# What is the Number of Samples Locations Readings for ISO14644?

> Learn how to calculate the correct number of sample locations under ISO 14644-1:2015 using room area tables and formulas to ensure accurate compliance.

- Author: Ashraf Almuhtaseb (Data Center Cleaning Specialist)
- Published: 2026-06-06
- Updated: 2026-10-24
- URL: https://idata.center/en/articles/number-samples-locations-readings-iso14644

## Key takeaways

- Under ISO 14644-1:2015, the minimum number of sample locations for rooms up to 1,000 m² is determined using Table A.1 of the standard.
- For rooms larger than 1,000 m², the minimum number of sample locations is calculated using the formula NL equals 27 multiplied by A divided by 1,000, rounded up.
- ISO 14644-1:2015 requires every sampling location to have a minimum sample volume of at least 2 liters of air and a sampling time of at least 1 minute.
- A room passes the ISO 14644 test when every individual sample location meets or falls below the class limit for each classified particle size.

## How Many Sample Locations Does ISO 14644-1 Require?

 

Under [ISO 14644-1:2015](https://www.iso.org/standard/53394.html), the number of sample locations comes from the floor area of the room. A 1,000 m² data hall needs at least 27 locations, and a room larger than 1,000 m² uses a simple formula.

 

Many reference tables online still show the older numbers. Those come from the 1999 edition, which used the square root of the area, so 1,000 m² gave 32 points. If your test report or tender still uses that table, it is quoting a withdrawn method.

 

## How do you calculate the number of sample locations?

 

For rooms up to 1,000 m², you do not calculate. You read the minimum number from Table A.1 of the standard. If your area falls between two rows, use the next larger area.

 

For rooms larger than 1,000 m², use this formula:

 

NL = 27 × (A ÷ 1,000)

 

NL is the minimum number of sample locations, rounded up to the next whole number. A is the area of the room in square meters.

 

Two quick examples:

 

- A 1,800 m² data hall: 27 × 1.8 = 48.6, so 49 locations.
- A 4,000 m² data hall: 27 × 4 = 108 locations.

 

## Suggested minimum sample locations by room area

 

| Room area up to (m²) | Minimum sample locations |
| --- | --- |
| 2 | 1 |
| 4 | 2 |
| 10 | 5 |
| 24 | 6 |
| 52 | 10 |
| 104 | 16 |
| 232 | 22 |
| 352 | 24 |
| 636 | 26 |
| 1,000 | 27 |
| Above 1,000 | 27 × (A ÷ 1,000) |

 

The table gives a suggested minimum sampling. You can always add locations, for example near a CRAH unit, a loading door or a cable cut-out where you expect more dust.

 

## Why are the new numbers lower than the old ones?

 

The 1999 edition calculated NL as the square root of A and then tested the results with a 95% upper confidence limit. The 2015 edition dropped both. It uses a statistical model based on a 95% confidence that at least 90% of the room does not exceed the class limit.

 

For a small room the two methods give similar numbers. For a mid-size room the difference is large: a 500 m² hall needed 23 points under the old rule and needs 26 under the new one, while a 1,000 m² hall dropped from 32 to 27.

 

## How long do you sample at each location?

 

ISO 14644-1:2015 sets two minimums for every location: at least 2 litres of air and at least 1 minute of sampling time.

 

The standard also gives a formula for the volume needed to count enough particles at the largest particle size you classify:

 

Vs = (20 ÷ Cn,m) × 1,000

 

Cn,m is the class limit for that size, in particles per cubic meter. For ISO Class 8 at 5 µm, the limit is 29,300 particles/m³, which gives about 0.7 litres. That is below 2 litres, so the 2 litre and 1 minute minimums are what apply in a normal data hall.

 

A counter that samples 2.83 L/min meets both minimums in one minute.

 

**Note: check the flow rate on your own counter's data sheet.** More on choosing the device is in [which particle counter to use for the sampling record](https://idata.center/en/articles/particle-counter-device-taking-sampling-record).

 

## How should the locations be spread across the room?

 

Split the room into sections of equal area, one per location, and place each point in the middle of its section at working height. In a data hall, working height is roughly the height of the server air intakes.

 

Keep these rules:

 

- Spread points over the whole room, not only the aisles you can reach easily.
- Avoid placing a point directly in front of a supply grille or perforated tile, where you would measure filtered supply air rather than room air.
- Mark every point on a floor plan, so the next test uses the same positions and the results can be compared.

 

The image below shows an example of marking sampling points on a simplified diagram of a data center.

 

![Sampling Layout](https://idata.center/uploads/2026/10/ce5cbd18fe7f684ce24d.webp)

 

## What counts as a pass?

 

Each location passes when its average concentration, at each size you classify, is at or below the class limit. The room passes when every location passes. There is no longer a room-wide average or confidence limit to calculate.

 

Data halls are usually tested at ISO Class 8, the level the [ASHRAE TC 9.9 particulate guidelines](https://www.ashrae.org/File%20Library/Technical%20Resources/Publication%20Errata%20and%20Updates/2011-Gaseous-and-Particulate-Guidelines.pdf) set for data centers. The class limits are explained in [ISO 14644 cleanroom classes](https://idata.center/en/articles/iso-14644-cleanroom-classes-555), and the timing of the test after cleaning is covered in [when to take the ISO 14644 reading](https://idata.center/en/articles/take-reading-iso-14644-cleaned-data).

 

Count your locations from the 2015 table, sample at least 2 litres for at least a minute at each one, and keep the same floor plan from test to test. That is what makes a certificate worth reading.

## Frequently asked questions

### How many sample locations does ISO 14644-1 require?

Under ISO 14644-1:2015, the number of sample locations is determined by the floor area of the room. Rooms up to 1,000 m² use Table A.1, while rooms larger than 1,000 m² use the formula NL = 27 × (A ÷ 1,000).

### How do you calculate sample locations for rooms larger than 1,000 m²?

You use the formula NL = 27 × (A ÷ 1,000), where NL is the minimum number of sample locations rounded up to the next whole number and A is the room area in square meters.

### How long do you need to sample at each location?

ISO 14644-1:2015 requires at least 2 liters of air and at least 1 minute of sampling time for every location.

### How should the sample locations be spread across the room?

Split the room into equal sections, placing each point in the middle of its section at working height. Spread points over the whole room and avoid placing them directly in front of supply grilles or perforated tiles.

### What determines if a room passes the test?

Each location passes when its average concentration at each size is at or below the class limit, and the room passes when every location passes. There is no longer a room-wide average or confidence limit to calculate.

---

Ashraf Almuhtaseb is a Data Center Technical Cleaning Expert with over 9 years of specialized field experience since 2017, and the only independent professional in the MENA region publishing dedicated technical content on data center cleaning, contamination control, and ISO 14644 compliance.

© 2026 Ashraf Almuhtaseb — https://idata.center/