Henry Febo from FM Global has written a good review for the AIDIC (Italian Association
of Chemical Engineering) which provides an excellent review of explosions associated with powder drying processes1, but there are a few issues associated with large scale spray driers in the dairy industry, that are not drawn out in his paper.
This short article aims to keep lessons learnt many years ago as current knowledge for plant operators.
Challenge #1 – Scale of the Operation
The first particular feature to note is the scale of some drying processes in the dairy industry. Spray drier chambers of 1000 m3 or even more are common. These volumes are typically not found in the chemical industry and bring in their train some significant problems.
Long standing advice on dust explosions from HSE (booklet HSG 103 pub 2003 which came from my desk)2 advised that ‘open air siting of dust handling plant is strongly recommended where the scale of the operation is large’. That way, you don’t have to worry about the response of the building to explosion pressure rise.
However, biosecurity and hygiene considerations mean this option is generally not acceptable to manufacturers of powdered dairy products. Instead we find driers which occupy a large fraction of the building room in which they are located. If the venting system for some reason releases pressure into the building, the walls are vulnerable to failure or collapse, causing danger to those both inside the building, and nearby the building outside. The implications for building design and location are evident.
Challenge #2 – Equipment and Plant Design Strength
The next big problem arose when the plant designers were asked to assign a pressure rating to these big chambers. Simplistic stress analysis analysis shows that as vessels get larger, for a given wall thickness they become weaker. Moreover, on a large vessel, not designed to pressure vessel codes, you have to correctly identify the failure mode under overpressure before you can either assign a value, or propose a modification to strengthen the vessel.
Early answers back from the plant designers gave such low values of over pressure rating (Pred in the terminology of explosion venting) that they could not be protected by venting or suppression, and the costs for strengthening these large dryers were very substantial. As a regulator, I never dared ask to see the design calculations, for fear that my mechanical engineering colleagues would find them faulty!
Challenge #3 – Coupling of the Dryer Equipment
The third challenge arrises from the vibro separators that are often installed downstream of the main drying chamber. These vibrate to keep the newly formed powder from agglomerating as it cools, but that means you have to provide some sort of flexible coupling between the dryer and the vibro fluidiser. This is a potentially weak link in the system with various possible failure modes.
If the vibro fluidisers have to be protected from explosions by venting (perhaps with flameless venting devices) or explosion suppression, more flexible couplings are needed. Fabric types are vulnerable to failing in fire conditions or simply prolonged use.
Challenge #4 – Layer Ignition Properties
If these design problems were not enough, the plant operators in the dairy industry also need to understand the self-heating properties of their products. The onset temperature for run away self-heating leading to burning is strongly dependent on the thickness of a layer of product; it is not a single number. Moreover the layer thickness is not easy to measure inside a drier while operating. The temptation to delay the cleaning cycle is very real. With some types of dairy products air inlet temperatures are close to the critical temperature for a modest layer thickness, and layers tend to collect at the top of a drier close to the air inlet.
My first experience of an explosion in a diary industry drier came at a premises where the operators had not appreciated the dangers of lifting the air inlet temperature just 5 degrees although they had been warned by the plant suppliers. These units provide a common example of the observation that where the fire starts, and where the explosion starts are not the same place.
Another explosion arose where dust accumulated in a hot part of a plant where it was not expected to be, because someone had left a valve open on the clean-in-place system. When the hot material was later dislodged, an explosion resulted. A HAZOP study may have identified this explosion scenario.
The conclusion is that spray dryers and other process equipment in the dairy industry are inherently difficult to design and operate safely.
1. Febo, Henry L. 2013. “Processes for drying powders – Hazards and Solutions“. Chemical Engineering Transactions, 31:709-714
2. Health and Safety Executive, 2003. “Safe Handling of Combustible Dusts: Precautions against explosions“. Website: http://www.hse.gov.uk/