Siphonic Roof Drainage Systems The Road To Priming

Siphonic roof drainage systems: the road to priming
Marc Buitenhuis

Hydraulic research engineer Akatherm International BV, Panningen, The Netherlands

16-01-2009

Abstract

In this article the start up of a siphonic roof drainage system is described. It is intended to give better insight in the
behavior of the system and help design systems that will work optimally and siphonic more often.
The development of the hydraulic jump and the elimination of air from the system by the suction power in the down
pipe are key items in the development of siphonic functioning.
To reach siphonic drainage as quickly as possible a longer vertical tail pipe as well as a quick deceleration in the
horizontal pipe is preferable.

1. Introduction

Siphonic roof drainage systems are designed to operate
with full bore flow. The computations made for
siphonic roof drainage systems therefore focus on
single phase water flow. For the proper design of the
maximum capacity of a siphonic system this is
sufficient.
However it is necessary to have a good insight in the
start up phase to siphonic functioning of the system as
well. Most of the times the rain intensity will not be
high enough to get the system to work siphonic. It is
preferable to have the system work siphonic once or
twice a year at least to self clean the system of debris.
Also when the system works siphonic the noise
production is lower than with two-phase flow
(combined water/air flow).
In this article the start up phase will be described to
have a better understanding of the phenomena and
develop measures to stimulate full bore flow.

2. Start up phase

When the rain starts the roof drainage system slowly
starts up. At first the water flows into the roof outlet at
a low rate and shallow water level. The water flows
along the vertical walls of the tail pipe creating an
annular flow. At the bend to the horizontal tail pipe or
collector pipe the flow collects at the bottom of the
bend resulting in a separated flow in the horizontal
pipe. In the down pipe behind the collector pipe the
water is forming an annular flow again.

Illustration 1. Flow regimes in vertical pipes: annular, slug
and bubble flow

Illustration 2. Flow regimes in horizontal pipes: straight, wavy
and slug flow

This is independent of the water velocity streaming into
this pipe. The flow can follow the inner contour of the
bend or splash onto the opposite wall. The point at
which the annular flow is reinstalled will differ, as will
the pressure distribution when the pipe is closed off by
the splashing water.
At some higher flow rates the separated flow in the
horizontal pipes will become wavy.

3. Hydraulic jump

When the water streams down the vertical tail pipe it is
accelerated by gravity. When it flows into the
horizontal pipe the flow is decelerated forming a
hydraulic jump.
The principle can be compared to the stream of
vehicles on highways or race tracks. Vehicles can
accelerate optimally on roads that are straight and
keep on being straight for miles. As soon as there is a
curve in the road the vehicles have to slow down.
When the first vehicle decelerates the one behind him
has to decelerate also and the distance between the
vehicles is decreasing. This is very often the moment for
accidents to happen: there is an increasing chance for
collision. Exactly this is the case for fluid particles in a
stream. When particles are redirected from the vertical
downfall to horizontal flow the fluid is decelerated. As
fluid particles have no brakes they will collide and the
only way they can go is up, creating height and thus a
hydraulic jump.

Illustration 3. Hydraulic jump

The above explains 2 things: first of all why an
increasing length of vertical tail pipe leads to earlier
priming, second why an increasing resistance in the
collector pipe leads to this same result.
An increasing length of tail pipe leads to more length
to accelerate the fluid coming from the roof, thus to
higher velocities in the bend to the horizontal pipe. This
will lead to a higher hydraulic jump when the flow is
decelerated in the horizontal pipe.
Also the more the flow is decelerated in the horizontal
pipe, thus the higher the resistance downstream of the
bend, the higher the hydraulic jump will be.
Eventually the hydraulic jump will close off the whole
pipe diameter, leading to below atmospheric pressures
in the system behind the closure and priming will start.

4. Start of priming

When the hydraulic jump closes off the whole
periphery of the pipe the air behind the jump has only
one way to leave the system and that is through the
down pipe. To transport the air through the down pipe
the friction forces between the water and the air have
to overcome the buoyancy forces of the air. In other
words the water has to drag the air along against its
tendency to rise. To make this happen the flow rate
has to increase further.

5. Measures to enhance priming

As stated above to enhance the priming of the system
a longer tail pipe can be chosen or the deceleration of
the flow in the horizontal pipe can be increased in
order to make the hydraulic jump close off the
periphery of the pipe as quickly as possible leading to
earlier priming.

6. Conclusions

In this article the start up of a siphonic roof drainage
system has been described. It is intended to give better
insight in the behavior of the system and help design
systems that will work optimally and siphonic more
often.
The development of the hydraulic jump and the
elimination of air from the system by the suction power
in the down pipe are key items in the development of
siphonic functioning.
To reach siphonic drainage as quickly as possible a
longer vertical tail pipe as well as a quick deceleration
in the horizontal pipe is preferable.

7. References

1. Fox, Robert W., McDonald, Alan T. Introduction to fluid mechanics, third edition, 1985, School of Mechanical
Engineering Purdue University, John Wiley & Sons
2. Scott Arthur, John A. Swaffield, Siphonic roof drainage: current understanding, 2001, Water research group,
Department of civil and offshore engineering, Heriot-Watt University, Edinburg, Scotland (UK)
3. G.B. Wright, S. Arthur, J.A. Swaffield, Numerical simulation of the dynamic operation of multi-outlet siphonic
roof drainage systems, 2005, Drainage and water supply research group, School of the build environment,
Heriot-Watt University, Edinburg, Scotland (UK)
4. Scott Arthur, The priming focused design of siphonic roof drainage, drainage research group, School of the
built environment, Heriot-Watt University, Edinburg, Scotland (UK)
5. S. Arthur, G.B. Wright, Siphonic roof drainage systems priming focused design, 2006, School of the built
environment, Heriot-Watt University, Edinburg, Scotland (UK)
6. S. Arthur, J.A. Swaffield, Siphonic roof drainage system analysis utilizing unsteady flow theory, 2000,
Department of building engineering and surveying, Heriot-Watt University, Edinburg, Scotland (UK)

Siphonic Roof Drainage Systems The Road To Priming

Uploaded by

Copyright:

Available Formats

Siphonic Roof Drainage Systems The Road To Priming

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Siphonic Roof Drainage Systems The Road To Priming

Uploaded by

Copyright:

Available Formats

Siphonic roof drainage systems: the road to priming

You might also like