct measuring deformation method can obtain a more accurate deformation and restraint degree result with little local damage. The external temperature deformation affecting the concrete specimen can be eliminated by adopting the temperature deformation compensation method with different considerations of steel materials. The concrete quality of different TSTMs can be guaranteed by being vibrated on the vibrating stand synchronously. The detachable mold design and assembled method has greatly overcome the difficulty of eccentric force and deformation.