Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

Shot Flow Control

Download as pdf or txt
Download as pdf or txt
You are on page 1of 13

1991020

Shot Flow Rate Monitors and Controls


(If you can't monitor it, you can't control it)
Jack Champaigne
President, Electronics Incorporated, 56790 Macnetic D r i v e
Mishawaka, IN 46545

ABSTRACT

The shot peening process relies on multiple impacts of spherical media to a surface to
achieve residual compressive surface stress. The flow rate of the media (Ibs./min.)
times the (effective) exposure time (min.) determines the number of impacts in a given
surface area. To achieve consistent peening performance it is therefore necessary to
continuously monitor the media delivery system. This can be achieved with inductive
sensors for ferrous media or capacitive sensors for non-ferrous media.

Beyond providing a consistent process, these devices will also promote process optimi-
zation whereby desired intensity and coverage can be maintained and productivity can
be enhanced.

The ability to accurately determine media flow rates then leads to the ability to accu-
rately control the flow rate. Screw (auger) conveyors, miniature belt conveyors (Pre-
cifeed), pneumatically controlled grit valves, and magnetic valves can be employed.

KEYWORDS

Shot flow rate, shot flow controls, shot flow monitors, valves, MagnaValves.

Introduction

Shot peening is performed by impacting a surface with hard round balls, causing plastic
flow and permanent deformation. The result is a residual layer of compressive stress
which prolongs cyclic flexing component life and reduces cyclic fatigue and stress
corrosion cracking. Complete coverage or denting of the surface is required to assure a
uniform compressive layer.

The need for continuous flow rate measurement is further promoted with the latest
edition of the U.S. Military Specification MIL-S-13165 Revision C. Section 4.2.5 iest
Records requires that shot flow rate (and air pressure) be noted when processing the
Almen strips.

While targeting and coverage tools, such as Peen Scan ( I ) , are needed for set-up and
post treatment inspection, in-process monitoring is necessary to assure consistent
process control throughout the entire peening exposure time.

High flow rates (for air peening machines) can decrease shot velocity and thereby
reduce peening intensity. Los flow rates can result in incomplete coverage and in some
cases promote surface damage due to the higher velocities available with lean flow
rates. See Fig. 1 and Table 1 for intensity and flow rate correlation. Variations in flow
rate during the peening exposure time can be detrimental and, therefore, it is helpful to
continuously monitor the shot flow rate.

Fig. 1. Effect of media feed rate on Almen intensity

Table 1

A t Moxirnum Intensity A t Maximum Flow Rote


PS st nord. ] 10 1 Y1 / 40 1 50 1 60

Older methods of flow rate control usina a fixed orifice or adiustable media v,alves that
are confirmed with a catch test do not Govide the continuo& monitoring needed for the
wocess.

Peening machines may be divided into three classes, gravity drop, wheel type and air
type. The gravity drop process usually uses a bucket elevator to lift the shot to the
prescribed dropping height. Flow rate is easily monitored and controlled by regulating
the elevator speed. Due to its basic simplicity additional methods of flow control are
seldom used.

The centrifugal wheel, which has evolved from the blast cleaning industry, is used in
peen forming and some high production volume, usually automotive, applications, Air
type machines tend to be more versatile and better suited for lower production levels,
usually found in aerospace applications.
Shot Flow Rate Monitors

Wheel ~ e e n i n amachines. Wheel peening applications historically have relied upon


wheel motor load current as a measure of flow rate. On some applications this may be
acceptable, but active flow rate sensing provides some advantages.

Flow rate measurements that rely on motor current are generally limited to fixed speed
wheels. This motor current, often referred to as "running-Amps", is one method of
displaying the amount of work done to throw the shot and it is a good indicator of the
amount of shot thrown. A typical ammeter may have a full scale range of 50 Amps
(typically 16). Different size (diameter) wheels, different wheel speeds and different size
or efficiency motors will affect the no-load and running-Amp values.

The ammeter readings are seldom converted to flow rate readings, but the implication of
flow rate is there. The peening process is usually tuned to optimum (or acceptable)
performance and the ammeter reading is noted for reference. As long as wheel speed
remains constant and the wheel blades are in good shape the ammeter reading reflects
the flow rate. But, care must be taken in using motor Amps as a process control. In
comparing two or more wheels for performance the motor types must be identical.
Different motor brands may have different efficiencies which are reflected in motor
amperage. Blade wear and bearing conditions can change motor amperage. The
running-Amps, for a constant flow rate, will change with motor temperature, stabilizing
after several hours. Motor Amps also depend upon line voltage.

Wheels that are driven by belt and pulley have inherent losses that add to both no-load
and running-Amps. Wheels capable of variable speed will use more current at higher
speeds, even though flow rate is constant, since it takes more energy to throw the same
shot amount at a higher velocity. Conversely, less shot can be thrown at higher RPM's.
See Fia. 2. Therefore. for variable SDeed wheels both the motor speed (RPM) and flow
rate ( A ~ ~are
S required
) to define th'e process.

Fig. 2. Wheel speed vs. shot flow rate


Use of an inductive shot flow monitor avoids the above problems and provides inde-
pendent measurement of flow rate. Flow rate information may be displayed either as a
percentage of full scale capacity or in engineering units, such as pounds-per-minute.
Not all flow rate measuring devices provide a linear convenience is negligible compared
to process repeatability and higher degree of measurement resolution. Maintaining a
flow rate that produces 13 Amps on a meter ranged to read 50 Amps full scale with 10
Amps for no-load is a challenge, especially since the running-Amps may decrease from
13 to 12 Amps due to motor heating. Use of a flow monitor with 1% resolution can
greatly improve process control. Since it monitors flow through a sensor and ignores
motor Amps, it can provide a more direct measure of the process.

Another benefit is the possible use of flow rate alarms associated with the flow rate
monitor. These alarms are not influenced by motor current or its changes, due to
factors like motor temperature. See Fig. 3.

Fig. 3. Shot flow display and sensor

Air peenina Machines. Air type peening systems cannot use an ammeter for flow rate
measurement (since there is no wheel motor current to monitor). Direct flow rate
methods must therefore be used.

Most flow rate devices actually measure media density at a point after flow rate regula-
tion. If the media maintains a constant velocity, then flow rate can be easily displayed.
If the measuring point is after acceleration by air pressure, then variable velocities will
affect calibration and a look-up table may be required.

1. Optical
2. Acoustic
3. Capactive
4. Inductive

1. O ~ t i c a l : Historically, optical approaches have been unsatisfactory since deteriora-


tion of the system components (sender, receiver, optical path) quickly reduces system
sensitivity and accuracy.
2. Acoustic: Passive acoustic systems, similar to listening to flow rate with a
microphone, degrade with hose deterioration and are also influenced by shot quality.

3 Capacitive flow monitors have been applied in air peening machines


with limited success due to their narrow range of sensing. Most capacitive sensors are
density measuring devices and are influenced by media velocity. The most common
capacitance transducer, the unbalanced bridge method, may be operated in the
constant velocity section of the blast machine and is therefore not responsive to air
pressure or velocity variations. These sensors have an upper limit of approximately
30% solidslair ratio. They are well suited to glass bead and ceramic bead applications
provided that proper sensor placement is available.

For flow rate applications under 16%, the noise-type capacitive sensor can provide flow
rate measurement; but care must be taken to consider velocity effects.

A novel capacitive sensor from Tealgate, the


capacitive-noise type sensors, works only in high velocity applications.This type of
sensor does not provide an output signal for static conditions, such as a plugged
sensor. The sensor, therefore, is located in the blast hose and its output signal must
be corrected for velocity.

4. Inductive: For shot peening applications using steel shot, the inductive sensor can
be used. Both single coil and multiple coil types are in use. The multi-coil transformer
type, first used in the mining of iron ore, is described by the U.S. patent in the 1930's. A
primary coil circles the flow path and it generates a magnetic field. A second coil, simi-
lar to the first, is place over the flow path near the first coil. With no shot flow in the path
the magnetic coupling between the coils is very poor and only a very small amplitude
signal is derived from the secondary coil. When shot flow rate is established an "iron
core" is introduced in the centers of the two coils and the magnetic coupling between
the two coils is increased. The signal "transformed" from the first coil to the second coil
is likewise increased and this can be displayed on a meter. While the technology is
simple to apply, it suffers from poor signal-to-noise ratio and it does not maintain long
term stability.

An improvement on this scheme (2) uses a stacked column of shot whose height is a
function of flow rate with an outlet flow rate equal to the inlet flow rate. A resultant
measurement of the shot column, using the change in the coil inductance and displayed
as a change in circuit current, is translated into flow rate information.

Another approach, the inductive oscillator, involves measurement of either the oscillator
amplitude or frequency. Since inductive proximity switches commonly use amplitude
measurement, this was the earlier approach. This is simple to implement; but various
shot sizes result in different responses and recalibration is occasionally required.

Another development, a single coil oscillator inductive shot flow sensor, is based on
frequency measurements. Sensors using this technology are available in sizes for both
air and wheel peening applications. Since the sensor is actually a steel density mea-
surement, shot flow rate is measured in a constant velocity zone. For wheel machines
this is between the media flow rate regulating device and the wheel inlet. For direct
pressure air peening systems, the sensor is located below the media regulator and
above the mixing chamber. The sensor can be used in venturi or suction type peening
machines if the velocity (pressure) is kept constant. If various pressures are used, then
a look-up table must be used in order to maintain calibration.

Since the magnetic properties of cast steel shot are different from wrought (cut wire)
shot, a recalibration is necessary when changing media. Some recalibration may be
required for different sizes of the same media, depending on packing densities.

Shot Flow Control Mechanisms

Shot flow rate controls can be divided into two classes, restriction devices and convey-
ance devices.

I. Restriction Devices
1. fixed orifice plates
2. mechanical valves
3. magnetic field modulated valves

II. Conveyor Devices


1. auger screw
2. micro conveyor

I. Restriction Devices

1. Fixed orifice. The fixed orifice is the simplest device in the first class. the orifice plate
is often constructed with an extension wherein a hole of the same size is located as an
identification aid. See Fig. 4. Fixed orifice plates provide constant flow rates provided
that:

a. the media flow properties are constant


b. pressure above the orifice is greater than below the orifice

'::pi
C. the air pressure differential remains constant
d. the orifice size remains constant
e. no air leaks are introduced.

,@;(
anLE SIZES riRE

:;
Rate '0

lwmin

0
20 +o

L l r Prcrrure

Fig. 4. Shot flow r a t e v s . air


pressure i n direct pressure
LO

- PSI
ba boo A:
B:
ORPICE
7-
REFERENCE HOLE FOR SIZE
IDENTIFICATION

s y s t e m w i t h f i x e d feed rate o r i f i c e . Fig. 4A. O r i f i c e place


2. Mechanical valves. Mechanical valves provide the ability to operate at various flow
rates without disassembly as required by the orifice plate. Mechanical valves used in
wheel peening applications include dipper or clam shell valves and slide gate valves.
These are generally operated by air cylinder for onloff action. The open position is
often limited by a positive stop which is readjusted manually. See Fig. 5. In some
applications electric drive motors are used to position the positive stoo and this allows
for remote setting.

-- nTPPER VALVE

Fig. 5. Dipper valves

Mechanical valves used in air peening are generally applied to direct pressure systems
and are either air or electric actuated. The air actuated devices usually employ a large
diaphragm to overcome the restraining force of a spring. See Fig. 6 . In some cases the
control air pressure is remotely regulated to provide variable flow rates instead of onloff
action. Another valve concept, often called a pinch tube, has no moving parts except
the flexible pinch tube. See Fig. 7. by applying a higher pressure to the outside of this
tube it collapses and restricts the shot. The tube will eventually erode due to the
capture of shot within the tube. Tube life is further decreased by excessively high air
pressure.
STORAGE
- -1 EOPP"

AIR
D m
VALVE

BOOST
HOSE

121Adjustable Brass
pinch Rollers
\ \ Metering Tube

Manual Sure-Flo D.

Fig. 6. Mechanical media valves


Fig. 7. Pinch tube shut-off valve

Venturi type peening systems generally use an adjustment for regulating the aspiration
air inlet to modulate the suction and hence the amount of media entrained into the flow
stream. See. Fig. 8. This is similar to a carburetor where the choke butterfly plate is
used to modulate the vacuum. One of the problems with this arrangement is efficiency.
The flow rate is determined by the amount of vacuum available to suck shot into the
blast hose. To increase the suction the air aspiration inlet is reduced. This allows more
shot to be entrained in the hose but with less air volume needed for conveyance. At
some point there is too little air volume at a sufficient velocity to convey the shot and the
blast hose either surges or, in extreme cases, shot flow may cease, causing blocking of
the hose. Typical flow rates for induction type nozzles are 5-9 lbs./min. for 318" nozzle
and 8-15 Ibs.1 min. for 112" nozzles.

STORAGE

--I---
NOZZLE (I+&

Fig. 8. Suction fed

Another limitation to venturi type machines is the energy required to convey the shot
from the shot hopper outlet to the nozzle. If the nozzle is located above the media
carburetor, then energy is expended to lift the shot. This energy loss can be greatly
reduced by raising the media carburetor above the nozzle level. Use of a bucket type
elevator for media transfer may be necessary to accomplish this. See Fig. 9.

HOPPER

FLOW REGUTORS

Fig. 9. Valve placement

If a pneumatic media recovery system is used care should be taken that the pressure
differential above and below the shot supply does not interfere with media discharge
from the carburetor. A large (compared to nozzle) vacuum from the dust collector may
not be noticeable with a large head of shot; but if the shot supply gets low the shot may
be held up in the hopper discharge. The aspiration air will travel up through the shot
towards the dust collector and reduce or block shot flow rate. See Fig. 10. Choking the
carburetor will increase the venturi nozzle suction and reestablish shot flow, but the
correct action would be to reduce the dust collector vacuum to a lower level.

Fig. 10. Dust collector backwash


Another approach to venturi nozzle design is to place a mini-hopper, or funnel, directly
above and attached to the nozzle. This alleviates the problem of dust collector back-
wash and relies on the nozzle design to regulate the flow rate. Flow rate will be con-
stant as long as air pressure is constant and the internal nozzle parts are not eroded.
This technique is also referred to as gravity-fed nozzles.

3. Maanetic valves. Peening machines using steel shot can be regulated with a mag-
netic valve. These devices do not have any moving parts. The valve is normally closed
due to a permanent magnet. Applying power to the valve cancels the magnetic field
and shot falls through the valve. These valves were first used for wheel applications
and later adapted to air peening machines. See Fig. 11. In both wheel and air peening
applications the valves can be either open loop or, by including an inductive sensor,
they can operate in the closed loop serve mode.

Fig. 11. Cut-away view of wheel peening MagnaValve

For direct pressure peening the MagnaValve is placed below the media hopper and
above the mixing tee. Care must be taken that there are no leaks above the Magna-
Valve, since this backwash of air can levitate the shot. See Fig. 12

Fig. 12 MagnaValve for direct pressure peening


When the MagnaValve is used in suction peening machines it acts like a fuel injector
instead of a carburetor. See Fig. 13. This provides the advantage of allowing large
amounts of aspiration at all flow rates and injecting the proper flow rate of shot as re-
quired. The servo loop controller is used to monitor and control flow rates without
influence of air pressure changes.

Fig. 13.

II. Convevance Devices

1. Auaer feed. The auger screw feed, borrowed from the agriculture industry, is a
simple mechanism capable of precisely metering shot. Its primary advantage is tis
ability to convey any granular media. It is easy to use with suction type air peening
machines, but direct pressure systems require more elaborate construction.

2. Micro-convevor. The Preci-feed precision feeder from Tealgate is a miniature con-


veyor with a depressed belt housed in a metal enclosure. It is suitable of all granular
media and can be used with direct pressure or suction peening machines. It has a
demonstrated accuracy and repeatability of 2% and is in use for glass bead peening
and plastic media blasting for paint stripping. Plastic media, which is very low density,
is notoriously difficult to control. The Preci-feed conveyor is especially suitable for this
application.

Combined Flow Rate and Velocitv Measurement

Wheel peening machines inherently provide shot velocity information, since the wheel
speed (RPM) is known. Air peening machines provide another challenge. We expect to
see nozzle shot velocity measuring devices incorporated into flow monitoring systems
within the next two years. Two systems are presently under development. The first is
useful for only steel shot and experimental data indicates a high degree of accuracy
compared to a laboratory double rotating disc.

The second device, using a different technique, provides both nozzle shot velocity and
mass flow rate information for any granular media. Laboratory models of this device
exhibit 2% velocity accuracy for glass beads. This technique is likely to see its first
application in plastic media blast cleaning for stripping paint from aircraft. Later
applications will be toward glass and ceramic bead shot peening.

Conclusion

If you can't monitor it, you can't control it. The direct and continuous measurement of
shot flow rate is necessary to assure homogeneous processing. A variety of tools is
available to accom~lishthe task.

References

(1) FELD, PAUL G., "Method of Inspecting Shot Peened Surfaces for Extent of
Coverage", U S . Patent 3,950,642.
(2) BIEHLE, WILLIAM C., "Apparatus for Measuring Flow Rate of Electromagnetic
Granular Media", U S . Patent 4,552,017.

You might also like