Every concrete sensor pitch for a decade has led with faster formwork stripping. This one leads with a $500,000 repair bill that happened because a crew trusted the cylinders instead.
GreenVibe, based in Ramat Gan, Israel, sells a multisensor unit tied to rebar before placement and left embedded in the element. It combines ultrasound, flexoelectric measurement, electrical resistivity and temperature, streams the data to a cloud platform, and runs machine-learning models to estimate in-place strength. ENR published the first substantive U.S. trade examination of its commercial rollout on August 14.
What happened on the Beyond tower
On the roughly 1,050-ft Beyond tower in Givatayim, the sensors said the concrete hadn’t reached required strength. The lab cylinders said proceed. The team followed the lab.
Shalom Buchnik, the project manager for contractor Tidhar, told ENR the resulting problem cost more than $500,000 in repair and replacement. On a later critical element the team went the other way and held post-tensioning until 11 days after placement. Subsequent core testing supported the sensor reading.
That’s the part worth sitting with. The system is being marketed on schedule savings, and Buchnik does say it saved 98 working days on Beyond in the end. But its value showed up when it contradicted the lab and got overruled.
The calibration curve is the real claim
Maturity testing under ASTM C1074 has existed for decades and almost nobody scales it, because every mix design needs its own calibration curve developed in advance. On a job with a dozen mixes, that’s a dozen separate lab programs before you pour anything.
GreenVibe’s claim is that it reads physical change in the material directly and doesn’t need the mix-specific curve. If that survives scrutiny, the constraint stops being technical and turns institutional, which is presumably why the company says it sits on ASTM Committee C09 and Subcommittee C09.64, the group that owns C1074 and is revising it now.
What it can’t do yet
Roee Reshef, GreenVibe’s CEO and co-founder, stated the limits on the record, and they’re worth repeating because they cut against his own sales case. The system is not approved to replace required concrete acceptance testing in the United States. It does not replace 28-day testing. Each sensor reads roughly an 8-inch radius, so a defect away from a sensor goes undetected.
The honest read for a U.S. general contractor today is that this is a supplemental QA layer whose main function is to tell you to slow down. That’s a harder thing to sell than schedule compression, and it’s probably the more valuable one.
Every operational figure above comes from the company or from a client relayed through a single article: the $500,000, the 98 days, the 11 paying projects, the roughly 10,000 sensors produced and the $12 million raised. Treat them as claims, not audited results.
See also our coverage of Mayo Clinic’s Rochester redevelopment, where concrete sequencing runs across nine active sites.