Wednesday, July 30, 2008

Panel-Level Dimming

This strategy involves installing a control system at the electric panel to uniformly control all light luminaires on the designated circuits.

You can control circuit dimming, manually or by inputs from occupancy sensors, photosensors, timers, or energy management systems. Panel-level dimming is a method for dimming HID systems as well as both electronically and magnetically ballasted fluorescent systems.

Continuous dimming is accomplished using a variable voltage transformer that reduces the voltage to the HID or fluorescent circuit.

For example, suppose you are using photosensors in a warehouse with skylights. The high-pressure sodium lighting system could be uniformly dimmed in response to the available daylight from the skylights, saving substantial amounts of energy.

Another application would include a wholesale merchandising outlet that requires higher light levels during normal business hours and reduced light levels during routine maintenance and stocking operations. The scheduling control system would automatically adjust the light levels based on the business-operating schedule.

Although slight improvements in efficiency can result from the dimming of fluorescent systems, slight reductions in efficiency result from the dimming of HID systems. Light output reductions are about 1.2 to 1.5 times the power reduction in metal halide systems and about 1.1-1.4 times the power reduction in high-pressure sodium systems. Manufacturers can provide the specific lumen-wattage performance curves for the specific systems being controlled.

Note that some panel-level dimming systems are incompatible with electronic ballasts. Check with the manufacturer to find out if their variable voltage system is compatible with electronic ballasts and whether the system introduces harmonic currents.

Be forewarned that dimming HID lamps below 50% power may result in a significant reduction in lamp life.

Friday, July 25, 2008

Do Energy Saving Motors..Always Save Energy?

This is an interesting phenomenon. Many types of energy efficient motors have a higher inrush current simply because they are manufactured with a lower rotor and stator resistance. This means that many types of energy efficient motors have a higher full load speed than their standard counterparts. This could be significant when these specific motors are utilized for continuous centrifugal loads such as in pumping and fan operation.

In centrifugal loads, speed changes greatly effect the required energy input, since variations in speed effect the required shaft power as the cube of the speed, while the flow of the air or liquid actually varies directly and linearly with the shaft speed.

Therefore an increase in the load of the driven equipment such as the fan or pump, can result in the increase of the kilowatt usage, despite the overall lower losses of the energy efficient motor.

There will actually then be a cost penalty of going to the energy efficient motor, since the equipment may have to operate at a higher full load speed. This will depend on the type of service the motor is actually utilized for. For example. If you are going to utilize the energy efficient motor on a pump which is used to fill a tank or similar usage where the service is more or less a start - stop type, the higher speed at full load will develop a greater flow. This will simply cause the tank to fill faster and cause the pump to be shut down faster. In this type of usage, the energy efficient motor will generate a cost savings. Although the input to the motor may be greater, the operating time will be shorter. This is where the cost savings can be developed.

If however, the motor is utilized in a system where there is continuous operation, as in a cooling or heating system where the extra flow developed by the faster speed is wasted and where it will often be throttled down by the balancing process, the actual system may generate an increase in energy usage by the installation of the energy efficient motor.

For the above reasons, the entire system must be evaluated when considering the installation of an energy efficient motor; not just the motor itself.


Sunday, February 10, 2008

Emergency Back-up Generation Systems for Buildings

The first Primer I wrote on Back-up emergency Generation was in January 2002 right after 9/11/2001. I am often asked why should any building use back-up generation if they are not forced to by local codes? I think 9/11 proved how important back-up generation is to many types of facilities. But remember if you go with back-up generation there are many important points you must consider.

After September 11, 2001 many organizations directly affected by that calamity were shocked to find that their back-up generation did not function. Many systems that did not operate were systems that were never properly tested as normally required. Most code required systems are tested on a weekly basses in accordance with local code requirements.

A few days after September 11, 2001 many facilities were reporting that their backup generation systems were not operative. Studies indicated that the problems were due to many causes. Contaminated fuel, lack of fuel, poor maintenance and lack of testing and poor design were the basic causes of most of the failures.

Of the cases I evaluated several failed due to improper care of long term storage of diesel fuel and natural gas engines failed due to improper gas pressure. One system failed due to the total harmonic distortion exceeding 30% of the total load in trying to operate three elevators. The generator failed and several people became trapped in the buildings elevators. It should be pointed out that this particular installation was only several months old prior to September 11, 2001.

Many facility managers make the big mistake of installing a new generator with out a proper design. They fail to understand what is involved in selecting a generator that can operate all loads including any motors; with out failing do to increased Harmonic Distortion.

In designing any back-up generator system for a commercial and Industrial building the designer must understand the purpose of the back-up system. The majority of such systems installed are normally legally required emergency generator back up systems. Such systems are primarily utilized for short duration power outages caused by a variety of power interrupting occurrences that are considered non standard as related to the reliability of the power utility system.

These short duration power disruptions are normally in the magnitude of minutes to many minutes but not normally exceeding 2 hours. For this reason most life safety codes require the system to be able to stay on line for back up purposes at least 2 hours. If the owner desires a system that will stay activated for longer than two hours than the designer must coordinate this requirement carefully with the owner so they are aware of the increase in cost that can result from designing a system for longer durations.

When power disruptions are caused by disasters it is essential that the power from back-up generation not be activated until a complete damage assessment is completed. Deaths have occurred after disasters have passed due to electrocution caused by the power restored by back-up generation before a complete building damage assessment is done and before damage is repaired. Fires, grounding problems and back feeding caused by back up generation is known to have occurred because of faulty damage repairs or the lack of proper repairs. NEVER activate your back-up generation system after a disaster with out first doing a complete damage assessment.

For more detailed information please see my book "Emergency Back-up Generation - Design, Operation, Maintenance, and Failures" available at


Monday, December 24, 2007

The Ground was Melting

The top of the world has many interesting features; high amounts of UV, lots and lots of snow, fierce wind storms and the interesting phenomenon that during the extremely short summer; the ground melts. Yes you heard me right. The ground melts. Try to build a building on soil that melts. We had to. And to make it even tougher, we had a critical deadline.

The deadline was selected so we could finish before the peak of the very short summer that occurs at the North Pole. Thing was, however, nothing went right that year and we began to fall way behind in getting our work done. To finish and be able to deal with the melting soil, also known as Permafrost we had to come up with some very innovative ideas.

In short Permafrost is a combination of soil, water, sand and crushed rock. It exists as natural soil in the Arctic area and is usually soil that has been frozen in this state for at least two years. You can build on it as long as you don't leave it exposed above 30 degrees F and exposed to the sun, for long periods of time. If you do, this it turns to mush and runs like mud. That is exactly what was happening because we fell behind in driving the piles for the buildings' foundations.

The air temperature was now in the 30's and the Permafrost was exposed to the sun. In order to drive the piles and keep the ground frozen we devised a method of freezing the ground with liquid Nitrogen. First, we mapped out the areas where we had to drive the piles. I then ran a series of calculations to see how much Nitrogen would be needed at each different site where a pile had to be driven. We had calculated that to get the heat transfer we needed we had to drive four 3 inch pipes around the spot where we were going to drive the piles. We then calculated the amount of Nitrogen we needed and added a 30% safety factor and pumped away. The pictures above indicate what the spots looked like where we had to refreeze the Permafrost with Nitrogen.

Though my calculations were fairly accurate we still had to experiment with the actual location and depth of the Nitrogen pipes. The calculations however gave us a good starting point. To perform the calculations I assumed that the Permafrost was a homogeneous material. This of course was not the case and introduced an error which we had to correct for in the actual placement of the Nitrogen pipes and the amount of Nitrogen required to freeze the earth.

The first parameters we tried caused the Nitrogen to escape much to fast without really freezing the amount of ground that it had to. We tried many different methods of injecting the Nitrogen so we could get just the right amount of Nitrogen liquid to enter at the right rate and to stay in the pipes long enough so as to freeze the right amount of permafrost. The depth of the pipes also had to be played with. First we tried 2 feet, no good...then we kept increasing the depth by 2 feet until we had all the parameters just right. It took forever to do this. By the time we got it just right, the Permafrost melted so much that it actually started to run down hill no matter how level the slope was. Just the slightest unevenness and the Permafrost started to flow.

It took several days to get it just right but because the makeup of the Permafrost varied from location to location we had to keep repeating the process to tweak it for each different pile location. It took at least 10 hours to do just one pile location. The Nitrogen pipes would get so cold that they would crack. The soil conditions were so different from location to location that it sometimes took us hours to get the right combination of parameters so the ground would freeze before the pipes would crack.

It was a long cold summer north of the Arctic Circle, and we used an endless supply of pipes and tons of Nitrogen. Working 24/7 we were able to keep the ground frozen until nature took its course.


Wednesday, December 12, 2007

The Desalination Process

Florida and many other coastal states are moving towards Desalination to supply some of their fresh water needs. Though there are many different desalination processes this review details some of the more popular methods utilized today.

Maintenance and operations for many of these systems can be extensive. On several High Temperature Distillation systems I have worked on the M & O costs were so extensive that efforts to keep them under control became all consuming. On reverse osmosis systems we had to keep very close control of all pumping costs. The following goes into more detail.

Process limitations.
The various desalination processes presently available have limitations that must be considered prior to selecting a desalination process for a particular site. These limitations apply only to the desalination processes themselves; pretreatment can be and is often used to bring a saline feed water within limits so that a desalination process can be used. The raw feed water chemistry for all desalination systems must be evaluated thoroughly for constituents that may precipitate in the desalination system.

a. High-temperature distillation. High-temperature distillation is limited by the saturation of alkaline earth metal salts, such as CaSO4, BaSO4, SrSO4, CaCO3, BaCO3, and SrCO3. Carbonate salt scaling can be controlled by acid addition. The recovery of water from a high temperature distillation plant is usually limited by calcium sulfate solubility. When the concentration of the sulfate and the limiting alkaline earth metal is one third of the saturated condition at ambient temperature, distillation design must include pretreatment to reduce or inhibit the scaling ions. High-temperature distillation is also limited to oil and grease levels below 1 milligram per liter. All other limitations on the high-temperature distillation process are equipment specific and require individual evaluation.

b. Low-temperature and mechanical distillation. Low-temperature and mechanical distillation systems are limited to operation below saturation of alkaline earth sulfates and carbonates. The lower operating temperature permits economical operation on waters that are at or below half saturation at ambient temperature. Oil and grease are limited to less than 1 milligram per liter. Any other limitations are equipment specific.

c. Reverse osmosis. The most severe limitation on reverse osmosis is the maximum limit of 50,000 milligrams per liter of total dissolved solids in the feed water. Another limitation is that there must be no iron in the feed water. This limitation is so rigid that only stainless steel and nonferric materials will be used downstream of the iron removal. The solubility of alkaline earth sulfates and carbonates limits reverse osmosis treatment. Any water containing less than 4,000 milligrams per liter of total dissolved solids that would be saturated with an alkaline earth sulfate when the concentration is multiplied by 1.5 should not be considered for reverse osmosis desalination. Reverse osmosis is limited to waters that do not have silica saturation in the reject brine. Silica chemistry is extremely complex. When the molybdenum reactive silica concentration exceeds 30 milligrams per liter as SiO2 or the pH exceeds 8.3 in the brine stream, an environmental chemist or engineer should be consulted. Reverse osmosis is also limited to the treatment of waters with less than 1 milligram per liter of oil and grease.

(1) Cellulose acetate membranes. Cellulose acetate membranes are usually limited to pH levels between 4.0 and 7.5. Cellulose acetate membranes require some form of continuous disinfection with the feed water to prevent microbial degradation of the membranes and can tolerate up to 1 milligram per liter of free chlorine. Therefore, cellulose acetate membranes are usually disinfected by maintaining 0.2 to 0.9 milligrams per liter of free chlorine in the feed water. Cellulose acetate membranes cannot be used on waters where the temperature exceeds 88 degrees Fahrenheit. Cellulose acetate membranes should not be used at pressures greater than the manufacturer's recommended pressure, since they are prone to membrane degradation by pressure compaction.

(2) Polyaromatic amide membranes. Brackish water polyaromatic amide membranes are generally limited to operation in feed waters between pH 4 and pH 11. Polyaromatic amide membranes are less pH tolerant and should not be used outside of the range pH 5 to pH 9. All polyaromatic amide membranes are limited to use on feed streams that are free of residual chlorine. If chlorination is necessary or desirable as a pretreatment option, complete dechlorination must be effected. Polyaromatic amide membranes are tolerant of water temperatures up to 95 degrees Fahrenheit. While polyaromatic amide membranes are not as
quickly or completely compacted as are cellulose acetate membranes, manufacturer's recommended pressures must be followed to prevent mechanical damage to membrane modules.

d. Electrodialysis reversal. While electrodialysis reversal has been used to treat water as saline as sea water, 4,000 milligrams per liter of total dissolved solids is considered to be an upper limit for economical operation. Some electrodialysis membranes can tolerate strong oxidants, like chlorine, but most cannot. The reversal of polarity used in electrodialysis reversal for removal of scale allows operation on water that is saturated with alkaline earth carbonates. Saturation with an alkaline sulfate with low carbonate alkalinity should be avoided.

Monday, September 03, 2007

Elevators and other Non-Linear Load problems for Generators

For emergency generator systems utilized for life safety the elevators may be the largest non-linear load. As important as these are the designer must make sure that these non-linear loads are not going to create a great deal of stress for the generators. In some cases if elevators are left to operate off the generator for a long duration they may cause the generators to burn out.

When designed to operate for standby power, the emergency generator must be capable of operating the elevators safely and with a great deal of reliability. If the elevators do not operate properly while on emergency power or if the generator can not handle the elevator load for a long duration serious problems may occur.

The most commonly ignored operation of an emergency generator is its ability to handle the elevator loads safely and over a long duration. Especially when the emergency generator system is designed for short duration power outages, let’s say, a class two system. If in fact the generator capacity was selected on a peak load expected to be occurring for short durations, say two hours, but instead the owner try’s to operate the emergency generator for a long period of time say, 48 hours at peek load, the generator may burn out.

Additionally, the owner must check to see if the local codes require elevator standby power testing annually. If more than one elevator is capable of running simultaneously, the local codes may require them to all be tested simultaneously.
The owner operator must also realize that with the advent of SCR and VFD drives for elevators and other devices, existing emergency generators may not be capable of providing the proper power to operate the elevators in an emergency over a long period of time. Additionally, older emergency generators may have a difficult time with the current demand changes that solid-state drives require.

With emergency generator systems current and voltage harmonics differ greatly from that produced by utility power. Improper grounding methods and increased impedance of the generator system can cause additional problems with sensitive devices on the emergency feeder system, due to increased harmonics and RFI. Studies and testing have shown that problems as follows can develop while operating elevators on emergency power:

The Total Harmonic distortion can increase substantially over that expected from utility power.

Grounding can be found to be insufficient.

Voltage regulation as the elevators operate can be found to vary by up to 15 to 25%.

Voltage may dip to values not tolerated by solid state drives.

It is very important that when upgrading to a modern solid state elevator drive that the owner also upgrade older emergency generators that they may be considering to operate all or any of the elevators in the event of a power disruption. If the emergency generator is not to be upgraded than an alternate type of motor drive may have to be specified.

In evaluating the emergency generator-elevator relationship other items effecting the generator operation may also have to be evaluated. Such items that could effect the operation of the generator and how it responds to the elevators are as follows:

What else is operating off the emergency system? Variable frequency drive motors?

Are any sensitive UPS systems expected to be served by the generators?

Are radio and emergency telephone services to be operated from the generator and if so how are they shielded and grounded?

Is the existing generator able to handle the regenerative power from the elevator SR and not have its performance adversely affected?

How will the emergency generators voltage regulator be effected and can it handle the major voltage variation it may be subject to?

Which type of elevator drive is to be installed and how will it effect the generator in actual emergency operation situations. The various types of common drives that may be utilized are as follows:

12 pulse SCR drives with out filters.
Variable Frequency-variable Voltage AC type Non Regenerative.
Motor Generator Set.
Six Pulse SCR drive with filters.
Hydraulic(Across the line)

Friday, August 31, 2007

Magnetic vs. Electronic Ballasts

If an existing lighting system is to be evaluated for changes it will be important to determine if the system is utilizing magnetic or electronic ballasts.

The full output electronic ballast is a high frequency version of the conventional magnetic "core-and coil" ballast. The electronic ballast operates fluorescent lamps more efficiently at frequencies greater than 20,000 Hz.

The full output electronic ballast is rated with a ballast factor of at least .85. This factor actually identifies the output of light from the ballast-lamp combination. The ballast factor is simply that percentage of the lamp's rated lumens actually produced by the ballast lamp combination.

Magnetic ballasts normally have a ballast factor of between .90 to .95. The electronic ballast however can be purchased in a large range of ballast factors. You can purchase an electronic ballast that may range from 1.00 to 1.30 which acts as a booster with the lamp and actually lets the lamp produce a greater amount of lumens then the lamp is actually rated for.

On the other hand, you could purchase an electronic ballast with a range of .45 to .85, which shows that some ballasts can be utilized to actually reduce the amount of light put out by the fixture. If the ballast is a full output type, it would have a ballast factor that would exceed .85.

A partial output electronic ballast is utilized to have a lighting fixture put out less light than that which it is rated. This can be useful for several different types of installations.

A simple way to determine if the installed ballast is electronic or magnetic is to utilize a "strobe top". Some ballast manufacturers supply these free of charge to designers and installers. The top is simply spun directly under the fixture in question. If you see pattern lines, the lights are operating at 60 Hz and therefore are utilizing magnetic ballasts. If the pattern turns out to be a smooth pattern with no lines, the fixtures are utilizing high-frequency electronic ballasts.

Thursday, March 29, 2007

Analog vs DDC HVAC Control

To be sure DDC control through central control rooms have a great advantage for the operators of large facilities. DDC allows you to interface with the control systems, it allows the operators to change the set-points and operating characteristics of terminal boxes, fans etc depending on the needs of the moment. This may be a major advantage in large complicated buildings with a variety of different types of usage's.

DDC control however, may not be required in smaller facilities with very simple operating requirements. In these types of facilities many times it is less expensive to install stand alone analog VAV control. These will include a controller and space thermostat that will operate respective terminal boxes, the combination may be stand alone analog while DDC may still be utilized for central fans, chillers etc.

For example, in a large facility say with 2000 terminal boxes, if stand alone analog is utilized for control of the terminal boxes in lieu of DDC the savings may be considerable. The cost of savings may be as great as that shown below.

Stand-alone Analog Terminal box Control:
Controller, Thermostat & Wiring = $225

DDC Terminal Box Control:
Controller, Thermostat & Wiring = $400

The difference for the Terminal box control can be as great as $175 per terminal box. So, for the 2000 box example the total savings to go stand alone analog may be as great as $350,000. Of course there are DDC short cuts that can be taken that will reduce this cost difference, so the designer will have to evaluate all possibilities.

Remember that DDC will allow interactive communication between the item being controlled and the central control computer. Stand alone sacrifices this communication mode to produce a reduction in price, however, all other control characteristics between the two methods are basically equal, additionally with many stand alone systems wireless or LAN modes can be utilized to tie the systems together.


Monday, February 12, 2007

Fate Was My Partner

It was a hectic year. The Vietnam buildup was in full swing. We were building major projects on the largest military base in the north east. The project we were working on now consisted of 30 barracks a Visitors officers Quarters and a variety of other supporting facilities including several mess halls, refrigeration buildings and a large power plant.

I had a team of 10, we were responsible to inspect, test and accept buildings and systems as the contractors finished the work. This week, everything was moving like clock work, one might even say we were all moving and interfacing like a professional ballet group. Everyone knew what they had to do and when they had to do it. Most of us were working together as teams all year so we knew each other very well and we knew what to expect from each other.

Marty and his team worked for the contractors, they were good very good. This week we were working on the refrigeration building and finishing up some remaining systems that we could not finish before, because we were waiting for equipment. Marty was the Chief welder, he could weld a 6 inch joint with his eyes closed upside down or on his back...he was very good.

It was the Friday before a three day weekend. We all wanted to leave that day at 2 P.M. First thing in the morning I had my usual safety meeting, I rushed it a little. I always covered safety tips during the meeting geared to what we were going to do that day. Marty and his guys were pros, they heard the speech a hundred times before..but they were always polite. After the meeting we had our usual coffee talked about what we were going to accomplish that day and than started work. As we worked we would joke and kid each other and talk about what we were each going to do that weekend. It was a typical day, nothing out of the ordinary.

We broke for lunch, joked, threw a ball around and everyone went back to work as if we were in some kind of a was just automatic. We were moving at a good clip. It was obvious that leaving at 2 P.M. was not going to be a problem. It was about 12:30, Marty and his team were way ahead of my team and we were trying to catch up. I bent down to pick up a water bottle, Ron on my team, who was standing about two feet from me on my right side was installing a test gage. Sidney on our team was just about a foot next to Ron. At that moment we were all lined up, as if we formed a valley and we three were the sides.

Marty and his guys just finished all the welding for the 6 inch connection for a five foot long, one foot diameter steel exhaust muffler for one of the large compressors, it weighed about 250 pounds. The electricians turned the power on, the compressor struggled at first, coughed and at the instant I bent down to get my bottle of water there was a loud explosion. I spun around, Ron threw his hands up to his ears and Sid just turned to look.

I froze as the 250# steel projectile came flying by me missing my backside by about an inch, Sid hit the deck and Ron moved his arm but not enough. The 250# projectile flew by him but managed to skim his arm, enough to give him a mean third degree burn and break his elbow.

The entire event must have taken just 2 or three minutes but to all of us it seemed like an eternity. I rushed to the emergency base phone and called emergency services. With them came a base safety inspector. We spent that three day weekend going over everything that happened piece by piece to see if we could figure out what happened. Marty's welding and silver soldering was torn apart and inspected over and over again. Nothing was found. The inspectors sent the joints out to a lab to be examined 100 different ways...nothing was ever found.

Fate watched over all of us that day. When we went back to work the following week, it was never the same. Why did it happen? How is it none of us were killed? What made that 250# projectile take the path that missed all of our vital parts?

We never found out, and we knew we would never know. But we all learned that day that once fate is the hunter....the event is taken out of our hands.


Saturday, February 10, 2007

Effects of Two Phase Steam Flow

Again, it is important to realize that the theoretical and working flow equations for all flow controls and meters are based on homogeneous flows. Depending on the flow controlling devices principal of operation, the effect of non-homogeneous flow can be considerable. Two phase fluids such as wet steam, play havoc with the accuracy of steam flow measuring devices and steam flow controls. For example. With a device that utilized an orifice to measure and or control steam flow, operating at a constant 150 psig going from saturated steam to about 10% moisture could cause data supplied to the control device to be considerably lower than actual flow, depending on how the moisture isdisbursedd within the steam.

Two significant differences with non-homogeneous flows are that first, the density is not easily derived or measured, and second, one phase of one or more of the fluids components may not be moving at the same velocity as the main flow. Therefore, in these cases some of the fluid may actually be flowing along the bottom of the pipe and be separated out, alternating between a separated flow and an entrained flow in different parts of the distribution system. When the liquid droplets separate out of the flow, the phenomena is commonly called slip. Slip is a complex function of viscosity, particle size, density differences, surface temperature and the superficial velocity of each component. The effects of gravity also play a part in slip and a part in altering various flow patterns. Horizontal, vertical and inclined pipes cause differing relative velocities for the varying components, so they each affect slip differently.

Whenever you have a two phase flow which must be controlled and metered, the void fraction, or percentage of each constitute, must be determined in order to predict the quantity or velocity of each component. Therefore, measuring such a two phase flow with a simple orifice designed under homogeneous equations and equations which apply to saturated and single phase flow, will contribute considerable errors in actual field measurements. Additionally, if the steam pressure supplied to a particular device is above the original design pressure of the measuring device and that device does not have pressure compensation, the combination of wet steam and increased steam pressure will generate a substantially reduced reading from any meter or flow control. For example. At 10% moisture content and 160 psig steam in lieu of 125 psig as the original measuring device may have called for, the differential pressure metering device will register only 85% of what the actual flow is.

One way to protect against meter and flow control inaccuracies created by a two phase steam flow is to utilize moisture separators upstream from the measuring device. If such separators have not been utilized and there are indications that excessive moisture exists within the distribution system, one should immediately suspect that meter readings, flow indications and flow control, are all being adversely affected.

If previously installed moisture separators are constantly overloading and damage to existing orifice plates, PRV stations and meters appear to be caused by excessive condensate, immediate action should be taken to determine the cause of the increase in the wetness of the steam.

Moisture in the steam flow that amounts to over 2% wetness has a tendency to also act as a grinding agent on orifice plates and other types of differential pressure measuring devices, causing enlargements in the measuring plate or device. The coefficients utilized therefore in the various equations and algorithms, though correct at the time they were selected since they were related to the design orifice diameters, will be incorrect when applied against the enlarged orifices which have enlarged as a result of the grinding effect caused by the wetness of the steam. The excessive moisture in the steam will also cause orifice plates to become warped. The excessive moisture, even if it is not in the form of large slugs, will have an eroding effect on orifices, control valves and on the PRV valves located as part of pressure reducing stations. Allowing this to happen can cause serious errors and dangers within a steam distribution system.

For more information on this subject please see the papers I have written and the book I wrote, Steam Distribution and Flow: A Guide for High, Low and Medium Pressure Systems, available at


Friday, August 18, 2006

A Cold Death

It was 1978 and we were building Crimson Towers, a 40 story office building in downtown NYC. It was one of those raw, cold, cloudy snowy days. We were working on the 20 th deck. The salamander propane heaters were firing away, but with the way the wind was blowing at that level nothing could keep us warm.

It was an ordinary day in all other ways. I was making the changes to what are called “As Built” drawings. I used a little wooden table that the large drawings just managed to fit on. On this day we were working with the plumbers. They would do their piping and I would keep track of everything by marking the changes on the drawings. The engineering inspector was Jimmy W.

Jimmy was the kind of a guy you liked one minute and disliked the other. Jimmy watched over all the work for the city. He was a real chop breaker, but he was always the first one to put his hand in his pocket and buy coffee or breakfast for everyone. And if he ran into us after work at the local bar, he would buy us all a beer. The next day he would break your chops like crazy. Jimmy was very over weight. He was 5’ 6” and weighed at least 275 pounds. His diet consisted of fried this and fried that. Jimmy never met a fried food he didn’t like. And Jimmy never missed a desert. And smoke, all day long Jimmy smoked. Never a day went by that we didn’t tease Jimmy about the short life he was going to have.

Jimmy would watch over all of us like we were his little kids. And Jimmy always yelled, do this, do its wrong do it this way and on and on. Every day he yelled and every morning he bought the coffee for all of us and after work he bought the beer. He was truly the kind of guy you always had trouble figuring out. Accept you knew that under all that screaming and yelling there was a guy that cared about what he did, and he did it very well.

This afternoon was ordinary. Why does it seem that so many tragedies start out as just plain ordinary days? This one did.

I was once again marking up the drawings. The plumbers were working on the piping and other related work. And, as usual, Jimmy was yelling. Suddenly Sal heard a funny scream, he was sure one of the guys was fooling around, we all looked around and could see nothing out of the ordinary, come on we joked to each other…”Who’s screwing around?”. After awhile it was obvious to all of us…Where was Jimmy?

Sal looked around…I ran over to the elevator was unfinished but had a wooden barrier around it. In one small area the barricade was smashed as if something or someone went through it, as I looked at it and motioned to the rest of the guys to come over, the call came in over the radio. Jimmy had fallen through the barrier and down the elevator shaft. He fall 200 feet to his death.

The next morning as we reported to the job…Jimmy wasn’t there waiting with the coffee for us or with any of his wise cracks. We all learned two things that morning.

One, Jimmy was good…very good at what he did and two, Jimmy will be missed….And boy was Jimmy missed.


Monday, July 10, 2006

Two Phase Steam Flow

Wet steam takes on the characteristic of a two phase fluid. Since two phase flow creates metering and control problems, changes in the moisture content of steam for any reason in a distribution system becomes very important and must constantly be evaluated. Two phase non-homogeneous fluid flow causes critical deviations from steam metering and control formulas which are based on single phase homogeneous flow.

The thermodynamic state of the steam in distribution systems is commonly defined by utilizing temperature and pressure as the two most important parameters. Pressure and temperature uniquely define the properties of superheated steam. For saturated steam, the additional parameter of steam quality must also be determined.

The problem with steam metering and steam flow control occurs when the steam is left as unsaturated wet steam which is a situation commonly found in many steam distribution systems. Under these conditions the pressure and temperature fail to define the steam properties. This situation occurs because the system actually contains a two phase fluid and not just a finely dispersed vapor fluid combination. Under this two phase flow situation, with wet steam flow, the mixtures specific volume could change radically depending on the ratio between the liquid and vapor within the steam.

From a practical standpoint, one never knows the exact liquid to vapor ratio in flowing steam. Therefore, there will always be a question concerning the make up of the mixture and the homogeneous nature of the mixture along the distribution system. Depending on the velocity of the steam within the pipeline, different amounts of condensate are picked up at different times and entrained along with the steam in different quantities.

The density of this two phase mixture varies along the length of the piping system. This density variation occurs due to steam velocity changes or changes in condensate development rate which occurs in different segments of the steam distribution system. Therefore, along the distribution system, the steam flow is made up of a two phase fluid which has within it water vapor existing as slugs, a non-uniform mixture of droplets or a total separated flow where the liquid itself has separated out of the vapor portion, accumulating on the bottom of the pipe. The effect therefore of this two phase flow on differential pressure meters and controls is very different depending on the device location in the steam distribution system.

Be aware that any kind of flow that is not homogeneous, whether it is slug flow or flow with randomly occurring droplets of water, will affect the accuracy of any type of pressure differential flow measuring device; especially those utilizing an orifice device or annubar, which appears to be the bulk of the measuring devices utilized today. The affects of this type of flow can be the loss of large sums of money in a commercial facility or very dangerous over or under shooting of sensitive controls in industrial and military facilities.
For more information on this subject please see the papers I have written and the book I wrote, Steam Distribution and Flow: A Guide for High, Low and Medium Pressure Systems, available at


Monday, June 19, 2006

DDC - VAV Electrical Controllers Can Produce Faulty Electrical Output

It is usually required that each electrical DDC controller be powered by its own 24 VAC power supply (transformer). All wiring connections to the controller should be via copper conductors unless the manufacturer specifically states that aluminum wire is acceptable.

The wiring connections must all be made in accordance with the NationalElectrical Code and/or the local code as specified. Do not run Level 2 bus wiring in the same conduits as line voltage wiring(30 VAC or above) or wiring that switches power to inductive loads, such as contactors, coils, motors, generators, etc.

Only use shielded cable to run bus wiring. If the runs will exceed 500 feet, use only twisted shielded pair type wire Belden #28 gauge, Beldfoil 8760 or the equivalent as specified by the manufacturer of the controllers. If the runs will be greater than 500 feet, they will usually require a properly designed and selected repeater.(For more details on this topic check out my books on listed as NRC Publications #79, 84 and 37)

To further minimize sensor error caused by field wiring, unless otherwise specified by the controller manufacturer, the total resistance of all passive sensor wiring should be less than 3 ohms. Shielded cable will not generally be required if the sensor wiring runs are less than 50 feet.

When you are dealing with runs of between 50 feet to 100 feet, it is recommended that you utilize Belden #22 gauge, Beldfoil 8761 or the equivalent. With runs up to 250 feet, consider utilizing Belden #18 gauge, Beldfoil 8760 or an equivalent. With runs of up to 500 feet, consider utilizing Belden #16 gauge or Beldfoil 8719.

Wherever a manufacturer is mentioned above, it is for example only. I do not recommend one manufacturer over another. All engineers, users, designers etc must do their own research concerning product manufacturers they wish to utilize.

Wednesday, June 07, 2006

Death Doing the Right Thing

It was a hot humid NYC August day. About 10 A.M. it started to rain and rain and rain. By 1 P.M. it was so dark it looked like midnight. Thunder and lightning occurred almost every 5 minutes and water kept coming and coming. By 2 A.M the next morning there was water everywhere.

Morgan Towers is a 12 story coop in Brooklyn Heights. At the time it was only three months old and was 80% occupied. Water was coming in to the basement area from everywhere. By about 4 A.M. the basement area was flooded and water was pouring down into the elevator pits. The electricity went out and the emergency generator came on. It stayed on for about 30 minutes and failed.

By 8 A.M the rain had stopped, the clouds were moving by fast and slight ribbons of sun light were streaming into the easterly windows.

By 12 P.M. electricians arrived to see what they could do to get the electricity on in the building. The water had receded and the buildings staff was in the process of cleaning up all the mud and dirt. The generator had tripped due to shorts in many of the circuits caused by the flooding.

The electricians were moving along checking circuits and repairing what they could. New wiring and circuit boards were required in many rooms. It was about 90 degrees outside now and about 110 in many of the basement rooms. They were now working in the main air conditioning equipment room. It was very hot, and late. It was now about 9 P.M. and the two electricians remaining in the building were very tired. They skipped meals and were existing on orange juice.

I investigated what happened next for 6 months, the following is a very short version of what I pieced together.

The master electrician went to the back of the A/C room while his helper stayed at the main panel to make sure no one turned any of the power on. They used no lock out tags and relied strictly on each other. It was now about 10 P.M., they were exhausted, but determined to get the building air conditioning on. The master electrician moved circuit by circuit checking for shorts, damaged wire and anything else he could find. He found and repaired many damaged circuits, making many temporary repairs. By this time the master electrician was now out of the line of sight from the helper standing guard by the panels.

From my investigation it appeared the electrician was resting on one of the pieces of equipment, drinking a bottle of juice. It must have been quiet, because about this time the helper yelled out for the electrician to see if he was O.K. We have no way of knowing what the electrician said next but the helper swore he heard something that sounded like "A.O.K. turn the power on" a phrase they had used many times before. The helper testified that almost at the instant he turned the power on he heard a loud scream. He shut down the power and ran to where he heard the scream; only to see his partner on the ground shaking. He immediately gave him CPR and called EMS on his cell. He lost much time as he fiddled with his phone shaking all over. The records showed EMS arrived in about 9 minutes. His partner died on the way to the hospital.

Here he was, a guy trying to do the right thing...and yet he died.

All to often, bad out comes are a result of many different small mistakes adding up to a major disaster. Try as we may, in attempting to get workers to pay attention to the smallest safety error, we find that they try and spend time watching out for the big mistake. They ignore the many small problems that can pile up and turn into a major disaster.

Allowing themselves to become exhausted set the stage; add in working with low blood sugar and the die was cast. They than proceeded with out utilizing lock outs and let themselves get out of each others direct line of sight. They were now heading down hill full speed. Add about 20 technical errors and you have the making of a major disaster, which of course it was.

On construction jobs major disasters are rare but accidents due to the compilation of many small things overlooked are very common. When we tear apart a major accident we find it has grown from ignoring many smaller signs that occurred from ignoring many smaller infractions. Pay attention to the micro event and you may avoid the major disaster.

Be safe,


Saturday, May 13, 2006

Basic Capture Hoods

Local exhaust which results in capture effect is such that the hood utilized; captures, contains or receives contaminates generated by the local source. The hood generates the capture effect by converting duct static pressure to velocity pressure and hood losses (e.g., slot and duct entry losses). The concept is described by the following formula:(Also See NRC Publication's #29, #31 and #84 at

Hood loss (HL) is equal to:EQ (1) HL = K x VP = {SP(h)} = VP Where K = loss factor
VP = Velocity pressure in exhaust duct ,{SP(h)} = Absolute static pressure approximately 4 to 6 duct diameters upstream from the hood entrance.

The capture hood's ability to convert this static pressure to velocity pressure is given by the hood's coefficient of entry C(e). This is further defined as follows:
EQ (2) C(e) = Q(ideal)/Q(actual) = divide VP/SP(h) = divide 1/(1+K)

Any time you enclose the material giving off the contaminate emissions, you will be able to greatly reduce the amount of air required to produce the required capture velocity. You must always keep the source of contaminate emissions as close to the local hood as possible. The hood must also be designed to allow a smooth entrance of air into the hood so that all of the air entering the hood will be capable of capturing the contaminates. The idea of the local exhaust system is to prevent worker inhalation of contaminates.

For this reason, the hood has to be located so that it does not cause the contaminates to move through the occupant's and/or worker's breathing zone in order to make its way to the hood entrance. This is especially true if the hood is to provide protection to workers leaning over an operation which involves utilizing an open surface tank or welding bench.

For a lot more detailed information you may want to review my book "Contamination Control Ventilation" available at Or send me an email if you just want to
explore some thoughts.



Thursday, April 27, 2006

End of Steam Main Collection Leg Analysis

When ever a steam main is started up and it is heated up from a standby cold temperature a large amount of condensate is formed and can work its way to the end of the steam main where a main steam trap is provided to remove all of this discharge.

If an initial warm-up is assumed to take place with a start temperature of 0 deg. F with a condensate forming temperature of 212 deg.F, pressure in the pipe will start to increase as the main stabilizes at the 212 deg.F temperature.

For this analysis we will assume 500 feet of 4" diameter steel main, with the temperature scenario indicated above you will generate 55 # of condensate per 100 ft. Using a factor of 1.1 to take into consideration fitting warm-up and the wetness of the steam the following will apply:

Volume of condensate will equal: 55 #/100 ft. X 5 X 1.1 = 300 #’s of total condensate will develop.
Volume of the condensate = 300 #/62.4 lbs per cu. ft
Volume = 4.8 cu.ft

The collecting leg should therefore be at least equal to 5 cu.ft in volume.
The following considers some alternate pipe sizes for possible collection leg sizes:

4" Pipe:
A = 3.14(Dia.sqd)., = 3.14(16.21)/144 = .355 sq.ft
Vol. = Area (Length) = .355(L) = 5 cu.ft
Length of the collection leg (L) with 4" pipe should equal 14 feet.

6" Pipe:
A = 3.14(Dia.sqd)., = .81 sq.ft
Length = 5/.81 = 6.2 feet

At this point you have several choices you can make; you could simply install the 14 foot 4" collection leg. In most cases however installing very long collection legs can present problems of there own. It is advisable to select the smallest collection leg that will do the job.

A popular way to reduce the size of the required leg is to assume that in long mains the main itself will act as a form of collection leg especially taking in the time factor for the condensate to work its way along the main. When utilizing this method it is often assumed that approximately 50% of the condensate will still be in the main as the collection leg collects the remaining condensate and removes it through its trap. If this is the case then you can utilize either seven feet of 4" pipe or 3 feet of 6" pipe.

As with any analysis of this type you must consider what the site conditions will be before making a final determination as to the proper size of the actual collection leg. Additional information can be found in my book “Steam Distribution and Flow” available at


Tuesday, April 25, 2006

Health problems caused by moisture from duct humidifiers

Moisture from duct humidifiers can constantly accumulate in downstream duct systems and in some cases actually cause large pools of water to form within the duct system.

Humidification in many HVAC systems is of prime importance.It is well recognized that most facilities should be at a relative humidity of between 35 to 60% RH and never above 60%. Fungi, black mold and bacteria develop very well at a humidity above 60%.

One of the more common methods of providing humidification in duct HVAC systems is through the use of clean steam humidifiers, usually classified as the dry type humidifier.

The duct type of humidifier is usually comprised of horizontal delivery manifolds. The humidifier's manifolds are usually placed in a horizontal position in the duct. It is important that when the manifolds are placed in the horizontal position, they
are in a perfectly level position and the discharge holes are facing into the direction of the air flow. In this way, the steam is injected into the air stream against the air flow.
It is also important that when utilizing duct humidification, the system is designed to eliminate the chance of water (condensate) from collecting in the duct and causing breeding grounds for bacteria and other microorganisms. For this reason, the vapor dissipation stage is important to define and evaluate.

When steam is discharged from the humidifier against the air flow in the air duct, it will change from the invisible gas which it is when it first discharges, into a moist vapor with large droplets.... sometimes 8 microns or greater in size. After
being carried along with the air, it will re-evaporate into an invisible steam gas once again.

When the steam first condenses out, it gives up its latent heat of 1000 btu/pound of vapor to the duct air. This in turn causes the air to warm up slightly. As the air vapor mixes, the heat previously given off, re-vaporizes the condensate particles
back into an invisible vapor. It is important for this to occur within a distance that has no obstructions or other devices that will cause the condensate particles to drop out of the air and cause duct wetness and possible areas of microorganism

It is important to have the humidifier controllers out of the area where the visible vapor zone occurs. This is because the combination of the locally warmer air in this zone coupled with the moist vapor particles will create a false indication for the
humidification controllers.

As a rule of thumb, the controller should be at least 15 feet downstream of the humidifier manifold. Other system characteristics will also have to be considered.

The ratio of the duct height to its width is an important factor and is known as the duct's aspect ratio. If all other parameters are equal, if we compare two ducts with the same cross sectional area, the duct with the higher height (larger aspect
ratio) will have a shorter manifold and therefore its vapor output comes in contact with a much smaller percentage of duct air, causing a longer visible vapor slip stream.

This must be considered in selecting the manifolds and in selecting the distance you need for the visible vapor stream to re-evaporate.

Duct temperature is another important consideration in selecting a humidifier. A duct with an air temperature of 75 deg.F can have a visible vapor zone of approximately 12 inches. If the duct air temperature is 55 deg.F the visible moisture zone can increase to as much as 15 feet.

Duct air velocity also affects the length of the visible moisture zone. The higher the air velocity the longer the length of the visible vapor zone.

Other then improper capacity, one of the major causes of actual operating problems is the improper calculation of the position of the visible vapor zone.

Placing controls, insulation and other important system items within the zone causes these items to become saturated with water and to fail.

In some cases I have worked on, final filters became so saturated that they facilitated the growth of aspergillus and other fungi causing serious air quality problems and many illnesses in several different types of buildings.

In many installations the practical method for reducing the length of the visible moisture zone is to utilize multiple manifolds. With multiple manifolds you can provide the full steam capacity you require but at a reduced visible moisture zone
length. This is very important when you have space constraints, airflow temperature is below 70 deg. F, duct air velocities are greater then 500 fpm, filters are utilized downstream of the humidifier, height of the duct exceeds 3 feet and the visible vapor length may impinge upon coils, fans,dampers, filters, insulation (any internal duct insulation that is within the visible vapor zone will have to be removed), duct work, turning vanes, etc. located downstream of the humidifier.

Sample Load Calculation:

Total Air Flow = 23,300 cfm @ 55 deg. F
Min. Outdoor Air = 3,725 cfm, Design = 10 deg. F @ 60% RH.
Return Air = 19,575 cfm, Design = 70 deg. F @ 45% RH.
Max. Outdoor air = 23,300 cfm at 10 deg.F
From standard humidity tables:
10 deg. F @ 60% RH. 0.40 lbs H2O per 100 cfm per
Net difference to be supplied 2.70 lbs H2O per 100 cfm
per Hr.
Under Min. 3,725 Outdoor cfm/100 x 2.70 = 100.6 lbs. H2O per
Hr. required.
For 55 deg. F supply air @ 90% RH. (maximum upper limit) the air
will hold 3.76 lbs. H2O per 100 cfm per Hr. We therefore
cannot attempt to have the air hold more then this amount. To
do so will cause moisture to fall out of the air stream.
19,575 cfm (return air) x 3.1 lbs. H2O/Hr/100 cfm = 606.8 lbs.
3,725 cfm (outside air) x 0.40 lbs. H2O/Hr/100 cfm = 0.40 lbs.
Moisture added by the Humidifier (load) = 100.6 lbs
Total moisture which will be contained by the air = 722.4 lbs.
H2O/Hr which for the 23,300 cfm (total supply air) at 55 deg. F =
722.4/23,300/100 = 3.1 lbs H20/100 cfm; a value below the
maximum which the air could hold which as indicated above is
3.76 lbs. H2O per 100 cfm.

Keep in mind that for human health it is very important in all air conditioned buildings to keep the R.H. below 60% and to prevent condensate from collecting in A/C ducts. If you would like more details on this topic please see the many books I have written on Air Contamination control and A/C system operations.