Water Supply Design Problem

What tells the dying machines to take water?

Whatever that is, hook the pump controls to that.

all the 5 machines run on their own stand alone PLC(barcosedo).

Let me restate the question to make sure I understand it.

You have 5 machines that each have a pump driven by a VFD - yes?

You are feeding water into vats of dye - yes?

You wish to provide feedback to the VFD (or PLC?) from an analog pressure switch?

or is your switch digital (has only an ON and a OFF output?)

Please clarify. Your help will only be as good as how you state the problem.

lyn, you hit the nail on the head. Bulls eye. GA for you. The best solution, simply said.

The advantages of using the PLC for control are several. In this case, you can use the level sensor to determine when to turn the pump off. And you can add time delays very easily as necessary.

A VFD, on the other hand, may have an analog input but that is generally used for speed control. Trying to shut a pump off using speed control may be desirable, but your control circuit will be very sensitive to issues that may or may not be controllable using other settings in the VFD. Also, you must consider what will happen if the sensor were to fail in the "Maximum Speed" condition. My guess is that you will have a lot of overflow problems that way.

A PLC gives you many more options to start and stop. So, if you must use an analog sensor, just buy the right kind of PLC card and connect it to that.

Also, if you intend to keep the vat full by adding water, it is most likely best to run the pump at a reasonable speed. How will you know when you need to add more dye?

If you try to dribble water in using the lowest frequency of the VFD you will find it very difficult to manage information with respect to the mix. Analog control systems are loaded with variables that you must define exactly or the result will drift off into other problems. Digital control systems are much easier to tune (or adjust).

How critical is the level in the machine?Will the level affect dwell time of the product,such as in a continuous rope-type dying operation?

In this instance liquid is constantly being removed by the product and must be replaced at the same rate as removal to prevent a difference in exposure time.

If so, proportional control is going to be the proper solution.Using a PLC and a controller with PID features,offset can be practically eliminated;with ON_OFF control there will always be an offset between set point and process variable.

If however, this is a batch dyeing operation, and you only fill and drain the tank for each cycle, then the low and high fill switches will work just fine.

A couple of magnetic reed float switches would suffice.Google GEMS Reed Switches for one source.There are many others.

I can't say much about your particular industry, but where I live (power plant), we turn on the RO system based on RO permeate storage tank level. Usually in the design of any water purification plant (especially and RO system), care must be given to 24-hour water demand, then match the size of the RO system to that, and add a storage tank with enough capacity to meet peak demand during the 24-hour day. In this way, the RO system runs for as many hours as possible (this is better for the RO system), and your dye vats can take on as much water as required without having a start-up and shut-down sequence on the RO system.

Or if you had rather not pay attention to the details, simply do what lyn suggested.

Q1 You have 5 machines that each have a pump driven by a VFD - yes? ans: all 5 have their own pump for filling, the pumps run by PLC command (a relay switches the magnetic contactor on ) Q2. You are feeding water into vats of dye - yes? ans: this is a batch process, after batch has been completed, machine takes in water for next batch, all 5 are connected on a water line comming from R.O plant, should i consider a maintained water pressure in the line? or consider only the filing time of machines? Q3.You wish to provide feedback to the VFD (or PLC?) from an analog pressure switch? ans: that is an option, is this a good option? Q4: or is your switch digital (has only an ON and a OFF output?) the electric panel in the machines have several relays, one is for filling control operated by the output of plc, should i use single pressure switch on main line to maintain reasonable pressure or use the normally open point of filling relay of all machines to switch on the VFD on the main pump in the R.O plant?

Yes, and make sure this instrumentation specialist also knows something about how an RO plant needs to operate, otherwise you might solve one problem only to cascade the problem into another area of the plant.

(I.E. - the RO system runs for a very short time each fill, does not get a proper chance to rinse in (yes ions don't stop flowing through the membrane at low flux, even though water flow stops), and puts too many starts and stops on the motors that pressurize the membranes, thus leading to early failures, valves wearing out, etc.; OR the RO system runs at too high a production rate, fills up a storage tank, then sits idle the rest of the day (also bad, since this will lead to biological growth on the membranes (fouling), and early failure of membranes.)

If the RO is producing too much water, a simple way is to remove membrane(s) from each vessel (has to be done in a way that preserves membrane array symmetry especially when the system has staged feed from one set of vessels to another), and put spool pieces (available from membrane vendor) in to take the place of the removed items. This can only be carried so far, without making too drastic a change on the flow conditions on the RO booster pumps. Consult with your membrane/ RO system provider about these options.

Well, the process is wrong, for starters. RO units should be kept running all the time, with the reject flowrate being managed by balancing the water recycling and make-up streams.

Forget the electrical stuff for a while, and get the process control philosophy written properly and agreed first.

Actually, the reject flow rate needs to be controlled based on membrane array parameters and water quality of the feed. Most medium sized industrial RO units can easily maintain upwards of 75% recovery of permeate (less than 25% reject/concentrate), as long as the water is properly conditioned, and the membrane array supports this. Household units typically would be doing really well to recover 50%.

Otherwise PWSlack and I agree completely - i.e.- get the process control philosophy figured out and documented, and then the electrical control part becomes more facile.