This manuscript presents an optimization of a previously published low-cost optoacoustic black carbon (BC) sensor designed for long-term BC measurements in highly polluted environments. The sensor is based on an ellipsoidal chamber that separates the quartz tuning fork transducer from the aerosol flow path. The optimization consists of a protective clean-air sheath flow that prevents BC from depositing on sensitive components while minimizing acoustic noise from the sheath flow. This results in a reduction in contamination by orders of magnitude, as demonstrated by an experimental study and statistical evaluation. The authors claim a significant reduction in maintenance requirements, enabling the deployment, monitoring, and regulation of BC emissions in harsh environments. 

In my opinion, the paper is well organized, fits the scope of the journal, and should be published, subject to minor revisions as detailed below:

Methodology:

• At which temperature(s) were the sensors operated?
• The tests were mainly conducted with „dry“ soot - Would humidity and/or volatiles affect contamination? 

Results section:

It would be interesting to see a performance comparison of the sensor with and without sheath flow regarding:

• *noise
• *LoD
• *rise time
• *linearity
• *Allan deviation

How is the baseline correction performed in more detail? Is a lock-in technique used? If yes, is the phase of the signal considered for the background subtraction?

Fig 3: the colors of the data points are hard to distinguish

Fig. 3a: What does the rising line represent?

Fig. 3b: Where are the spikes in the control and IDSS coming from? 

Discussion:

• How was the sensor calibrated (MAC vs. BC mass concentration), and what would a calibration scenario for field deployment look like?
• Would there be a possibility to track possible contamination (e.g. resonance frequency of the tuning fork) during field deployments?