5
negative-pressure impulse (in direction) is likely to occur as
the impeller passes from the Organic side to the Aqueous
side. This impulse is calculated to be around 0.141 psi and
would occur locally at these transitions around 4 times per
second or at the impeller vane pass frequency.
The R301 impeller provides a 1.56 psi pressure pulse to
the top of the floor plate as the flow migrates through the
radial vanes into the open flow of the mix box at the stated
flow and rotational speed.
The backward curved radial vanes of the R320D impel-
ler provide a much smoother flow transition and resulting
pressure characteristics are much more favorable for the
mix box structure. See Figure 5. Chamber wall momen-
tum- derived pressures calculate at 0.102 psig. Aqueous to
Organic impeller pass impulses calculate at 0.0465 psi and
flow migration pressures through the curved radial vanes
results in a 0.152 psi pulse to the false floor plates.
Determination of Fatigue Limit
A literature search3 for ASTM A240 316L SS plate gave
an ultimate tensile strength Sut of 70,000 psi for the base
material of the mix box. A suggested endurance limit in
the literature was given as 35% of Sut or an S’e of 24,500
psi. A purely analytical approach was also used to corrobo-
rate this number as a design basis from a power curve fit of
data from a strength/cycle diagram or S-N diagram and the
stated Sut of 70,000 psi.4 Results yielded 35.28 kpsi at one
million cycles,
29.1 kpsi at ten million cycles and 23.99 kpsi at one
hundred million cycles. It should be noted that these calcu-
lated values are from an idealized completely reversed axial
fatigue test of a round bar in simple tension. Because of
this, precise results are not obtained, but rather should be
used as a guide. The S’e of 24,500 psi appears to be cor-
roborated with the analytical approach and would at least
be conservative. This value was used in the analysis.
The fatigue strength is determined by applying a sur-
face factor, size factor, temperature factor and miscellaneous
effects factor (i.e., notch/defect sensitivities factor) to the
endurance limit4 which yields a maximum allowable stress
of 5,683 psi. With an additional applied factor of safety of
1.5, the target maximum stress under fatigue loading con-
ditions is suggested to be 3,800 psi.
For non-cyclically loaded areas, the allowable stress
determination was taken from the ASTM A240 316L
stainless steel value of Sy =25,000 psi. Taking 0.66 Sy,
yields 16,500 psi. This is the value utilized for maximum
allowable stress for pure static loads.
Plate Analysis
The static and dynamic loads were applied to the concep-
tual platework as distributed loads. Plate spans and rein-
forcement sections were modeled as both simply supported
and fixed-fixed structures depending on position and orien-
tation. Plate thicknesses and reinforcement section geom-
etries were repositioned or thickened to ensure plate stresses
were below the determined fatigue limit of 3,800 psi or
16,500 psi given the loading condition.
RESULTS AND DISCUSSION
Based on the analysis, the preliminary design presented
by CAID required some slight, key modifications for the
Metcalf E1B replacement.
The analysis identified that the reinforcing on the
existing and proposed design appeared insufficient while
utilizing the R301 impeller design. The suggested floor rein-
forcement redesign consisted of a composite arrangement
of the upper 3/8” plate with an 8-3/8” deep Tee with 8” x
3/8” flange and 1/4” web, continuously welded throughout
in two locations spanning the entire false floor. In addition,
a 6”x3/8” flat bar welded on edge to the underside of the
floor as stiffeners on approx. 33” centers (3 total) in the
opposite direction as the Tees will be required to stiffen the
floor plate sufficiently against the cyclic hydraulic load. If
the R320D impeller is used, the proposed CAID design
qualifies without modification. It is recommended that the
reinforcing be added in order to keep the flexibility of uti-
lizing the R301 or R320D impellers in the process.
It is also suggested that the orifice be kept at a mini-
mum of 48” diameter for the plate design considered. A
larger opening, if possible, would improve the pressure
design of the system. A larger opening would likely result
in a slight loss in efficiency of the pump mixer, but it is felt
that it would be “small.”5 A cone inlet is recommended to
stiffen the free edge of the orifice and to improve the head
losses through the orifice over the current flat-plate orifice
design proposed. Stiffening of the Divider Wall is recom-
mended. Based on the analysis, a suggested Divider Wall
reinforcement is suggested as two 4”x3/8” flat bars oriented
vertically on either side of the orifice cone as shown in the
sketch, continuously welded on the Organic side. The pro-
posed continuously welded angle sections running along
the perimeter and angled across the face of the Divider Wall
should remain. The Divider Wall should be welded con-
tinuously to the False Floor and box walls.
Stiffening of the lower chamber exterior walls was not
required over the proposed CAID design. It was recom-
mended though to lower the lowest horizontal stiffener by
two inches to obtain a maximum plate stress of 16,500 psi
negative-pressure impulse (in direction) is likely to occur as
the impeller passes from the Organic side to the Aqueous
side. This impulse is calculated to be around 0.141 psi and
would occur locally at these transitions around 4 times per
second or at the impeller vane pass frequency.
The R301 impeller provides a 1.56 psi pressure pulse to
the top of the floor plate as the flow migrates through the
radial vanes into the open flow of the mix box at the stated
flow and rotational speed.
The backward curved radial vanes of the R320D impel-
ler provide a much smoother flow transition and resulting
pressure characteristics are much more favorable for the
mix box structure. See Figure 5. Chamber wall momen-
tum- derived pressures calculate at 0.102 psig. Aqueous to
Organic impeller pass impulses calculate at 0.0465 psi and
flow migration pressures through the curved radial vanes
results in a 0.152 psi pulse to the false floor plates.
Determination of Fatigue Limit
A literature search3 for ASTM A240 316L SS plate gave
an ultimate tensile strength Sut of 70,000 psi for the base
material of the mix box. A suggested endurance limit in
the literature was given as 35% of Sut or an S’e of 24,500
psi. A purely analytical approach was also used to corrobo-
rate this number as a design basis from a power curve fit of
data from a strength/cycle diagram or S-N diagram and the
stated Sut of 70,000 psi.4 Results yielded 35.28 kpsi at one
million cycles,
29.1 kpsi at ten million cycles and 23.99 kpsi at one
hundred million cycles. It should be noted that these calcu-
lated values are from an idealized completely reversed axial
fatigue test of a round bar in simple tension. Because of
this, precise results are not obtained, but rather should be
used as a guide. The S’e of 24,500 psi appears to be cor-
roborated with the analytical approach and would at least
be conservative. This value was used in the analysis.
The fatigue strength is determined by applying a sur-
face factor, size factor, temperature factor and miscellaneous
effects factor (i.e., notch/defect sensitivities factor) to the
endurance limit4 which yields a maximum allowable stress
of 5,683 psi. With an additional applied factor of safety of
1.5, the target maximum stress under fatigue loading con-
ditions is suggested to be 3,800 psi.
For non-cyclically loaded areas, the allowable stress
determination was taken from the ASTM A240 316L
stainless steel value of Sy =25,000 psi. Taking 0.66 Sy,
yields 16,500 psi. This is the value utilized for maximum
allowable stress for pure static loads.
Plate Analysis
The static and dynamic loads were applied to the concep-
tual platework as distributed loads. Plate spans and rein-
forcement sections were modeled as both simply supported
and fixed-fixed structures depending on position and orien-
tation. Plate thicknesses and reinforcement section geom-
etries were repositioned or thickened to ensure plate stresses
were below the determined fatigue limit of 3,800 psi or
16,500 psi given the loading condition.
RESULTS AND DISCUSSION
Based on the analysis, the preliminary design presented
by CAID required some slight, key modifications for the
Metcalf E1B replacement.
The analysis identified that the reinforcing on the
existing and proposed design appeared insufficient while
utilizing the R301 impeller design. The suggested floor rein-
forcement redesign consisted of a composite arrangement
of the upper 3/8” plate with an 8-3/8” deep Tee with 8” x
3/8” flange and 1/4” web, continuously welded throughout
in two locations spanning the entire false floor. In addition,
a 6”x3/8” flat bar welded on edge to the underside of the
floor as stiffeners on approx. 33” centers (3 total) in the
opposite direction as the Tees will be required to stiffen the
floor plate sufficiently against the cyclic hydraulic load. If
the R320D impeller is used, the proposed CAID design
qualifies without modification. It is recommended that the
reinforcing be added in order to keep the flexibility of uti-
lizing the R301 or R320D impellers in the process.
It is also suggested that the orifice be kept at a mini-
mum of 48” diameter for the plate design considered. A
larger opening, if possible, would improve the pressure
design of the system. A larger opening would likely result
in a slight loss in efficiency of the pump mixer, but it is felt
that it would be “small.”5 A cone inlet is recommended to
stiffen the free edge of the orifice and to improve the head
losses through the orifice over the current flat-plate orifice
design proposed. Stiffening of the Divider Wall is recom-
mended. Based on the analysis, a suggested Divider Wall
reinforcement is suggested as two 4”x3/8” flat bars oriented
vertically on either side of the orifice cone as shown in the
sketch, continuously welded on the Organic side. The pro-
posed continuously welded angle sections running along
the perimeter and angled across the face of the Divider Wall
should remain. The Divider Wall should be welded con-
tinuously to the False Floor and box walls.
Stiffening of the lower chamber exterior walls was not
required over the proposed CAID design. It was recom-
mended though to lower the lowest horizontal stiffener by
two inches to obtain a maximum plate stress of 16,500 psi