Energy

Jul

Virtjoule Vibe Monitors Natural Gas Pipeline Compressor Stations

Posted by randy at 6:14 pm

Summary

You might call a 300 hp Caterpiller natural gas fired engine, powering a huge fan and compressor, the ultimate piece of HVAC equipment. We call it just another day of critical machine monitoring for Virtjoule. Instead of monitoring machines to cool contents of walk-in refrigerators or occupants of buildings, this time the compression and cooling that is occurring is for the benefit of natural gas being pumped into the interstate gas pipeline system.

A solar powered version of Virtjoule Vibe, is uniquely suited to monitor these large pieces of equipment at Priority Oil and Gas pipeline locations in western Kansas. Using a vibration microphone, Virtjoule Vibe monitors vibration from the huge compressor engines and can tell the oil field service, LaRana Resources, when they are down. A previous system requiring a long distance radio connection was much more power hungry, expensive, and less reliable than the Virtjoule system.

Caterpillar engines that drive the compressors can fail from a variety of reasons including engine oil temperatures, bitter cold and frozen gas fields in the middle of the winter which starve the gas supply, blown turbo chargers, cracked cylinder heads, cracked manifolds, etc. Virtjoule isn’t figuring out what those reasons are, that’s LaRana Resources’ job. But Virtjoule is on guard 24 hrs a day ready to alert LaRana field engineers if the compressor shuts down. If you thought they just punch holes in the ground and watch the bank account grow, you would be wrong.

Natural gas fired compressor motor for natural gas pipeline

Key Points

Virtjoule Vibe gives more information and is more reliable than other oil and gas industry specific solutions.

Multiple alert mechanisms give Virtjoule Vibe flexible 24 hour alert capability including text, email, and voice callout.

Applications where there is no electrical proxy for runtime activity makes Virtjoule Vibe a clear choice.

Cellular coverage from our national carrier, Verizon, is amazingly good and the sensors can pull in almost anything if it’s there. The nearest cell tower that we know about is about 15-20 miles away from the various locations that we’re monitoring.

Remote locations can really benefit from our cellular approach. Even though there is cellular coverage, there is no local internet connection or phone line anywhere close to these machines.

Role of Natural Gas

It doesn’t matter where you stand on the expansion of natural gas production in the United States, most people agree that domestic natural gas production has helped create a certain amount of energy independence and is one of the cleanest methods of producing energy from carbon sources that we have. The underground Niobrara formations in eastern Colorado (Phillips and Yuma counties, and others) and Western Kansas (Cheyenne County), have been prolific producers and Priority Oil & Gas and LaRana Resources are teamed up on wells that are tapping these resources. I grew up in northeastern Colorado and have seen natural gas production expand many fold over the years in that area. My chemical engineering and petroleum refining education has given me good insights into the processes required to bring natural gas to market.

The Compressor Station

The natural gas compressor station provides a critical role in getting natural gas to the interstate gas pipeline and on to the market. These stations collect gas from many wellheads in the area and compress that gas up to the pressure required to move it into the pipeline. The gas is compressed in several stages as it would be much too difficult to compress to pipeline conditions all in one stage. Also, as you compress any type of gas, it will heat up. Gas that is compressed must be cooled some amount after it is compressed to avoid adding a lot of extra heat to the pipeline system.

The photo below is of one compressor station Virtjoule is monitoring. 12 VDC power is used to power the sensor that can run on anything between 12 VDC to 24 VAC. That 12 VDC power is provided by the solar panel you see on the building. Inside the gray box under the solar panel is a 12 V deep cycle marine battery and a solar charge controller.

This gray box is also housing a much more expensive monitoring system specifically designed for oil and gas companies that Virtjoule is replacing. Both monitoring systems are being run in parallel, but much of the time the other system is not operational.

The huge fan that you see on the left hand side of the photo is part of a gas cooler that itself is attached to the big Caterpiller engine inside the building. Air is exchanged across tubes in the cooler while the hot gas flows through the tubes. Virtjoule’s vibration microphone is literally attached to this cooler just inside the building. The cooler has a hard connection to the compressor and engine and so it vibrates with the engine. If the engine isn’t running, nothing else is happening in this building. Compressed gas is then routed outside the building to go through a dehydrator to remove extra water vapor that comes up naturally with the gas. Some fields have more water in the gas than others and the amount of dehydration needed varies from field to field.

As you can imagine, safety is important when working around this much natural gas. There is no other electricity in the building except from the solar DC system. In a smaller booster pump station that Virtjoule is monitoring in the same field for an owner based in McAllen, TX, Virtjoule is powered off of a standalone battery and works for many weeks before they replace the battery with a charged one. LaRana engineers visit these sites at least once a day, so occasionally putting recharged 12 V battery in is no big deal. Virtjoule’s sensors are very low power and so the drain is not large. Adding Virtjoule to the current solar setup has not caused any problems.

Natural gas pipeline compressor station

Fault Detection

Although these engines have throttles and they are sometimes operated at less than full throttle, for the most part they are running all-out all of the time. This is pretty straightforward runtime fault detection for Virtjoule. It’s always supposed to be running. There’s no purposeful cycling and no short cycling faults like you might see in HVAC machines.

As a point of interest, if you’re standing in the building shown above, the ground shakes. The building shakes. There is no missing the fact that the compressor is running. These effects far overwhelm any noise from wind or rain. We’ve had no problems distinguishing the difference.

LaRana Resources has a small group of super smart field mechanics. Kevin Andrews, VP of LaRana, has the mechanics on shift rotations as they’ve found that these compressors can go down any time of day or night. They also purposely bring them down roughly once a month for routine maintenance for either the engine, pipeline, or other mechanical issues. Just this morning, the compressor they dub “Cherry Creek” went down about 5:00 am for a little over an hour. We see the calls go out at all times of day or night and know that someone may have gotten out of bed and are driving to the site. Depending on the problem, the mechanic may be there for a very long shift or even making a mad dash to Denver to pick up parts for the engines.

Alerts also go out to Melissa Gray, business and operations manager at Priority Oil & Gas. Everyone wants to know when these things go down. Melissa has mentioned to me in the past that it can cost $500/hr or more when a compressor goes down.

Natural gas pipeline compressor failure

Other Benefits

Melissa has also seen other benefits by using the Virtjoule system. For starters, since Virtjoule keeps all the runtime history of the machine, she is able to determine the service level of the compressor. Priority Oil & Gas used to own these compressors, but now actually rent them from a supplier. Let’s just say the rental would make your rent or mortgage look small. As such, it’s important to get the most out of these machines and if they’re not performing up to the service level set by the supplier she can get a rebate on the monthly fees. Virtjoule machine history makes our $29.95/mo look really cheap when our service does both fault detection and provides the runtime history for such an important and expensive piece of equipment. Virtjoule never throws away runtime data, no matter how old it is.

Conclusions

Virtjoule’s ability to monitor the vibration of a machine is particularly well suited for industrial gas compression where compression is done not by electric motor, but by a natural gas fired engine. Because Virtjoule is based on a cellular chip in each device, there is no need for an internet connection to be brought in or any need for a telephone line. These compressors are amongst farmland, pasture, and prairie land. They’re in range of a Verizon cell tower, but any other communication method would be thousands of dollars more expensive…one of the reasons Virtjoule is a good fit in oil & gas applications.

[Randy Cox - CEO and co-founder of Virtjoule - He is the software designer and analytics engineer for Virtjoule Sense sensors. He studied Chemical Engineering and Petroleum Refining at the Colorado School of Mines. You may contact Randy at: randy at virtjoule dot com] – See more at: http://blog.virtjoule.com

Jul

Virtjoule Juice Cellular CT Monitors Pool Pumps and Chlorine Generators

Posted by randy at 10:38 am

Summary

Virtjoule Juice Cellular CT’s are ideal for monitoring pool pump and chlorine generation equipment. Using a Virtjoule Juice Cellular CT sensor, you can now monitor amperage of the pool pump and other electrical components critical to the system such as chlorine generators. This is suitable not only for large pool equipment, but it’s affordable enough to monitor smaller apartment complex pools and even residential pools.

Virtjoule Juice can monitor activity of a Turbo Cell chlorine generator

Key Points

Pool pump equipment is often unattended for days. It seems when you need the pool ready for the weekend is when you find out the pump has failed.

The first step in maintaining a healthy pool is keeping the pool pump running. If the pump isn’t running you’ll see that heating, chlorine generation, and water circulation simply stops. What else is there to a pool besides water and people?

Virtjoule Juice Cellular CT’s are uniquely capable of monitoring all the electrical components of a pool plant.

A Small Residential Pool

What you see here is a common residential pool plant setup. There is a gas fired heater, a WhisperFlo pump, Turbo Cell chlorine generator, and a large filter.

A typical residential pool pump setup with chlorine generator

In this setup, a single Virtjoule Juice Cellular CT was installed in the disconnect box for the plant. Activity level of both the pump and the chlorine generator can be monitored with just one sensor.

Below is a routine chart generated by Virtjoule showing the activity of both pump and chlorine generator.

Runtime graph of pool pump and chlorine generator

It’s plain to see on this chart that there is a base level of around 9 amps that is contributed by the pool pump. The label on this pump says it runs from 8.8 to 9.8 amps, right in the range we’re measuring. This particular pump is a WhisperFlo pump.

The chlorine generator is a model called “Turbo Cell”. The Turbo Cell doesn’t continuously generate chlorine. It intermittently runs depending on the level setting on the cell which is determined by the pool maintenance service, how often the pool is used, and whether or not the pool is covered when not in use.

Chlorine Generator Energy Use

People often ask, “How much energy does my chlorine generator use?” If you search the Internet there are a variety of answers that amount to “It depends.” People do want to know since what you’re doing with a chlorine generator is using electricity to break apart NaCL (salt) to temporarily generate chlorine which kills pathogens in the water. Just as quickly, that chlorine recombines forming salt again.

You may have heard of salt water pools. These aren’t done to mimic the ocean. It’s done as a more economical and environmentally friendly way to produce chlorine. The net result is a pool without chlorine and the chlorine smell and burn when you swim in it. Chlorine generators do cost money to run because it takes electricity to run the cell. Now there is an easy way to quantify how much energy your chlorine generator is using and also whether it’s running or not.

In this example, the chlorine generator is running every 2 hrs for 10 minutes taking roughly 2 amps. That’s 120 minutes at 2 amps on 220 voltage. 2 amps on 220 is 440 watts. Times 2 hrs is 0.880 kWh. That costs you roughly $0.10/day in most places in the US. Virtually nothing.

Pool Pump Energy Use

The pool pump, on the other hand, is where all the electricity cost is going. 9 amps, 220 V is 1,980 watts. 24 hrs/day is 47 kWh per day which is between $4-$5 / day or anywhere from $100 – $150 / month to run depending on your part of the country…and that’s a small pump.

Fault Detection

Most pools have a lot of problems that start when the pool pump shuts down. This can happen for a variety of reasons including lightning storms and pool pump failure. Chlorination and circulation cease. Water quality can get cloudy very quickly. Virtjoule Juice can alert you when your equipment isn’t running by either text message, email, or even voice callout.

Conclusion

With just one Virtjoule Juice Cellular CT, it’s possible to watch over two critical pieces of equipment that keep your pool clean and healthy.

Jul

Announcing Virtjoule Juice – World’s first dedicated cellular CT

Posted by randy at 9:10 am

Summary

Virtjoule has been in production and in the field with a new Cellular CT device called Virtjoule Juice. It’s a new continuous monitoring system for amperage readings used on any type of equipment electric motors or other types alternating current draw.

The new Virtjoule Cellular CT has been calibrated with the Magnelab SCT-0750 line of industrial current transducers (CT) making it both accurate and flexible for reading amperages from 0 to 3,000 amps.

With the Virtjoule Cellular CT sensor and monitoring service, you can find problems such as short or long cycles, over or under amperage conditions, and after hours operations, all of which may indicate operational problems which should be corrected before becoming expensive repairs, large utility bills, or interrupting critical business processes.

Monitor Amperage and Power

Current transducers are a special type of sensor that can not only tell if current is flowing through an alternating current wire, it can also tell you how much amperage is flowing through the wire which is useful for both fault detection and energy studies.

First dedicated cellular CT on the market

This is the first low-cost dedicated cellular CT available on the market. Other CT products are available on the market, but they are either unconnected loggers that have to be retrieved and uploaded or are connected to gateway devices that cost well over $1,000 and don’t stand alone.

CT’s are sized and based on the electrical current used on one leg of an alternating current wire. Sizes can go from 5 to 200 amps in the SCT-0750 line and up to 3,000 amps on other parts of the Magnelab CT lineup.

Virtjoule cellular CT with Magnelab 20 amp CT

A Must-Have Tool for Energy Auditors, Retro-Commissioning, and Service Providers

A cellular CT will become a must-have tool. Energy auditors, retro-commissioning specialists, and service providers are all aware of the basic usefulness of a regular handheld CT. 24/7/365 monitoring via CT gives you the ability to isolate a single component of a machine or the flexibility to find the variable energy use of an entire machine.

Virtjoule has been known for excellent fault detection capabilities and using a CT is now one more way to tap into fault detection capability in addition to understanding the energy usage of a machine.

True RMS

The Virtjoule Juice Cellular CT is a True RMS sensor. In addition to normal sinusoidal power feeds, it can be used to accurately measure amperage on variable speed devices and other pulse width modulated machines where computing the True RMS is key to knowing the correct amperage. Knowing correct amperage is critical to knowing how much power a machine is using and, for that matter, what parts of the machine are running.

How can you use a Cellular CT?

Cycling and hours of operation

Use the Cellular CT to pick up runtime information which includes amperage levels, cycling behavior, and hours of operation.

Failing compressors

It’s a well known fact that most aging compressors will begin to draw more amps, not only on startup, but at runtime as well. We have seen a recent example where amperage monitoring showed that a compressor was at the end of its life and we caught the actual failure when it happened. Monitoring amperage can give you insight into machine life expectancy issues.

Locating energy wasters

Use the Cellular CT to do sub-metering on selected units. With power output estimates from known amperage, it’s now possible to find those energy hogging machines.

Use the Cellular CT to demonstrate how much after hours operations cost. By monitoring the machine 24/7, you know when it is running after hours. Because you now have amp information, along with voltage and power factor it’s now possible to closely estimate how much after hours operations are costing. Virtjoule can help you do that.

Monitor complex behavior of individual components

Cellular CTs can be used to monitor specific electrical components of a much more complex machine. For instance, you can know conclusively at any time how many stages of compressors are being used in a large package unit. Also, estimating energy usage on larger and more complex machines can be very inexact because you can only estimate just exactly how the machine is used. Rules-of-thumb break down, particularly on larger machines. Now you can find out exactly how the machine is being used and how it’s performing.

Because amperage can be turned into power information and because we’re taking 10 second averages of amperage, you can get very accurate estimates of power usage on any electrical machine, motor, or other electrical component of a machine.

Multi-tenant situation, expense sharing

Do you have a multi-tenant situation, but sub-metering is not possible? Use Cellular CTs to understand power use across shared machines or electrical input and allocate costs appropriately and fairly.

Comparing a standard “clamp meter” with Virtjoule Juice Cellular CT

Clamp vs Cellular CT

Available now

Virtjoule Juice Cellular CT is available now. Call us today to discuss how Virtjoule Juice Cellular CT can help you or your clients and get your order in for this first of a kind monitoring device.

Aug

Newly installed air conditioners can be the worst energy offenders

Posted by randy at 5:25 am

Summary

Newly installed machines are often not programmed. A client will either replace or add a machine to an existing space and the thermostat that comes with the machine is installed and no one ever programs it. Everyone seems happy because the machine is working well and there is new cooling, but the machine is left running 24/7. It’s as if the thermostat owners manual is tossed in the trash as the installer leaves the premises. For a 10 ton machine running an extra 12 hrs a day, this could be a $3,000 per year oversight.

Takeaways

If you have thermostat controlled machines, know where the thermostat manuals are.

If you can’t find the thermostat manual, note its model number (open the thermostat if necessary) and find the thermostat manual on the web. Most reputable makers have repositories of the original operator manual online. Print it out, use it, and remember where you put it.

Make sure that the unit is programmed before the installer leaves the building or assign someone on your staff to get that unit programmed right away.

The days are gone when you can afford to ignore problems like this. We live in a technology society and there is someone in your organization, young or old, capable of knowing how to do basic thermostat programming. Find them.

Ah…A new machine

Who doesn’t love a new air conditioner? You feel good because you might have just replaced an old with with a new unit with a really high SEER rating. Or you may have finally taken care of a hotspot in a building and bolstered the overall cooling capacity.

It’s not unusual for clients to replace equipment expecting that their utility bill will be lower. Normally you should see a decrease if you’re replacing a similar size machine with a newer, more efficient model. But the devil is in the details if there isn’t enough follow through by you or your HVAC installer.

We’ve seen a few instances this summer where a client has replaced and/or added machines. New Virtjoule HVAC monitors were added and quickly discovered the new machines were working overtime, 24/7. No one had bothered to set the thermostat in automatic mode or to program the occupancy times and setbacks.

It’s easy to be complacent once the new machine is in and working. The complaints stop. Things get back to normal.

Realize that a 10 ton machine might be burning at 7.5 kW. If that machine is running extra time then your utility bill can add up really quickly.

Let’s say that your 10 ton machine is running an extra 12 hrs a day because no one bothered to program it. That machine could be using an extra 90 kWh / day. An extra $8.10 / day at $0.09/kWh. Doesn’t sound like much. Multiply by 365 days/yr = $2,956.50. There’s your $3,000/yr problem.

Will you be able to find someone to spend 15 minutes to program the thermostat to save you $3,000 per year?

Aug

Walk-in Refrigerator Door Left Open

Posted by randy at 10:27 am

Summary

Refrigeration faults aren’t always blamed on the condenser or evaporator. Sometimes there is a perfectly good explanation and a cheap fix. This article shows the effect on cycle times when a walk-in refrigerator door was accidentally left open overnight after a produce delivery.

Key Concepts

Virtjoule detects extended cycle times that can be due to problems with the walk-in cooler box itself.

When door seal gaskets wear out and doors need an adjustment, the unit will have to work harder to make up for heat leaking into the cooler.

The first chart shows the normal cycle pattern for this Heatcraft HyperCore. Cycle times are between 8 and 15 minutes long and the unit cycled 33 times.

Normal walk-in cycling pattern

The next day, in the early morning hours, the unit began a very long 3 hr and 40 minute cycle followed by a 33 minute cycle, a 47 minute cycle, and two 52 minute cycles. By that time the morning crew figured out the cause of the problem and got the door closed.

Extended cycle times when door is open

Virtjoule caught this problem through the use of cycle duration rules and recognizing several extended duration cycles. No crisis call was made to this client because of the early morning time and the fact that the unit was still cycling.

When we see a unit extend cycle times like this, but still cycle off and on, we can safely assume the unit is still meeting demand. It’s obviously taking a lot longer to meet the demand and the unit should be looked at.

The fix in this case was cheap and obvious…close the door. Other situations could be more insidious. Walk-in refrigerator doors are used a lot. Door seals wear out. Hinges sag. Latches break. When door seals start to fail or the door doesn’t fit as well, it can be hard to know.

When you’re monitoring the cycle count and cycle durations, you can see the unit begin to work harder and harder to make up for the cooling loss.

Feb

Thermostat – Friend Or Foe?

Posted by randy at 4:48 pm

Summary

It’s hard to imagine where control of our indoor environments would be without the lowly thermostat. Programmable thermostats have been a mainstay of both commercial and residential heating and cooling. But like the VCR “Blinking Twelve Problem”, the proliferation of thermostat interfaces has caused many to be completely misunderstood. Property owners and tenants ignore the thermostat to their own detriment.

This article discusses a client who was heating and cooling 24/7 on two of their Carrier HVAC units for a retail store in a small strip mall. The cause was mis-programmed Carrier Debonair thermostats. They had no idea. Without the HVAC monitoring capability of Virtjoule, this condition may have gone unnoticed for years.

Upon properly programming the thermostat, the units are now running just 12% of the time they were running before.

Carrier is a big and trusted name in HVAC, without a doubt. But that hasn’t solved the problem of complex programming interfaces for mere mortals. I have a 29+ year career working with computers and the interfaces on most thermostats can still be mind boggling. This store was using a Carrier Debonair thermostat. Within the Debonar lineup there are no less than 10 models to choose from. The model our client had was similar to this one:

Carrier Debonair thermostat

Since this client was very close to where I live, I took the opportunity to go check on it myself. First thing I did was to go on the Internet and find the owners manual for this model. I was able to find it here:

http://www.docs.hvacpartners.com/idc/groups/public/documents/techlit/33cs-7so.pdf

As is often the case, I was expecting that there would be some schedule problem or perhaps a misunderstanding about what happens when the fan mode is set to “On”. After all, we’ve seen that there is no predicting what the blower is actually doing when the heating and cooling mode is set to “Off” or even “Auto”. We’ve seen blowers continue to run when the thermostat mode is set to “Off” and it is thermostat dependent. Very counter intuitive, but who is going to go on the roof or hold a tissue in the air stream to notice if the blower is running after you shut down for the night?

What I found on this thermostat is that it has three different occupancy programming modes. The first and default mode, Occupancy 1, has the unique feature that the occupancy schedule can be overridden by a light sensor. That was a red flag to me as that means the thermostat was depending on yet another sensor which might or might not be working, rather than the building schedule that it also contains. It could also be getting confused by other inputs like street lights through windows and start times would change radically depending on what time of year it was. Why would you want that if the building hours are nearly identical throughout the year?

Without enough time or a good way to determine if the light sensor was working, it was clear to me that a retail establishment like this should not be depending on light to control its space.

Switching the thermostat to “Occupancy 2″ mode allowed it to be controlled by the occupancy schedule. On this thermostat each day has occupancy times and unoccupied times. Luckily there is a feature that allows you to copy one day to the next, but how you do that is completely inscrutable if all you had was your eyes on the thermostat. It took some careful reading of the owners manual to figure that out. By the way, where do you think the owners manual was on site? I don’t know either.

Besides the occupancy mode issue, both thermostats had their time and day set wrong. In the case of this seven day a week establishment, having the day set wrong was harmless. But here in February, both clocks were still set to daylight saving time.

Let’s take a look at the results. During this time the temperatures here in Colorado were temperate with daytime highs in the 40′s and 50′s and lows in the mid-20′s. The first image is a typical daily runtime graph of one of the units before the thermostat change. The blower was running 24/7 punctuated by cycles of heat. There was simply no schedule on this unit and it was always running.

Runtime before thermostat change

The “after” chart is one that is much easier on the eyes. You can see some periodic runtime during the night when the setback temperature of 58 was hit. But by late morning there was no runtime at all. So not only was the tenant saving night time and after hours runtime, the majority of their savings was actually achieved during business hours when outside air and activity in the building was working for them.

Runtime after thermostat change

Across the two units, I was able to compute that total runtime after the thermostat change was just 12% of what it was before. That’s what you call saving money. Without HVAC monitoring of these units, this money would have been wasted month after month.

[Randy Cox - CEO and co-founder of Virtjoule - He is the software designer and analytics engineering for Virtjoule Sense sensors. He studied Chemical Engineering and Petroleum Refining at the Colorado School of Mines. You may contact Randy at: randy at virtjoule dot com]

Jan

Property owner and tenant incur extra 20% HVAC expense on BAS building

Posted by randy at 1:57 pm

Summary

A Trane Tracker controlled office and warehouse building had a rogue schedule, resulting in a HVAC bill that was 20 percent higher than necessary for the tenant. The building had 10 roof-top units (RTU’s), including one CRAC (Computer Room Air Conditioning) unit and nine package units. In addition to running during normal operating hours, all nine RTU’s were showing runtime on Saturdays — when the building was unoccupied. Neither the owner nor tenant were aware of the issue. Simply put, these machines were operating 20 percent longer than needed which not only led to higher HVAC expense for the tenant but, ultimately, would have shortened the lifetime of the equipment, reducing the property owner’s ROI.

Although BAS (Building Automation Systems) can streamline and help control a big facility, mistakes in BAS programming are often compounded — or missed altogether — without monitoring tools. As buildings with BAS get passed from tenant to tenant and HVAC service to HVAC service, valuable knowledge is often lost as to how the BAS and subsequent schedules were set up. It can be tedious and expensive to recommission the building and do a full audit of schedules within the BAS. Until there is a complaint, these sort of problem situations can drag on for months or years, wasting energy and money.

Key Concepts

Mistakes in BAS programming can be magnified across a building system, resulting in higher than necessary expenses, and still not be readily apparent.
Programming of BAS, if properly done, has benefits for both the property owner and the tenant(s).
Many BAS, if not most, lack the kind of internal monitoring controls that can uncover these money-wasting mistakes.
BAS systems are complicated and system knowledge is often lost when new tenants move in or when HVAC services are switched.

The facility we were working on is a typical high tech office building with an attached warehouse area. It has a total of 10 units, one Trane Voyager used as a CRAC unit and 9 other Trane Voyagers of various vintages. All the units that we were monitoring were on the roof of the building.

BAS controlled building running after hours

Two story office, one story warehouse space

Soon after Virtjoule’s HVAC monitoring sensors were installed on the Trane Voyagers, it became apparent that there was a scheduling problem with the building. All the units would start up on Saturday mornings at, but not exactly on, their normal weekday hour. The slightly different start time was the first clue that an extra schedule was in the system. Shutdowns were often at the same time as the weekday schedule for each unit.

This is a professional building with product developers and executives and featuring some manufacturing and warehouse space. Although employees can come in on the weekend, there is no need for the building to be completely heated or cooled for just a few people. Employees do have access to thermostats to regulate heating or cooling if they wish.

The building engineer was alerted and initially puzzled by the situation since he thought the building should be in “unoccupied mode”. He proceeded to work carefully through the BAS system and located the rogue schedule.

Because the rogue Saturday schedules were so similar to the weekday schedule, it’s easy to see that the owners of this facility have now reduced the wear and tear on the equipment they own by one day a week — or 52 days a year! The tenant is saving over 16 percent on their HVAC expenses (having paid 20% more than they should have) and presumably, over a longer period of time, will save on maintenance as the number of calls should decrease.

Jun

It’s 2 am. Do you know what your building automation system is doing?

Posted by randy at 12:56 pm

Summary

If you have a BAS (Building Automation System) you can’t assume that everything is always well. Our experience has shown there can be numerous control problems with BAS systems even when staffed by full time employees (earlier blog article).

In this article we’ll discuss problems with a Trane Tracker BAS used on a small 12,500 ft2 office and retail building in Niwot, CO.

Without the knowledge of the building owner and operator or their HVAC service company, three out of the four Trane Voyager units were running 24 hrs a day multiple days a week, including weekends, when the building was not occupied. The BAS programming interface was obtuse enough that even an experienced HVAC control technician failed to correct the problem on the first trip. The problem was fixed on all four units on the second trip and our monitoring showed that the building owner would save over 5,600 hours of runtime over the course of a year.

This building is managed like many others where the building owner hires an HVAC service provider to provide varying levels of service, primarily to handle complaints and do some routine maintenance a few times a year. The building is not large enough to justify a full time facility manager or maintenance person.

All four package units were the same Trane YCD120B3HCEB, a 10 ton package unit, installed in a small quad on the roof of this building.

After the first few weeks of monitoring, a clear picture emerged that suggested that all the units were running on flawed schedules. Here’s a summary of what we found:

Unit 1:
Sunday through Tuesday – Unit ran 24 hrs each day
Wednesday – Saturday – 5:00 am – 11:00 pm

Unit 2:
Monday – Midnight to 7:30 pm
Tuesday – Friday – 4:30 am – 7:30 pm
Sat – on demand
Sun – 24 hrs midnight to midnight

Unit 3:
Monday – Midnight to 10:00 pm
Tuesday – Friday – Starts ranged from 3:00 am to 5:00 am with stops at 10:00 pm
Saturday – 6:00 am – 7:00 pm
Sunday – 24 hrs midnight to midnight

Unit 4:
Monday – Friday – 4:30 am – 6:30 pm
Saturday – Sunday – 3:00 am – 5:00 am and then on demand

To summarize that list, there were several units running 24 hrs per day for several days, several units with startup and shutdown times well before and after the building was occupied, and unexpected weekend runtime.

Keep in mind that this was a professional building that had some empty suites and the rest were 9-5 offices and a doctor’s office. There was rarely any activity outside of normal business hours.

The byzantine BAS interface

This particular vintage of Trane Tracker BAS had a serial interface to the system. The HVAC technician to had to “jack” into it with his laptop computer and was presented with a command line interface. The building is divided into zones and groups and any particular suite would belong to both a zone and a group.

The HVAC service company for the building had only been in charge for about a year and was never asked to fully commission the building. They had only been at the building a few times and never to fully explore the current BAS programming. This particular BAS was old enough that their experience with it was out of date.

Everything is not as it seems

The technician discovered that there were several zones assigned to multiple groups, almost certainly caused by tenants moving in and out followed by layer upon layer of changes being made to the system. Some of those groups had the obsolete schedules and somewhere along the line a programmer didn’t reconcile what was going on with all the zones and all the groups. Who knows, perhaps someone did notice something amiss, but left it alone assuming the last person knew what they were doing and the problems kept stacking on.

Once we found this nest of issues we were sure that the problem would be fixed. In the command line interface, the technician changed the schedules from things like 03:00 to –:– which was his latest understanding of how to zero out a schedule entry.

With much tedium through this interface, day by day, zone by zone, group by group, the technician dutifully found and “zeroed” out all the offending schedules by putting in –:–. We wrapped up quite sure all was going to be well again. It turned out it wasn’t. Virtjoule was still detecting bad schedules, but this time it was a different set of bad schedules and all four units had the same bad schedule. That was a disappointment, but also a clue.

The technician returned a few days later after conferring with a colleague who used to work at Trane and was an expert in these systems. It was suggested that putting –:– to zero out a schedule entry left the Trane Tracker system assuming that it should continue whatever the last state was. If the last state was that the building was occupied then it would go through the next schedule with the same state. The new schedules were leaving the building in an “occupied” state at the wrong times.

The fix

Since it was not possible to tell this version of the Trane Tracker that a specific day was unoccupied, the technician had to set up very short run times on Saturday and Sunday such that the units would come on, but they would not stay on for very long. Correcting all of the weekday schedules and double checking that the same zone did not belong to multiple groups cleaned up the other issues. It became obvious that the new schedules were in place and correct.

Wrap up

Without the Virtjoule monitoring system, the schedule flaws programmed into this BAS would have gone unnoticed for years. No one really knows how long it had been like that. Left unchecked this could take years off the life of the equipment not to mention the extra utility expenses most often passed on to the tenants.

Even after a trained technician made changes, things were still not right. Without the monitoring capability to actually know the machine was running, there would have been little resolve or patience to notice that the service call didn’t actually fix the problem. Virtjoule not only found the problem, but it was able to verify that the problem hadn’t been fixed initially and was fixed on the second trip.

Feb

Announcing economizer opportunity metrics

Posted by randy at 8:52 pm

Summary
An economizer is a set of controls and dampers, usually fitted to a rooftop package unit, that can allow variable amounts of outside air into the system to supply cooling when outside air temperatures allow. This is an energy saving tactic to take advantage of “free cooling” by using cool outside air rather than running compressors and the refrigeration circuit.

With Virtjoule’s economizer opportunity metrics you can discover the following issues and opportunities:
– Identify units with economizers running refrigeration circuits when they could be cooling from outside air instead
– Compute the potential savings for an existing unit to have an economizer installed or be replaced with a unit that has a built-in economizer
– Find the total amount of time the outside environment spent below a certain temperature threshold.

An economizer is one of the most misunderstood HVAC components there is. They can be complex, prone to failure, and controlled improperly. What starts off to be an energy and money saving device can turn into an achilles heel of extra equipment expense, maintenance costs, and lost energy saving opportunities.

In an excellent case study recently published by the Western Cooling Efficiency Center (WCEC) at UC Davis, Davis, CA, economizer faults made up two of the top 10 most common HVAC faults.

http://wcec.ucdavis.edu/sandbox/search/ResearchBriefsPDF/Case%20Study_FaultDetectionDiagnostics.pdf

We especially like this paper because the Virtjoule sensor got a front page picture and we were included in a feature roundup of fault detection products. I would encourage you to take a look at the paper where Kristin Heinemeier reviews the state of standards development for fault detection, particularly as it relates to California Title 24 initiatives.

According to the WCEC report, the two main economizer failures are incorrect or sub-optimal set point and economizer damper failure.

The WCEC and CA Title 24 recommend a 75 degree set point. That means that when outside air temperature is 75 degrees or less, the dampers for the economizer should be coming open.

There is a huge range of capability with economizers. Some are completely manual where a building technician will manually adjust the outside air damper to a certain degree. But to be useful, the damper position should be able to be automatically set by economizer controls.

A 75 degree set point makes sense in the dry climate of California, but there is this thing called enthalpy that affects how much good you can get out of air with a certain humidity and temperature. It turns out that the usefulness of cool outside air goes down when the humidity of that air goes up. That’s because enthalpy, or the total amount of heat in that air, is higher at higher humidities.

Because humidity is a measure of how much water is in gaseous form, there is extra heat content in air that has higher humidity due to the latent heat required to hold water in a gaseous form. That heat is in addition to the sensible heat, the heat you can feel, in the air, raising the total enthalpy and heat content of the air making it harder to cool. There is no linear relationship between temperature, humidity, and enthalpy. It has to be experimentally determined and is the focus of the engineering field of psychrometrics (not psychometrics…another topic, another blog).

In many parts of the country, a useful set point temperature for the economizer would have to be much lower, meaning that the outside air temperature would have to be much lower before you can take advantage of the economizer. For warm and humid climates, an economizer may never make sense to install.

Ultimately, a good economizer is one that can select an air stream, or mix an airstream, with the lowest enthalpy, the lowest total heat, so the air going in will take less energy to cool to the desired temperature. A good economizer will measure the enthalpy of the return air as well as the outside air by using temperature and humidity sensors. A combination of temperature and humidity sensors are required to compute enthalpy and faulty temperature and humidity sensors are a common cause for improper economizer function.

Example

An economizer failure will show up in a couple of ways in run time statistics that Virtjoule provides. If the economizer is stuck closed or the set point is set too low, then Virtjoule will see compressor run time at temperatures below the desired set point. If the economizer is stuck open, then on warm days the run time of the unit will be longer than normal for a given condition because warmer air than is called for is being fed into the RTU. Both ends of the failure spectrum can be noticed. What is more subtle, and something we are not chasing at this time, is figuring out whether the compressors and economizer are working together to create an optimal mix and the subtle degradations that might be shown with more specific gauges.

Let’s look at an example. The following graph shows compressor cycles for a 40 ton McQuay on an executive office building in San Diego. The accompanying temperature chart shows the temperature trend throughout the day.

As you can see, a significant amount of compressor time was used for cooling for several hours between 9:00 am and noon and again after 3 pm. And that’s time when it was below 55 degrees. Humidity levels were 45-55% during that time. Most HVAC people will tell you that humidity doesn’t make much difference at temperatures below 55 degrees and so computing enthalpy is a waste of time…just use the air less than 55 degrees to cool all you can.

These McQuays are fitted with economizer capability. It looks to us like it’s not being used or controlled incorrectly as outside air temperature peaked at 57 degrees and most of the day was at 55 degrees or below. There should have been ample cool air to supply cooling for most of the day. Given that California recommendations are to use the economizer up to 75 degrees, minimizing compressor time, it looks like a lost energy and money saving opportunity.

In tables in the Virtjoule web application, we’ve tallied 6 hrs 16 minutes of compressor time on this unit this week. It hasn’t been too warm in San Diego this week. But we’ve also tallied 5 hrs and 18 minutes of compressor time when the temperature was below 55 degrees. 85% of the compressor run time this week has been at the same time outside air temperature was 55 degrees or less. It’s almost certain the economizer is not paying for itself and could use a call to check the set point, temperature and humidity sensors, and the physical operation of the dampers at various levels.

Summary

In summary, new analytics functionality in Virtjoule makes it possible to identify lost opportunities to save money and energy for machines with economizers. It’s also possible to use this same functionality to evaluate the potential benefit if retrofitting an economizer or replacing a machine by monitoring the cooling behavior of a machine when temperatures are less than 55 degrees.

Feb

HVAC and refrigeration monitoring projects

Posted by randy at 11:48 am

Summary:

As a young company, we’re still learning the range of things that are possible with the Virtjoule solution. In the last year and a half, we’ve discovered that not only has the technology met our original vision of functionality and price, the types of projects that can be accomplished are even more varied than we thought. The Virtjoule solution has been installed on everything from beverage coolers to a 6,000 ton cooling system on a major Las Vegas hotel.

We often get asked, “What kind of projects can you do?” In the earlier days of the company, the answer was in the form of suggestions of what we ought to be able to do. After all, we had a device that could pick up vibration and the ability to track that vibration over time and see patterns emerge.

These days, the answer is educated by field experience with the problems and the economics of solving those problems for our customers. At the highest level, the answer is “If it vibrates, we can monitor it.” As a company, we’ve done work in the various segments you’ll see in the list below.

To get a handle on the range of applications, perhaps it’s useful to take a scan across the types of units that we’re currently tracking.

Function	Model
Beverage Coolers	Micro Matic MMPP4301
Walk-in refrigeration	Numerous models of Harford Duracool Harford Duracool – H1984A8 Trenton – TEHA025L6-HS2A-F – 2.5 hp Trenton – TEHA006E6-HS2B-B – .6 hp
Air conditioning – Split systems	Ducane – 2AC13L60P – 2A – 5 ton
Package units	Trane YSC048A3EMA2U – 4 ton Trane YSC060A3EMA3 – 5 ton Rheem RKKA-A073CL10E – 6 ton Carrier – 48TCEA07A2A5A0A0A0 Bard PH13062 Bard PH1060-B – 4.6 ton York D1NA042N05625C – 3.5 ton York D1EB048A25B – 4 ton York D2NA060N09025D – 5 ton York DM090N10A2AAA4B – 7.5 ton York BP090C00A2AAA4A – 7.5 ton York D1EB060A25B – 5 ton Johnson Controls – J05ZJN10D2AAA1C – 5 ton Johnson Controls – J07ZHN15P2AAA4B – 7.5 ton Johnson Controls – J10ZHN20S2AAA4B – 10 ton Johnson Controls – J25ZJN32S2BZZ10001 – 25 ton York ECO2 50 ton Carrier 48P5 Horizontal VAV 100 ton McQuay 40 ton McQuay 70 ton
Heat pump packages and split systems	Heil Tempstar – NHP060AKC1 – 5 ton Coleman/York – HP060X1021A, 5 ton split system Goodman CKL60-1 – 4.75 ton Goodman CPLE60-1 5 ton Goodman GPH1324M41AB 2 ton York 20 ton
CRAC-(Computer room air conditioners)	Liebert
Evaporative Coolers – Water pumps	Can’t divulge manufacturer at this time
Air handlers	Goodman AH120-00
Commercial Refrigeration Compressors	Copeland Copelametic 4RA3-100A-TSK-800 Copeland Copelametic 6RA4-200A-TSK-800 Copeland Discus 3DB3F33KE-TFC-800 Copeland Discus 4DL3F63KE-TSK-800
Chillers	York – 18 ton Carrier 30GTN090 90 ton
Cooling towers	Baltimore Air Coil, 125 hp electric pump and 100 hp VFD fan – 2,000 ton

One of the most satisfying things that we’ve seen is that this is a huge range of equipment, all being monitored by the same Virtoule Sense technology.

If you’re familiar with some of the pieces on this list, you’ll know that some put off so much noise and vibration that you need ear protection. Other pieces are so quiet and smooth that it’s very difficult to tell if they’re running when standing right next to them or even putting your hand on the machine. I’ve been particularly impressed with the Trane series (YSC060A3EMA3) as very smooth operators in package units.

Because the Virtjoule sensor is self-calibrating, once it’s installed it can start off as a very sensitive device and self calibrate to the point where it can handle large magnitude vibrations. This self-calibration means that the installation is roughly the same for all of these installations, stick it on, power it up, and start getting data. An operating signature will always emerge that can be analyzed, reported on, and alerted on.

Customer and facility types

The table above is specific to equipment and equipment applications. But what sort of customers and markets does this represent? Here’s one look at the type of facility and customer:

Executive office buildings
Restaurants
Strip malls
Small market and convenience store refrigeration
Data centers
Hotels

Some are very high energy users where control problems or equipment degradation can cost thousands of dollars a year if the problem isn’t corrected. Some have equipment that might not burn a lot of energy, but the unit is serving high value contents. Food, pharmaceuticals, and computers are good examples of high value contents that need to be protected from catastrophic loss.

In a recent example, Virtjoule was able to give the owner of a walk-in refrigerator days of notice that their unit was degrading. The customer was able to get a refrigerant leak fixed before they lost control of the temperature of the cooler. One of my upcoming blog articles will discuss this “catch”.

If I were to sum up our results since we started, I would say we’ve helped customers save money by identifying control problems or system degradation that were racking up extended hours and energy bills. And we’ve helped some customers avoid serious loss of high value contents that were being refrigerated.

What new applications will we see this year? I can’t wait to find out.

[Randy Cox - CEO and co-founder of Virtjoule - He has been the software designer and analytics engineering for Virtjoule Sense sensors. He studied Chemical Engineering and Petroleum Refining at the Colorado School of Mines. You may contact Randy at: randy at virtjoule dot com]

Oct

Are your compressors excessively cycling?

Posted by randy at 2:38 am

Summary:

Is your package unit or split system cycling too much? How do you know? The correct answer is that it depends on the unit and the manufacturer. However, common sense can play a huge role in figuring out if your machines are excessively cycling. Finding out how often your machine is cycling and how long the cycle times are can tell you a lot about how healthy your machine is or whether you need to change your control regime.

In this article we’ll look at a case study of a 90 ton chiller from Carrier and how we helped a customer cut over 14,000 cycles per year in normal operation even when the chiller was operated on a building automation system (BAS) with a dedicated maintenance staff.

Key concepts: Excessive cycling and compressor short cycling can be controlled. Avoid excessive wear and tear on HVAC equipment. Stop excessive HVAC energy consumption and expense.

Let’s set the scene. This 90 ton Carrier chiller has normal operating hours of 6 am to 6 pm. We could tell that its start up and shut down times were programmed correctly because it’s obvious from the Virtjoule beat chart below that the unit is running all the time between those hours. The building was not occupied outside of 6 am to 6 pm and the owners of the building did not expect to see any unit operations in the off hours. However, you can see from the graph below that even though there was a noticeable shutdown, the unit continued to cycle on and off throughout the night and early morning hours.

The extra cycles were typically 3-4 minutes in duration and numbered 40 or more per day and many more than that on weekends adding up to over 14,000 cycles per year of extraneous cycling and run time. That’s 14,000+ cycles and over 700 hours of extra run time not to mention that electric motors can take up to three times the amount of electricity to start them than it takes to run them. At common electricity rates all of this could add up to around $5,000 per year not to mention the wear and tear on a very expensive asset.

Here’s a snapshot of the run time graph for a typical day with out-of-hours cycling. You can see out-of-hours cycling through the early morning hours up to 6:00 am and then solid operation between the hours of 6 am and 6 pm. Out-of-hours cycling begins again at 6 pm and continues through midnight on this chart. The excessive cycling continues until 6 am the following day.

Excessive cycles and out of hours operation on a 90 ton Carrier chiller

The following graph is how the machine operated on Saturdays and Sundays. Two out of the seven days of the week had close to 90 extraneous short cycles.

Extra cycles and runtime on weekends when there should be none

Keep in mind that this was the main cooling unit and was operated on a building automation system. Soon after this run time behavior was noted the building engineers were able to make control adjustments that completely eliminated the extra cycles. Now you can see a very clean start up and shutdown of this chiller each day. No extra cycles. No wasted energy. No unnecessary wear and tear.

Control problem fixed for out-of-hours operations and excessive cycles

If you’ve operated BAS before you are probably aware of how much work it can be to extract and analyze the data points that are available. We often hear that the BAS should catch these kinds of problems, but case after case has shown us that it isn’t happening. BAS has proven many times that it’s better at control than monitoring. Even when it’s used for monitoring it can cost hundreds of dollars per data point to extract and then someone has to interpret and monitor the results regularly. Maintenance organizations often have more urgent needs to attend to in their building and this sort of problem doesn’t usually cause immediate comfort problems in the building.

The steady burning of electricity and asset wear should make for a foul smell of burning money to someone in the building and so this should be a comfort problem under someone’s seat eventually. The top maintenance organizations that we see deal with the issues of comfort, maintenance, energy conservation, and cost every day in their operations. They like things to run well and to cost the least amount possible. Fortunately those things usually go hand in hand. With Virtjoule, after a 1 hr installation and setup, a few days later we had enough information to show that a change was needed. The owners of the building were able to get their maintenance organization to make the changes and make an immediate difference on the healthy operation of this unit.

[Randy Cox - CTO and VP of Software Engineering, Virtjoule - is the software designer and analytics engineering for Virtjoule Sense sensors. You may contact Randy at: randy at virtjoule dot com]

Virtjoule installation on Carrier 90 ton chiller

Oct

HVAC Sensor Reporting Intervals – Short Cycling Example

Posted by landon at 11:26 am

Summary: We’re often asked why Virtjoule HVAC vibration sensors report on a one-second interval. Why not five minutes or even longer? This article will cover an actual fault case we discovered on a customer site and demonstrate why a one-second update interval is beneficial for diagnosing HVAC problems such as short-cycling.

Key concepts: HVAC vibration analysis techniques, HVAC sensor reporting intervals, signal aliasing, HVAC short-cycling.

It’s fairly common to find short-cycling HVAC units, especially after we first install the Virtjoule sensor system. They’re easy to see visually in the Virtjoule web application by looking at the sparklines for the building’s HVAC sensors. The following is a typical sparkline of a short-cycling HVAC unit – it’s almost solidly filled with lines going from off to on and back to off:

Short cycling HVAC sparkline

Graph of Short cycling HVAC Unit

A more typical HVAC cycling pattern is shown in this sparkline:

Normal cycling HVAC

Graph of regular cycling HVAC Unit

The reason I’m using sparklines above is mainly to illustrate that the short cycling problem stands out from the visual contrast of these two graphs without regard to the actual timing of the cycle. However, the reason it can stand out in such a simple way is because of a fairly rapid reporting interval by the sensors.

Here’s the short-cycling graph in a bit more detail – this represents about a 12 hour period of time.

24 hour short cycling HVAC waveform

Zooming in more closely, you can see the actual short-cycling wave form begin to take shape along with more detail with respect to the period of short cycling.

Closeup 1 short cycling

Another zoom level deeper:

Closeup HVAC short-cycling

And finally, a zoom level of the short-cycling waveform focusing on a single period or cycle of the HVAC unit. From this, it’s easy to see the entire short cycling period is about two minutes. Moreover, you can see the “click” of the HVAC unit attempting to turn on the condenser fan at about mid-way through the 10:20 mark. The fan failed to start and a bit later, the compressors kicked on and shortly turned off, potentially due to a high head-pressure fault in the system which protects itself by shutting down the compressor.

Closeup 3 HVAC short cycling

Virtjoule-Sense sensors sample the vibrations from the HVAC unit nearly 10,000 times per second and report the average magnitude of the data over that one-second period. If we sampled data once every minute or two minutes, we could easily miss short-cycling events such as those captured above.

A sampling period of 2 minutes could leave you believing the unit is running non-stop if the sample period happened to fall on the regular peak shown in this real-world example. Or alternatively, a 2-minute sample could leave you believing the unit never ran at all if it happened to sample on the interval when the unit was not running between peaks.

Finally, the world is never so punctual, so the more likely scenario of sampling on a 2 minute interval is that you would see a mix of highs and lows which would show inaccurate, sporadic run time.

Using a one-second reporting interval, it’s possible to quickly capture subtle state changes in HVAC equipment and also form a very accurate picture of what’s happening within the unit.

Each HVAC unit will have a somewhat different acoustic signature when you get down to the small details of the waveform, but the overall picture of HVAC short-cycling, built on one second data, becomes very clear.

[Landon Cox - VP of Embedded Engineering, Virtjoule - is the hardware designer for Virtjoule Sense sensors. You may contact Landon at: landon at virtjoule dot com]