Blending Using Recirculation

Eductors. They really work.

Here is one source:

Eductors Eductors (also known as jet pumps ejectors, and Venturi pumps) are the most efficient way to pump or move many types of liquids and gases in the petrochemical, process, and power industries. Each ultra-smooth unit complies with ASME/ANSI B16.5 standards. The Tank Liquid Agitator (TLA), with no moving parts, can be used in a wide variety of open vessels or closed tanks. "In-tank" mounting eliminates the need for costly mounting structures above tanks, while investment costs are minimized by utilizing existing transfer pumps.

Water jet eductors utilize the kinetic energy of one liquid to cause the flow of another. Eductors consist of a converging nozzle, a body and a diffuser and resemble syphons in appearance. In operation, the pressure energy of the motive liquid is converted to velocity energy by the converging nozzle. The high velocity liquid flow then entrains the suction liquid. Complete mixing of the motive and suction is performed in the body and diffuser section. The mixture of liquids is then converted back to an intermediate pressure after passing through the diffuser.

Tank Eductors are designed for "in-tank"applications. The TLA operates on the principle of flow dynamics pressurized fluid is accelerated through the nozzle to become a high velocity stream that entrains tank contents and intimately mixes with them. This combined stream exits the TLA at a high velocity creating a flow field capable of causing additional agitation and mixing the tank contents.

Go to:Search GlobalSpec for more.

Nice answer LynLynch.

The OP wants to do this in the tank.

Along with your eductors, Why not in line static mixers?

http://www.coleparmer.com/products/mixersandoverheadstirrers/inline-static-mixers.asp

milo

Yes, might be better, depending on what the OP is up to. Static mixers would indeed blend the product. You might not even need both.

Cheers.

just hv suction of tank and add a pump and delievry of pump should be in tank .

delievry should not b direct entry

say you have delievry of 3 " than take 3" pipe line from pump and make a coil or trident like coil in base of tank with a reducer and line should be 1 or 1/2". the outlet should be dummied or closed.

then make some small holes in coil or trident like system. say hole of 4 mm .

this is placed on bottem of tank. so the pressure will be high it will act like nozzles and thata will agitate your material.

or you can use medium pressure steam also for agitating

Thanks guys! these are definitely some creative solutions, which I will have to look into. However, I didn't communicate my question properly. What I got is a tank with baffles, agitator (that I can't use for this particular case), and an external pump that recirculates the material from the bottom of the tank back to the bottom of the tank. I imagine there will be some mixing due to the turbulence from the recirculation. What I want to figure out is how much and what quality of turbulence this recirculation will create. In other words, how much volume I can agitate using just the external pump or how long to get the desired mixing for given volume. I figure these would be factor of viscosities, rate of recirculation (gpm), the height difference between the recirculation discharge and suction ... Ideally, I would have liked to just do an experiment and figure out... much easier and reliable...but I don't have that option. So, if there is some method of estimating on the mixing one can achieve due to recirculation of fluid...that would be great...does any of this make sense? Sorry about the confusion

I would suggest that you do a simple arithmetic estimate of Tank "turns" by dividing quantity tank holds by gpm (gallons per minute) to see how many minutes it takes to do one turn or turn over.

My guess is that in the time it takes you to get 4 or five turn's there would be no measureable segregation.

Your goal is homogenization with in the tank, right?

milo

As usual, Milo is spot on. I believe his solution is simple and effective.

How do you know you have the right combination of chemicals?

You could always automate the process, maybe. There are sensors for pH, ORP, conductivity or whatever quality is important to you. Then hook it to a dosing pump with a simple control circuit and maintain a constant balance by adding make-up chemicals as needed. Depending on where you place the sensor you might want to check the control system lag and make allowances.

Remember, the more detail you can supply the closer we can come to a solution.

Good Luck.

Thanks guys! Yes, I am looking for homogenous solution.

I will get to "turn-over" calcs on Monday when I get back to work. I am just getting introduced to blending; so, let's see if I understand the "science" a bit here.

So, if I can dare compare with impeller mixing:

1) with impeller, the turbulence is created at its surface and depending on the type of impeller this turbulence is transferred to certain region of the bulk

2) with certain impeller rpm, the turbulence intensity can be ccontrolled to see how far from the impeller the turbulence is transferred

These will determined the mixing time and quality with the impeller.

With recirculation loop,

1) the discharge will create turbulence and depending on the discharge location and angle the turbulence will transferred to near by bulk region

2) with certain gpm (and/or nozzle restriction as suggested by a guest), the turbulence can be transferred to certain distance; however, not in the entire bulk region

3) enuf turn-overs of the bulk volume will ensure the "transfer" of turbulence to entire bulk

What determines the number of turn-overs? The miscibility of the liquids? the viscosities? The volume to be mixed (I imagine that will primarily impact the time rather than number of turnovers)? How did 4-5 turnovers come about?

Let me know what kind of information will help you guys educate me. I am bit wary of providing any information unless needed, as I am not sure what would be too much.

Thanks!