
Re: How Do I Estimate The Torque For An Impeller Agitator?
25 February 2013 at 11:27amYour anchor impeller does not fit your tank. An anchor impeller for a 23" diameter tank should be 20" to 22" in diameter. The 17" diameter impeller will do little more than rotate the contents of the tank, which will not provide much mixing. However, I can give you a method for estimating the amount of torque required.
The viscous power number for your impeller is 176. The viscous power number is a characteristic of the impeller geometry. I assumed that each of the four blades is 1" wide. An anchor impeller will have a constant viscous power number for impeller Reynolds numbers less than 1.0. Impeller Reynolds number (dimensionless) can be calculated by the following formula:
Re = 10.7 * D^2 * N * sp.gr. / visc
The Reynolds number is dimensionless, the constant 10.7 is a conversion factor and D is your impeller diameter in inches, which is squared, N is the rotational speed of the impeller in revolutions per minute, sp.gr. is the density of the fluid as specific gravity with respect to water (grams/cubic centimeter), and that quantity is divided by the visc the viscosity in centipoise.
The torque for the impeller is as follows:
Torque [inchpounds] = 6.8x10^8 * visc * N * D^3
The torque is in inchpounds, the coefficient is 6.8 times 10 to the 8 power (0.000000068) includes your viscous power number, times the viscosity in centipoise, times the rotational speed in rpm, times the impeller diameter in inches cubed.
Remember the formula for torque is only valid for Reynolds numbers less than 1.0. As the Reynolds number increases, so does the calculated torque. The impeller torque gradually increases as the flow becomes more turbulent as defined by a larger Reynolds number. The torque in the turbulent range could be more than 1000 times the torque calculated by this formula, depending on the amount of rotation of the fluid. Your small impeller may not increase the torque that much since you will not be mixing, only spinning the fluid.
The answers by this expert are based on the best available interpretation of the information provided. The consequences of the application of this information are the responsibility of the user. If clarification is needed, please submit a further question. 
Re: How Do I Estimate The Torque For An Impeller Agitator?
25 February 2013 at 3:04pmThe product mixed is so sensitive, that is why that design has been used. The rotational speed is 100RPM, the viscosity 20 cp and the sp. gr is 1.045. I attach a file with further information on the impeller design. By using the formula you gave on a precedent topic (May 31, 2011), I estimated the Reynolds = 1.8 x 10^4. As the Reynolds was for a turbulent flow, I then estimated the correction factor, then after I've been able to estimate the power of the impeller (0.134 HP). Finally, knowing that : Torque [lbf.in] = 396 000 P [HP] / 2*pi* N [RPM], I've been able to found out the required torque of 85 lbf.in Do you think, I've got the right path? Thank you for your help. 
Re: How Do I Estimate The Torque For An Impeller Agitator?
26 February 2013 at 11:15amThe attached impeller drawing looks nothing like what is typically known an an anchor impeller. The impeller drawing shows more of a wideblade paddle (or turbine) impeller. Unfortunately, using typical correlations and corrections for modified anchor impellers and wideblade paddle impellers give different results for the same dimensional extremes. Because the impeller looks more like a fourblade, wideblade paddle impeller, working from similar turbine impellers seems reasonable. The impellertotank diameter ratio and fluid properties are closer to those commonly mixed with turbine impellers. The best data for anchor impellers is at much higher viscosities and larger impellertotank diameter ratios.
Using a modified correlation for a wideblade impeller, the Reynolds number is about 16,000, the power estimate is for 0.66 hp and the resulting torque at 100 rpm is 418 lbfin. The power and torque estimates could be high compared with actual measurements if the fluid rotates freely and does not experience much vertical and radial flow. The Reynolds number does indicate turbulent conditions, but the flow pattern may be more like Couette flow, which could mean a lower torque.
This situation has a sufficient number of unique characteristics that estimates based on more conventional mixing equipment may not be accurate. The reason for the question about torque may influence how conservative the estimate should be. If the mixer drive is sized based on required torque, a higher estimate is recommended.The answers by this expert are based on the best available interpretation of the information provided. The consequences of the application of this information are the responsibility of the user. If clarification is needed, please submit a further question.