| Nexus Pty Ltd is a furniture manufacturing company located at Victor
Harbor approximately 80 Kms south of Adelaide. The company has been in operation for 25
years and employs 70 people. Nexus produces product for the major Austalian furniture
retailers with a small amount of product exported to New Zealand and Japan.
In December 1997 the Company engaged a local consulting organisation to develop a
finishing line to improve both productivity and the quality of finished product.
Unfortunately, the system failed to achieve these objectives and the capital investment
has not delivered a return. In order to get the system operable in some form our General
Manager worked in conjunction with technical staff from several lacquer companies and
during this process discovered a number of reasons why the system was incapable of doing
the job.
Process Outline
- Staining
The process of finishing the product involves the sprayed application
of a stain to enhance the natural features of the timber. The stain consists of various
dyes dissolved in a hydrocarbon solvent. After application the solvents evaporates leaving
the timber surface dry and coloured.
- Seal Coat
When the stain has dryed a lacquer formulated to make sanding easier
is spray applied to provide a 'build' to the surface. These materials are also dissolved
in solvents and after evaporation the surface is dry. After drying the sealer needs
additional time to cure (harden off) before being sanded.
- Sanding
This pocess involves sanding the surface with fine grade sand papers
to de-nib the raised timber fibres and key the surface for the application of the final
lacquer coat.
- Final Coat
The final lacquer coat which is formulated to provide a lustred
finish in various gloss levels is spray applied in the same way as the seal coat and again
drying is achieved by the evaporation of solvents followed by a time to cure.
Finishing Line
Prior to the implementation of the finishing line the product was 'manhandled' between
the various processes, and drying and curing occurred in the ambient atmosphere of the
factory.
With the advent of the line, product is transported between the processes using roller
conveyor which is manually controlled in some areas and power controlled in others.
Two heated tunnels (one after seal coat and the other after final coat) are used to
accelerate the drying and curing of the lacquered surface. The tunnels consist of an
enclosure into which heated air is pumped in at the bottom and exhausted through the roof
with some air being recycled and some exhausted to the atmosphere. The reason this is done
is to keep the amount of solvent in the air below flash point. The air is pumped by a fan
across a heat exchanger which is heated by a hot water boiler.
The work flow through the line is:
Raw product is
loaded onto the conveyor on chipboard pallets (2.0 metres x 0.8 metres).
The pallet is
then conveyed by hand to the stain booth where the stain is applied.
After staining the product is moved by hand in board increments through the drying area
which consists of a spray booth through which air is being drawn and exhausted to the
atmosphere.
The pallet
arrives at the seal coat booth where the sealer is applied before entering the first
tunnel.
After the seal
coat has been applied the furniture is moved to the flash off area before entering the
seal-coat tunnel.
The journey
through the tunnel is on a powered conveyer.
When the pallet
arrives at the end of the tunnel it is transported by hand through the sanding area. The
product is removed from the line for sanding and reloaded after sanding.
It is then
transported by hand to the final coat booth where lacquer is applied.
After lacquering the product is moved through the second tunnel by powered conveyor.
Problems
The major problems with the system are:
- too much manual handling
- too much rework
By far the major problem is the rework which is caused by a fundamental flaw in the
design of the system. Unfortunately, the line developers did not understand the correct
conditions required to dry and cure lacquers; their only information was 20 minutes at
25°C at 50% humidity in laboratory conditions. The reality is that the process of drying
involves the evaporation of various solvents which have differing rates of evaporation
under different climatic conditions. The varying factors which affect this process are:
- Temperature
- Ambient temperature affects:
- the spraying process if it is too hot, solvents vaporise before contacting the
product surface
- the amount of energy required to generate heatif the ambient temperature is too
low the tunnels cannot reach operating temperature
- the upper and lower limits at which the process will worktoo high and solvent
boils; too low and blooming occurs
- Generated temperature affects are influenced the method used to generate heat (hot water
or steam with heat exchanger or direct gas firing)
- Lacquer temperature
- too cold and the material won't pump
- too hot and the solvent vaporises pre-maturely
- Surface Temperature of the product
- too cold it assists blooming
- too hot the solvent vaporises pre-maturely
- Relative Humiditytoo high slows evaporation and causes blooming
- Air flow across the surfacelow air flow allows low evaporation rate
- Evaporation rate of solventsaccording to the lacquer suppliers the ideal
conditions are:
- ambient temperature of 20°C to 25°C for the application of stains and lacquers
- good air movement of dry air (20°C to 25°C) over the product's surface during flash
off (evaporation of fast solvents)
- for lacquers, ramp up temperature from flash off to 40°C over about a 7 minute period,
sustain temperature for another 7 minutes and reduce back to 25°C over another 7 minutes
while sustaining good air movement.
Outcome
Our concern is that this information is not specific enough. We would like a model
developed which:
- shows the relationship between the variables
- establishes the optimal conditions for drying and curing the materials
From the model we could then determine what modifications are required to the line.
References
Howison, S. D., Moriarty, J. A., Ockendon, J. R., Terrill, E. L. & Wilson, S. K.
1997, A mathematical model for drying paint layers , Journal of Engineering Mathematics,
vol. 32, no. 4, pp. 377-394. |