[ News ] [ Services ] [ Products ] [ Mixing Forum ] [ About Us ]

[ Up ]

Importance of baffles

Baffles are needed to stop the swirl in a mixing tank.  Almost all impellers rotate in the clockwise or counter-clockwise direction.  Without baffles, the tangential velocities coming from any impeller(s) causes the entire fluid mass to spin.  It may look good from the surface seeing that vortex all the way down to the impeller, but this is the worst kind of mixing.  There is very little shear and the particles go around and around like in a Merry-Go-Round.  This is more like a centrifuge than a mixer.

Most common baffle type

Most common baffles are straight flat plates of metal that run along the straight side of vertically oriented cylindrical tank or vessel.  This baffle design is often referred to as "standard baffles".  There are many nuances.  Some may have the baffles extend into the bottom dish or bottom head.  Some may go up the straight side only partially (so-called partial baffles).  Some baffles are flush with the side wall, but the majority have a space between the baffles and the tank's wall.  They are off-wall.

Number of baffles

Most vessels will have at least 3 baffles.  4 is most common and is often referred to as the "fully baffled" condition.  This basically means that adding any more baffles doesn't significantly add to the power consumption of the impellers.

Baffle schematics courtesy of P. Csiszar

[Top of this page]

Baffle width

Baffle width, w_B, is a function of the viscosity.  For very high viscous fluids and broths, baffles are not even required, because there is enough resistance to flow at the walls.  As the viscosity decreases, baffling becomes important and the baffle width gets larger.  Guidelines for baffle width will be here soon.

In turbulent flows two standards of baffle widths have emerged:  Metric standard (w_B=1/10) and the American standard (w_B=1/12).  The difference is not large, but it is worth noting.  Since the metric system is built on tenths, it is easy to determine baffle width.  If the tank diameter, T, is 3 meters, then the baffle width is 300 mm.  Simple, right?  That is not so simple in those countries using inches and feet for linear dimensions.  1/10th of 9 feet is 0.9 feet and is not easily converted to inches.  Using the 1/12th rule (0.083), the baffles for a 9 foot tank will be 9 inches in width.

It is important to understand this small difference, 0.083 vs. 0.1, when looking at power numbers and flow patterns.  If the author of the paper comes from Europe, Asia, or Africa, chances are that the Metric standard is used (1/10).  North American, Central American, South American, and British may be using the American standard (1/12).

[Top of this page]

Baffle height

Most baffles start at the bottom tangent line of the lower head and run up to the top tangent line of the upper head.  Some do extend down into the head.  If you have consistent batch volumes or weights, it is best not to extend the baffles through the liquid surface, unless you want to entrain air into the batch.  Baffles extending through the surface can create vortexes behind the baffles and entrain air.  In viscous media, the baffles can create dead zones on the surface, and pack-up the material in front of the baffle (in flow direction).  If you are adding sticky solids on the surface, the baffles are a great place for materials to agglomerate.

[Top of this page]

Other types of baffles

Baffles for glass-lined tanks

Straight flat baffles are not found in glass-lined tanks.  The types of baffles for glass-lined tanks often depend on the manufacture.  In general, though, the baffles are supported from the tank top.  Baffles usually take on names that describe what the baffle looks like, such as Beavertail, F-baffle, and D-baffle.  

The Pfaudler Werke GmbH in Schwetzingen, Germany has kindly notified us that the F- and D-baffles are designs from the past.  They have come out with a brand new patent-pending baffle design called the C-baffle, that looks to us like it would be ideal in metal tanks, too.  Like all other baffles for glass-lined tanks, it attaches to the top of the tank through an available port.  They can also add probes to this design.  The orientation of the C-baffle to the flow is interesting and is for baffling what the Smith Turbine or RS-6 is to radial flow turbines. Because of its design, the port can still be used to add liquid and gasses and be used as a vent if required.  This makes it ideal for gas-liquid reactions in glass-lined tanks that utilize the mass transfer across the fluid surface from the headspace, like hydrogenations.

This baffle is shaped like an open C whose opening is directed against the direction of flow of the fluid moved by the agitator.  The geometry was optimized with respect to minimized bending stress using the Finite Element Method (FEM).  Have you ever seen a flat baffle buckle and bend under the stress of the mixing action?  We have!  This shape will keep that from happening.  The c_w-value of a C-section is significantly higher compared to the profile of a classical paddle-type baffle. The c_w-value is a characteristic quantity that describes the flow resistance of a body.  Higher c_w-values indicate a higher resistance to flow, and a better disturbing effect in agitating applications.  The importance of baffling has already been discussed above. 

Since glass-lined baffles are attached to the tank top through an available port, ports can be scarce, especially on the lids of small tanks.  Well Pfaudler came up with another patent-pending idea they call the BaffleRing.  It has a ring-shaped component with two C-baffles attached to it, which is directly inserted between the lid and the lower reactor body and fixed by the reactor clamps. Thus, all reactor nozzles are available as process ports.  Furthermore, a so far unknown level of flow disturbance can be reached inside glass-lined reactors which reduces the mixing times by up to 60%.  Not bad!  Finally a great way of reducing the swirl in glass-lined tanks.  Look at the table below how it improves mixing times!

You can read all about these innovations in a newly released brochure from the Pfaudler Company.  This PDF-file is in English.  If you want it in German please contact Pfaudler directly.  If you reside outside of Europe and Asia and you want to know more about this interesting development please contact Pfaudler in the USA.

Another baffle-type is called the Quatro-Pipe.  This unit is a dip pipe with the outer shape of the classic Paddle-type baffle. Thus, the Quatro-Pipe acts like a baffle, but offers the added value of introducing or removing a fluid into/out of the reactor through an additional nozzle.  But think about this for subsurface aeration, hydrogenation and other gas-liquid applications, too.

[Top of this page]

Baffles in the tank bottom

When axial flow down-pumping impellers are used, a crucifix baffle in the base may be all that is needed.  The crucifix baffle is comprised of two plates that cross (usually in the center).  They should not be placed directly on the bottom of the tank, so that solids don't get stuck.

[Top of this page]

Baffles in side-entry tanks

A flat plate baffle in the outlet stream of a side-entry mixer can help to straighten the flow and reduce the formation of a vortex emitting from the impeller.

[Top of this page]

Fluid forces on baffles and baffle plates

To design an agitation system - tank, shaft, impellers, and baffles (or baffle plate) - each component must be built strong enough to withstand the fluid forces generated by the impellers.  Many calculations and programs are based on conservative hand-waving ideas based on the torque on the impeller shaft times some multiplier.  This really isn't necessary anymore.  The Acusolve CFD program that we use, can accurately determine the fluid forces on baffles, shafts, and the tank walls!  Email us if you would like a quote to determine the fluid forces of your mixing system.

[ Home ] [ Up ] [ Mixing Definition ] [ Nomenclature ] [ Static Mixers ] [ Baffles ] [ Impellers ] [ Coils ] [ Tanks ] [ Macro Level Mixing ] [ Micro Level Mixing ] [ Phase Properties ] [ Scale-Up ] [ CFD & CFM ] [ Meetings ] [ Mixing Companies ] [ Literature ]

Send mail to webmaster@postmixing.com with questions or comments about this web site.
Copyright © 2003,2004,2005 Post Mixing Optimization and Solutions
Last modified: Feb 6, 2005