The paper has been evaluated by three reviewers. All referees rate the importance and quality of the paper as good, with the exception of referee #3 who considers the scientific quality to be only fair. Referees #1 and #3 recommend publication subject to minor revisions, whereas referee #2 recommends to reject the paper because of overlap with technical documentation from the manufacturer. In their reports, Referee #1 and #3 have raised a number of scientific and technical questions about the instrument, which have been adequately answered in the author's reply. Corresponding information and improvements have been added by the authors to the manuscript. After revision, the manuscript contains considerably more scientific information on the functioning and quality of the ozone generator described than can be found in the technical documentation available on the manufacturer's website. Therefore, I agree with referees #1 and #3 that the paper after revision merits publication in AMT.
I have a few minor points that should be changed before the paper can be published.
1. Lines 206 - 208: Cantrell et al. (1997) have investigated the absorption cross section of water vapor, but not of O2. Besides Creasey et al. (2000), other appropriate references are Hofzumahaus et al. (1997) and Lantzendorf et al. (1997) as pointed out by referee #3.
2. Line 153: Carulite is a registered trademark for the catalyst. The material (a combination of copper and manganese oxides) should be explicitly mentioned in the paper (as in the reply to referee #3).
3. The ozone generator can provide accurate ozone mixing ratios at reduced atmospheric pressure as is shown by tests at different altitudes (Figure 5). Can the instrument also provide calibration gas at higher pressure (> 1atm), which could be useful in laboratory applications, and what would be the maximum useful outlet pressure?