How many sensors should I use in a temperature mapping study?

Well, although there is no predicate rule establishing the amount of measuring sensors to be used when mapping a volume, there is guidance to help you determine sensor amount and placement. Guidance is available from The International Society for Pharmaceutical Engineering (ISPE), the World Health Organization (WHO) and the United States Pharmacopeia (USP), among others, as listed at the end of this article. This article summarizes them for you.

Commercial-off-the-shelf freezers, refrigerators, incubators and environmental chambers

Recommended sensor placement for internal volumes up to 2 cubic meters

Figure 1 - Recommended sensor placement for internal volumes up to 2 cubic meters

Use 10 temperature sensors in chambers with internal volume of up to 2 cubic meters. Place 9 sensors within the usable volume. Figure 1 suggests sensor placement. Place an additional sensor near the probe of the device that monitors and/or controls the system.

Why is this? You are mapping worst-case locations under the assumption that any other location contained within the usable volume will not experience temperatures exceeding these extremes. The corners represent the intersection of the most planes (3) that are exposed to outside air temperature. Also, corners tend to have the least air circulation. Therefore, each corner represents a potential worst-case location. Conversely, the center of the usable volume is farthest from the outside air. It is also most likely to have good air circulation. Likewise, it represents a potential worst-case location. Together, these locations are most likely to experience temperature extremes. Include a rationale in the temperature mapping protocol to explain all assumptions.

Recommended sensor placement for internal volumes up to 20 cubic meters

Figure 2 - Recommended sensor placement for internal volumes up to 20 cubic meters

Use 16 temperature sensors in chambers with internal volume of up to 20 cubic meters. Place 15 sensors within the usable volume as recommended in Figure 2. The extended length of the sides in bigger chambers makes the center of each pane a point of interest since its temperature might differ significantly from that of the corners. This is due to its proximity to the center of the usable volume. Place an additional sensor near the probe of the device that monitors and/or controls the system. This will be needed to determine alarm and control offsets by correlating temperature differences between these locations and the worst-case locations identified in the study.

Warehouses, cold rooms, walk-in freezers and walk-in refrigerators

Distribute sensors in a grid format every 5 to 10 meters along the width and length of the area3. For very large facilities, sensors can be placed up to 20 or 30 meters apart. At each point on the grid, arrange additional sensors vertically forming layers. Use three layers for rooms with ceiling up to 3.6 meters. Consider adding more layers for rooms with ceiling above 3.6 meters. Place an additional sensor near the probe of the device that monitors and/or controls the HVAC. Map product storage areas only. Provide rationale if adding sensors in other areas of interest. Temperature mapping should consider other factors that may contribute to temperature variability such as the size of the space, location of HVAC equipment, sun-facing walls, low ceilings or roofs, geographic location of the area being mapped, airflow inside the storage location, temperature variability outside the storage location, workflow variation and movement of equipment (weekday vs. weekend), loading or storage patterns of product, equipment capabilities (e.g., defrost mode, cycle mode), and operating procedures.

Bibliography:

  • 1 - ISPE (2016). Good Practice Guide: Controlled Temperature Chamber Mapping and Monitoring.
  • 2 - USP (2018). USP41-NF36 <1079>. Good Storage and Distribution Practices for Drug Products.
  • 3 - World Health Organization (2015). Technical Supplement 8 to WHO Technical Report Series, No. 961, 2011. Temperature Mapping of Storage Areas.
Read 4236 times Last modified on Wednesday, 23 March 2022 07:51

comments (10)

  • Cesar Mirabal
    Cesar Mirabal Comment Link

    Hi Subal, thank you for the question. The WHO Technical Supplement referenced in the blog, and the source of the recommendation in question, does not define a threshold for what very large facility (e.g., distribution center, finished product warehouse at a multi-building manufacturing site, etc.) dimensions are. As guidance, it provides a framework you can use when designing your mapping study for temperature-controlled storage areas. Keep in mind that mapping design should have enough detail to support your study conclusions based on area layout, HVAC system design, and workflow variation, storage location design (racks, shelves), and other factors that outline the boundaries of the storage system. It is understood that spacing between sensors can expand as the area becomes larger as long as it is reasonably covered. You should always include in the mapping protocol the rationale you used and any risk-based approach followed. The number of loggers selected should be commensurate to the storage area design and intended mapping test objectives. I hope this helps.

    Thursday, 04 August 2022 10:01
  • Subal Saha
    Subal Saha Comment Link

    As per above statements for large area we can take 5-10 meters apart a data logger and for very large area we can take 20-30 meters apart a data loggers. But the my confusion is which volume we can consider as large area and very large area? Is there any approximate scale. As above mentioned that up to 2 cub.meter we can set 9 data loggers and up to 20 cub.meter it will be 15. So is there any range for large area and very large area for consider volume size? Please advice!

    Wednesday, 20 July 2022 07:56
  • Cesar Mirabal
    Cesar Mirabal Comment Link

    Hi Astrie, yes, you can use the recommendations made in the article when defining placement of EDLMs and study duration. They are in alignment with WHO. Moreover, regarding duration, WHO’s Technical Supplement 8 listed in the bibliography establishes that “There is no formal time limit for a mapping study. Typically, for warehouses and other ambient storage areas, it should be run for a minimum of seven consecutive days – including five working days and two weekend days.”.

    Wednesday, 08 June 2022 13:58
  • Astrie Ivo
    Astrie Ivo Comment Link

    HI, I'm Astrie Ivo, thank you for your sharing and I want to ask relate temperature Mapping for Ambient Room. How to define the point (placement of EDLMs) of study Mapping? Is it same with the controlled Room? Mapping study with 7 days continously? Thank you so much!

    Friday, 20 May 2022 09:12
  • Cesar Mirabal
    Cesar Mirabal Comment Link

    Hi Paul, there are no regulations establishing an actual number. There are two good guidelines I suggest you use to make that determination based on your room design, support systems, operational workflow and other factors: 1. ISPE (2016). Good Practice Guide: Controlled Temperature Chamber Mapping and Monitoring. 2. World Health Organization (2015). Technical Supplement 8 to WHO Technical Report Series, No. 961, 2011. Temperature Mapping of Storage Areas. You should also check chapter of the latest USP version (2022) for updated information (if serving the USA market). The recommendations in this article are based on them. I recommend using data loggers that measure both temperature and relative humidity and following the recommendations in the blog for their amount and placement. Should you decide to use a different amount of loggers for relative humidity measurement than those for temperature, you should include a rationale in your test protocol explaining your choices considering HVAC process psychrometrics and room characteristics.

    Wednesday, 30 March 2022 07:21
  • Paul Wheeleer
    Paul Wheeleer Comment Link

    I was wondering if there is a regulation regarding how many EVM sensors are required for a room. We map with 55 each data sensors but only have one EVM probe monitoring this room. Thank you for any information you might have, Paul

    Monday, 21 March 2022 15:44
  • kay
    kay Comment Link

    Good! Learn more,thanks!But i still have some question how to choose a machine meet my request.Here is a machine Temperature and Humidity Chamber,i want to know more . For the conditioning of samples prior to testing, also applied for Water Vapour Permeability Testing when equipped with related kits.

    Monday, 19 July 2021 08:42
  • Gary
    Gary Comment Link

    What is the reference source for the following statement: Distribute sensors in vertical arrays with a minimum amount equal to the square root of the room area in square meters plus 1 for each layer. For very large facilities, sensors can be placed up to 30 meters apart. Use three layers for rooms with ceiling up to 3.6 meters. Consider adding more layers for rooms with ceiling above 3.6 meters

    Tuesday, 30 June 2020 12:31
  • Cesar Mirabal
    Cesar Mirabal Comment Link

    Amer, Thank you for your question. There is no cookie-cutter solution for these intermediate volumes. Here is my line of thought and recommendations: The use of the square root of the room area in square meters plus 1 per layer as an estimate for the minimum number of mapping sensors to use is a widely adopted industry interpretation of WHO’s guidance for large temperature-controlled storage areas that hold pharmaceutical products. For these large areas, WHO recommends installing sensors every 5 to 10 meters, hence the formula, in at least three layers along the width and length of the area. For very large facilities, this can be up to 20 or 30 meters. This is intended to ensure that the volume is mapped symmetrically in all three dimensions while reasonably covering the volume under study. This method of estimation is intended for large volumes. For high ceilings, the relatively small volume of the 34 and 70 cubic-meter stability chambers might result in a calculation yielding less sensors than those recommended by ISPE for a 20 cubic meter volume. If this is the case, you might consider using 16 sensors following the same ISPE rationale as for a 20 cubic meter volume. For low ceilings, the calculation might yield an adequate number of sensors provided that the area layout and intended product location still support symmetrical mapping in all dimensions. Applying the calculation to the 90 cubic-meter stability chamber will most likely yield an adequate number of sensors to symmetrically map the volume. Keep in mind that you can use more or less sensors depending on the chamber height, area layout, intended product location and other factors discussed in the article. You should consider mapping only the intended product locations based on the product rack configuration and layout. There is no need to map corridors if product will never be stored in them. If in doubt, err on the safe side. Finally, it is important to always include a rationale in the mapping protocol explaining the selected number of sensors. This will help auditors understand your decision. I hope this helps. Cesar

    Thursday, 19 March 2020 20:04
  • Amer
    Amer Comment Link

    Dear Sir : Kindly according to this article ( Use 16 temperature sensors in chambers with internal volume of up to 20 cubic meters. ) , what about stability chambers of volume above 20 m3 (34, 70 and 90 m3) ?how many sensors shall we use for these volumes? can I use the square root of the room area in square meters plus 1 for each layer as mentioned above ? Kindly advise .

    Wednesday, 26 February 2020 17:13

Leave a comment

Make sure you enter all the required information, indicated by an asterisk (*). HTML code is not allowed.