Is anyone familiar with evacuation experiments and/or measurements of flow rates of people and/or population densities with revolving doors (see http://en.wikipedia.org/wiki/Revolving_door)? In Sweden these are becoming more and more popular (and the designs vary) and it is not entirely easy to figure out how they should be considered in the fire safety design, nor evacuation analysis.

Please post here if you have any data on the topic. Thanks!

Hi Karl,

I know that revolving doors are included in Fruin's studies based on observations in public spaces
[J.J. Fruin. Service Pedestrian Planning and Design. Maudep (1971)].

Fruin categorized the doorways according to the mechanism employed (instead of width) and then he calculated an equivalent pedestrian volume for the flow rates (expressed in occ/min). For revolving doors this is equivalent to 25-35 occ/min.

You can also find some interpretation of his data in this paper:
S.M.V. Gwynne, E.D. Kuligowski, J. Kratchman, J.A. Milke (2009) Questioning the linear relationship between doorway width and achievable flow rate. Fire Safety Journal 44:1, January 2009, Pages 80–87.

Hi Karl,

"CIBSE Guide D: Transportation Systems in buildings" also states flow-rates of 25-35 occ/min for an opening that was 1m wide (no further details of the study or what this is based on is mentioned).

Good luck with the research!

Michael

Interesting to note that there was a revolving door used at the main exit to the Cocoanut Grove Fire in 1942 which jammed due to the high number of people trying to use it:

"The club’s main entrance/exit was a revolving door where many people made their escape — until it was overwhelmed by the crush of people and jammed, trapping everyone behind it."

http://nfpa.org/cocoanutgrove

We did some flow experiments at our consultancy firm in 2010 (together with Boon Edam if I remember correctly). The experiments was on a revolving door with collapsable sides. We found that the way that the flow through the door was much higher than expected, and higher than we are currently using for our calculations. We have, however, not yet implemented these higher flow rates yet becuase we would like to see more research done on the subject. So if anyone has anything else on this it would interesting to see their results.

Cheers!
Axel