Despite the fact that our cooling towers are galvanized steel, sitting on steel I-beams attached to 8" steel pipe, which in turn is connected to a lighting arrestor system, we seem to have a problem with some type of electrical/radio noise that is being transmitted through the water to some test equipment. (See VFD FRI/EMI in this forum)We are trying numerous filters on the VFD drive outputs and inputs and so far are having no luck in filtering out whatever it is. The electrical system grounds from the motors back through all panels to the SE and main building ground have all been tested at under 0.02 ohms. Is there any way of trying to trap the interface/voltage/current in the water and take it to ground? Would this cause an unsafe parallel ground situation?
I don't understand what you are asking. On what conductor(s) is this undesirable signal?
How is the equipment malfunctioning?
How did you trace it to the cooling tower?
On the other hand, if you want to attach the water tower to an additional grounding electrode system that is bonded back to the electrical service, I see no problem. I also at this time do not see how this will help, unless the tower is in direct line of a microwave transmitter.
Dealing with parallel equipment grounding paths: Parallel paths are to be avoided in telecommunications, namely radio. Typically even in a telecommunications room in a building there are 2 paths for ground (with exceptions), one the electrical power circuit (and may be isolated type) and the other for bonding the racks etc., and both are required to be bonded at the service equipment.
#130482 - 06/29/0608:45 AMRe: Best way to ground water????
The undesirable signal/noise appears to be comming from the VFDs, the fan motrs that they drive, or the fan itself perhaps creating a static charge in the water. It does not appear to be reaching this meter through anything but the water that the meter is monitoring. WE have one more test to perform to verify this beyond all reasonable doubt. We are trying to get the Powrs that Be to allocate additional funds for testing to try to figure out what type of noise we are dealing with, but if it doesn't affect a critical aspect of the daily operation they don't see the point. I can keep trying to throw different filters on the drives which in the end will cost more.
#130483 - 06/29/0610:04 AMRe: Best way to ground water????
Part 2: Sorry had to run and take care of another issue. As for the other questions. The TDS meter samples water from the cooling tower. This is the meter that is affected. The reading is normally stable and increases as water evaporates from the cooling tower. From the calibration point the meter usually rises at a steady rate changing by a factor of 10 over a period of about 30 minutes. When the fans and drives are ramping up and down, it can change by a factor of 300 in the same time frame. At a pre-set conductivity level, the meter trips a valve which allows water to escape from the towers and fresh water is added to lower the amount of dissolved solids and there-by the conductivity of the water. What has been happening is, that when the tower fans ramp up to maintain the temperature the conductivity reading starts to fall and vice versa. As long as the fan speed is stable, so is the indicated conductivity. There seems to be a time delay factor with the reading chasing the fan speed at about a 45 second interval. We moved the meter out of the tower yard and put it in the chiller mechanical room inside the building. When water is supplied to the meter from downstream from the chillers the meter tracks normally, powered from any source, including power from the tower yard. When water is supplied from the tower yard or anyplace up to where it enters the chillers, the meter operates erratically. It appears that the steel chiller barrels and the copper tubing is somehow damping down the noise to a point where the meter can operate. Any attempt to date to ground the towers, motors, conduit, water pipe and anything else we could think of has proven to be ineffective. Until I can find a way to get money for a signal analyzer test authorized, I don’t know exactly what I’m chasing. We surmise that the drives are the root of the problem, but why now after 6 years? Without knowing the nature of the noise, you can’t choose a filter to do the job. We are trying to figure out a way of taking the noise out of the water at the tower as a means of fighting a holding battle until we can identify exactly what we are dealing with. How much surface area will be needed to capture the noise? Will taking it to earth ground in the tower yard work if we attach it to the lighting arrestor system? Any and All suggestions are welcome.
#130484 - 06/29/0602:24 PMRe: Best way to ground water????
There seems to be a time delay factor with the reading chasing the fan speed at about a 45 second interval.
That indicates to me that the issue is not electrical, but PHYSICAL. I am guessing that the time it takes for the water to travel from the cooling tower to the sensor is about 45 seconds.
when the tower fans ramp up to maintain the temperature the conductivity reading starts to fall and vice versa.
When the water temperature changes, the concentration of the H3O+ and the OH- ions changes. That’s why pH, conductivity, & salinity probes have to be temperature compensated. When the fan speeds change, the water temperature changes. Is downstream of the chillers a constant temperature? If so, that’s why the meter tracks actual changes in TDS.
1) Test this theory by sampling the water downstream of the chiller. Then heat up and cool down the sampling line. Verify the meter readings increase with heat and decrease with cold 2) Verify the TDS meter probe temperature compensation is turned on and working.
#130485 - 06/29/0605:08 PMRe: Best way to ground water????
The TDS Probes are temperature compensated.The Towers maintain one of two temperature:75 or 80 degrees,depending on chiller load and can usually hold it to within 1 degree. Maximum delta T for any set temperature is three degrees. This holds true until the tower approach curve starts to fall as the dewpoint rises above 45 dreegs or 105 degress drybulb which ever occurs first. Maximum shift in TDS is 20 micro siemens per degree. Four different TDS meters all respond similarly in this location. I have been working with this equipment for six years, 5 days a week and monitor the meter and cooling system at least once an hour while at work. This is not normal temperture/TDS swings by any stretch of the imagination. Temperture increases are by one or two degrees over 4 or 5 minutes. The TDS swings are hundreds of micro siemens in seconds. I can't correlate the rate of rise and fall in temperature with the rise and fall of TDS.
#130486 - 06/29/0606:34 PMRe: Best way to ground water????
How about a crud burst? Perhaps as the fans change speed, they produce a resonant frequency that shakes loose deposits that are in a low flow region?
Has there been any changes in the water system that would have released plated out deposits? Any recent pipe replacement, changes in water chemistry, physical work on the cooling towers, additional loads added to the system, etc.?
#130487 - 06/30/0608:50 AMRe: Best way to ground water????
Larry, The system has been in operation for 6 years in its present configuration. There have been no changes that we are aware of other than aging. We are thinking that somehow the tower has become insulated enough from ground and is acting like a battery or capacitor and that whatever this noise is being "stored" in the water. For the record, a hand held MYRON TDS meter is used to check calibration and its operation is unaffected. We are going to move the meter in the chiller room back outside along with its water source from the chillers. If it remains stable, that will confirm that the water is the source of the drift. In previous tests, this meter was moved to multiple locations with-in the tower yard and suffered drift despite multiple power sources including UPS clean power. This at first led us to believe we were dealing with RFI which may still be the vector fot the noise entering the water. If the meter starts to drift again under these new conditions we believe that would indicate that an RFI problem does exist and is strongest in the tower yard enclosure. The only added variable is that we have to use different, longer hoses to make the new connections. The present hoses are a white vinyl and the replacement hoses are rubber. The interior lining of the rubber hoses is black and in rubber products this could mean the presence of carbon? I couldn’t get any continuity indications with a megger so I’m hoping it doesn’t act like a wave guide. How’s that for paranoia? Thanks for all your input. It is good to bounce this off of someone else with a fresh viewpoint. I have pictures of the tower yard, power schematics etc. available if I can figure out how to post them.
Many thanks, Jim
#130488 - 07/01/0612:08 AMRe: Best way to ground water????
Could be anything...That's why it is so frustrating. But all seems to indicate that it's in the water. We have tried various schemes to try to improve the contact between the water and the towers and the towers and earth ground and nothing seems to have worked. We contacted the testing company that did the original baseline ground data for us. They will come back out to take another look see. I'm hoping that with the additional test that we have run, they will be able to take a fresh look. I tried using my meter to see if I could read any voltage potential between the water and earth ground, but with the water in constant motion the readings on the digital are ambiguous at best.