Theatre Design & Technology - Winter 2005 - 21

Obviously, the selection of an appropriate hose was critical to the success of the project. It had to meet the following
criteria: the ability to handle high pressure (between 400 and
600 psi); withstand abrasion and the roller action along its
walls; bend tightly without kinking and be adaptable to standard hydraulic fittings. After several missteps with hoses that
only seemed suitable, we settled on a 12˝ diameter, double
jacket fire hose, model DP15-600-50-ARI from Key Fire Hose
Corporation. Manufactured to stringent NFPA codes, the hose
met the following specifications:
*
*
*
*
*
*
*
*

Test pressure: 600 psi.
Service pressure: 400 psi.
Elongation: 6%
Twist per foot: 15° right
Warp: 10˝
Rise: 0
Burst test: 1000 psi.
Kink test: 500 psi.

For this study, we assumed pressure and rise were the
most important. Pressure is obvious, but rise is the tendency
for a hose to bend under pressure, which would present problems when running a linear system. Elongation, it turned out,
would have a major impact on the system as well. Note: this is
not the type of fire hose one typically finds in a hose cabinet in
public buildings (usually rated for only 125 psi).
Having acquired a suitable hose, the next step in the system design was the selection of the hydraulic power unit and
ancillary control equipment. In order to properly size the
power unit, we needed to determine the desired speed for the
system. A typical stage winch operates at 2 to 3 feet/second,
which is walking speed. This seemed like a good starting point
for our system. The formula for determining actuator velocity
in a hydraulic system is:
V = .3208Q ÷ A
Where V = velocity (feet per second)
Q = flow rate (gallons per minute)
A = area of the piston (square inches)
The formula reveals two important details: velocity is directly proportional to flow rate and inversely proportional to
the area of the actuator. (This fact would come to haunt us
later on in the investigation). Knowing the velocity desired and
the area of our fire hose, we solved for Q.
Q = VA ÷ .3208
Q = 2(π(0.75)2) ÷ .3208
Q = 2(1.76) ÷ .3208
Q = 3.53 ÷ .3208
Q = 11 GPM
Next we needed to determine the size of the motor and pump.
The formula for hydraulic power provide by the pump is:
W = QP ÷ 1714
Where W = power (hp)
Q = flow rate (gpm)
P = pressure (psi)
20

W I N T E R

2 0 0 5

TD & T

Given the capacity of the hose, pressure was limited to 600 psi.
Therefore:
W = 11 (600) ÷ 1714
W = 3.8 hp
I approached an engineer at Parker Hannifin with my calculations and, after some initial skepticism about the project
("You want to do what with a fire hose?!"), he became a quite
enthusiastic participant. He verified my calculations, but, given the
experimental nature of the project, suggested increasing the flow
rate to 15 gpm and doubling the horsepower. It was my turn to be
skeptical, but after a lengthy discussion of the potential pitfalls of
undersizing the system, I was convinced. He wrote the specification for the power unit, which included a Parker PVP33 pressure
compensated, variable volume piston pump; a 10 hp motor and a
35 gallon reservoir. This was a very sophisticated power unit.
Pressure compensation meant that when the system reached
its maximum set pressure, the pump would automatically stop
the flow, resulting in high efficiency and much quieter running. In addition, flow could be remotely enabled or inhibited
electrically, which would make control easier. Unfortunately,
being somewhat of a neophyte in the field of hydraulics, and
the fact that the power unit was delivered without any sort of
manual, that very sophistication caused us a great deal of difficulty when the system did not initially perform as expected.
For directional control I ordered both a manual lever-operated valve, Parker Model D1VL9CN and a high-end, electronic proportional valve, Parker Model D1FXE01HCNDJ0030.
The manual valve was intended for initial set up and trials and
the proportional valve for when control was handled by a
computer. The schematic of the system (fig. 2) shows the proportional valve. There is no schematic diagram for the hose
and pinch roller, so I just made one up.
In addition to the hydraulic power unit, I also built the
equivalent of an electrically controlled, 4-way directional valve
from sprinkler valves and PVC pipe for use when powering the
system with domestic water supply (fig. 3). This valve and lowpressure water supply represented our fall-back position in
case the pressures generated by the hydraulic power unit
proved too much for the hose.
With the power unit and hose on hand, we turned our attention to the design of the pinch rollers. For the rollers we selected
polyurethane, cut-to-width wheel stock from McMaster-Carr. The
rollers come in 20˝ lengths, and diameters ranging from 1˝ to 3˝
with a variety of axle sizes.
Our first attempt was intended only to test the basic concept and so had no scenery attachment points. The hose lay
over two 3˝ diameter rollers in a clumsy aluminium frame and
a smaller, 12˝ diameter roller with adjustable pressure bolts
squeezed the hose against the larger rollers (fig. 4). With high
expectations we assembled the rollers on the hose, tightened them
down, turned on the power unit and promptly bent the smaller
shaft without any perceptible movement from the rollers. "Hmm,"
I thought, "this is not a failure. It's just more data."
In our second attempt, we replaced the smaller roller
with one the same size as the other two, increased the axle size
to w˝ and used a hardened steel shaft. We also simplified the

Table of Contents for the Digital Edition of Theatre Design & Technology - Winter 2005