Sometime in 2003 GE Power purchased Osmonics, a maker of reverse osmosis (RO) membranes as well as complete, skid-mounted RO systems. At the time, Osmonics had an excellent software program (as well as excellent equipment) called Winflows that allowed you to design and model RO systems. From the beginning this software has been one of my favorite RO programs. It had many features, including an auto-design feature, but even though this software was feature-laden the program was always very easy and intuitive to use. After GE acquired Osmonics further development of the free Winflows program “seemed” to stop. Or, for a long time, that’s what I thought. But then GE came out with an updated version of the Winflows program and it’s better than ever. It’s now GE Winflows 3.0.
You can download the free Winflows Membrane System Design Software by going here: http://www.gewater.com/winflows.jsp.
When you first open Winflows you are presented with the screen shot shown below.
You can begin using the program in many ways but really, the most important initial input is a detailed water quality analysis. The Winflows screen for this is shown below. Once you’ve entered this key data you can quickly get down to the business of designing and modeling your RO system.
If there are any issues with your system design or any water quality concerns the program will alert you as shown below.
The Winflows program produces an excellent analysis of scale potential in the feed water and the graphs are great for use in presentations. You can modify your design using pH adjustment, softening, and antiscalant addition and then regenerate your design to check again for any problems. This is yet another highly recommended, free program. If you do any work with RO systems this is must-have software.
I want to show you just a few more screen shots. For these I’ve picked a different model, different water source. The process flow diagram for this example is shown below. You can see this is a small RO system with a feed water flow rate of only 70 gpm and a product water flow rate of 35 gpm resulting in an overall system recovery of only 50 percent. Note that no significant pretreatment is provided as indicated by the “Untreated” box and arrow shown on the graphic.
When you “calculate” this model, a required step to have Winflows analyze the system and determine if there are errors, I generate a lot of errors as shown below. The error message below shows 11 problems!! We’ve got our hands full trying to figure out what to do with this system. But before trying to correct these problems let’s look at the scale potential graph this program does such a good job with. That’s shown in the graphic below the error message box.
Note that I’m taking advantage of a great tool built into the GE Winflows program. It’s GE’s Argo Analyzer. And not again that the graph showing scaling potential, which shows we have a problem with scale formation from calcium carbonate, calcium fluoride, and calcium phosphate, is based on an “untreated” feed water.
So now I’m going to introduce some treatment options. The first approach is to step through a wide selection of GE’s antiscalant products. One product that looks very good is the MDC 150 antiscalant. When I select that all of my scaling concerns from the graphic above are now gone. I even have a product feed rate. In the graphic below you can see that I’m going to need to dose the MDC 150 at 2.59 ppm. That’s equivalent to 0.99 kg of antiscalant per day. Not a bad way to reduce the scale potential. We’ve got some other problems to address with the SDI, membrane flux rates, etc. So we still need to model different combinations of pressure vessels and membrane elements. That’s just a step-by-step process you go through in modeling any RO system. In this example I just wanted to show another aspect of how complete this software program is in terms of selecting pretreatment options which can include the addition of a softener, pH adjustment, and the selection from a wide range of antiscalant products. I think you can see this is a program with a lot of modeling capability.
If you search the Web you’ll come across different versions of this filtration spectrum. But in all the years since I got this particular spectrum from Osmonics (now part of GE) I have not seen a better one than this. If you click here (or on the image) you will open up a nice full-page, high-quality, PDF version that prints beautifully on a color printer. I often print this a heavy stock or thicker high-gloss paper for a really nice look and feel. This is a handy chart that I keep in a “technical” folder for quick reference.