Retrocommissioning Findings: Reducing Boiler Purge Cycle Losses

Greetings after another break in the action in terms of my posting rate.  As those of you who follow this blog know, sometimes, I get pretty busy and since writing this is not exactly my “day job’, my posting rate suffers.

Having said that, this post started out as an e-mail the students taking the current Existing Building Commissioning workshop class at the Pacific Energy Center and was intended to provide them with some follow-up information for the lab sessions we are currently working through.  Then, I realized I should share the same information with a different lab class I am involved with.  That made me realize that the information may generally be of interest, so I decided to spend the time on a blog post and point the students to it for the information I was trying to share.

The links below will jump you to the indicated topic.  The “Back to Contents” link at the end of each section will bring you back here.

If you read through this, you will discover that even though the answer is based on a fairly simple equation, obtaining the correct answer requires that the equation be applied properly and there  are a myriad of  details that come up when when you think through exactly what needs to be done.

This is typically the case with most energy calculations.  Thus, while illustrating the specific steps in a specific calculation process, this  post also  illustrates in the general case, what you need to consider as you develop an energy calculation and how to deal with the issues that will come up.  Thus, it may be useful in that  context if you are contemplating how to develop an energy calculation, even if it is not one about boiler purge losses.


Developing the post caused me to realize I needed to do a separate post on Aliasing, so I just put that up to support this.  All of that means these posts are out of sequence in the context of the series I was working on that was focusing on economizer analysis via scatter plots.   But I will be returning to that soon, I hope.  Meanwhile, in this post, we will look at a potential “low cost/no cost” improvement that might be made by fine tuning the sequencing of the boilers in an existing building heating hot water system.

Back to Contents

Background Information

One of the classes I teach is using a hotel in Columbus Ohio as a living lab to teach Directors of Engineering, Chief Engineers, and Engineering technicians retrocommissioning field and analysis techniques.   The class is very similar to the Pacific Energy Center Existing Building Commissioning Workshop series but is delivered in a concentrated, three week form.

The facility in question is a 485,000 square foot, 408 guest room, 22 story high-rise hotel that was originally built in the 1960’s and then went through a renovation cycle in 1996 during which the current mechanical systems were installed.  One of those systems is a heating hot water central plant that serves the building’s preheat, reheat, fan coil, finned tube radiation, and domestic hot water loads.  Here is a draft of the system diagram I have been working on to give you a sense of what the system looks like.



The diagram was created in PowerPoint and this link takes you to the PowerPoint file in case you want to look at it in detail or grab some of the symbols for your own system diagrams.

Back to Contents

Utility Consumption Patterns; My First Clue of an Opportunity

Even before I had gone  on site, I was planning to spend some of my field time and logging capabilities on the heating hot water system.    One reason was simply the age of the equipment;  older equipment can often benefit from some fine tuning if nobody has been paying attention to it for a while.

The obvious tuning opportunity is making sure that the combustion efficiency of the boilers is optimized. A less obvious tuning opportunity is related to minimizing boiler cycling, which will minimize the parasitic  losses associated with the pre-fire and post-fire purge that occurs for most large boilers.  More on that to follow.

My point here is that tuning opportunities are perfect RCx opportunities in terms of bang for the buck.  So one of the items on my list for field investigation was to see if I could determine how well the boilers where tuned.

Another reason that I  wanted to focus on the system was because it was the primary consumer of gas on the site.  The average daily consumption analysis I performed with the California Commissioning Collaborative Utility Consumption Analysis Tool (UCAT) revealed a relatively high baseline during the summer months.  Those are months when you would not expect a system to need to deliver much actual heating energy (i.e. offset losses through the envelope).


Notice how the average consumption in the summer months is about 40-50% of the peak consumption in the winter months, even though the need for heat, as indicated by the heating degree day data does not exist.   Note also that the consumption is independent of occupancy, something that is not always true for a hotel due to the domestic hot water loads.

In addition, the thermal benchmark was not particularly good relative to other lodging type facilities in the United States.


If you are wondering where the data in the chart came from, I generated it using the Building Performance Database, which is a free, online tool.

Back to Contents

Equipment Room Sounds, An On-site Clue

When I went into the facility for the first time, the Chief engineer gave me a quick orientation tour of the mechanical spaces and then left me to further explore them while he ran off to respond to a radio call for assistance with an HVAC problem in one of the meeting rooms.

The central plants were located on the 22nd floor, so I headed up there and started working my way through the mechanical room, observing the condition of things, documenting nameplate data, and taking pictures of things that captured my attention.  At this point in the scoping process, I am typically just “following my nose” and getting a sense of the plant and how it is configured.

That means that when I start, I may not have any particular focus and am just letting the clues I see lead me to things.  For most of us, the things we are seeing probably catch our eye more than the sounds we are hearing.  But the background noises in a mechanical room can also lead you to important information once they penetrate your consciousness.

Such was the case for me as I worked through the mechanical room;  at one point, it dawned on my that I was hearing a repetitive sound of some sort, not particularly loud or alarming, but definitely some sort of change in pitch, almost a shriek of sorts, that had a distinct pattern to it.

You may be able to get a sense of what that might be like by playing the video clip below while reading further into the post rather than watching it.   (I’ll send you back to watch the video clip later in the post to support the discussion at that point.)

But for now, by playing it while reading on, you will have the equipment room noise in the background but your focus will be on what you are reading, which may give you a sense of the mindset I was in at the time I realized that the sound was a clue.

The “Light Bulb” Comes On

Right before the light-bulb came on, I had discovered the door to the cooling tower area and gone out to see them.  As I walked back into the mechanical room, I happened to open the door just as the pitch of the sound that I mentioned above changed.

It could have been the contrast between the sound level outside (fairly quite with only the splash of water over the tower fill) and the general noise level in the equipment room that made it penetrate my consciousness, but what ever the reason, I found myself thinking:

What in the heck was that?  Now that I think about it, that has been going on the entire time I have been in this room.

So, off I went to find it.  That was a bit challenging at first because the shriek was of fairly short duration and there were breaks of a minute or more between the sound events.  In fact, I realized that while there was a pattern to it, the pattern was not a regular pattern.  Specifically, there would be a number of closely spaced shrieks, then a long break and then a single shriek, then a long break and the pattern would repeat.

The Discovery


  • The boilers were in my line of sight and
  • Because boilers will cycle at part load and
  • Because it was a mild day and thus, a likely part load day,

… I decided to take a look at them to see what they were doing.

As I walked up, it seemed like they were not doing much of anything.  But then my curiosity was rewarded by a click, then the sound of the combustion air fan starting, followed by the shriek I had been hearing.   I had found my mystery sound;  here is what I was looking at when I made my discovery.


Specifically, the actuator to the left of the air intake box varies the flow of combustion air and gas via a linkage system that moves an air damper and the gas valve as the actuator rotates in response to the firing control system.

At certain point in the cycle for the boilers in this facility, towards the low fire end of the capacity range, the air flow through the damper system creates a shriek that goes away as the dampers open towards high fire.  That shriek was my mystery sound.

The Pace Picks Up

Having found what appeared to be the source of the mystery sound, I decided to watch the boiler go through its firing cycle to see what I could learn.

I had little time to rest on my laurels as they say. Within moments of my initial discovery, I heard another click, another fan start and another shriek;  a second boiler had staged on with-in moments of the start-up of the lead boiler

In fact, the lead boiler had not even started to fire yet, which implied the second boiler may be starting earlier than it needed to, which implied a potential low-cost/no-cost improvement.  More on that in a moment.

This  sudden and rapid turn of events lead to a bunch of frantic note taking and IPhone stop-watch use on my part as I frantically ran back and forth between the boilers trying to capture what was going on.

Ultimately, I realized that a lot of the events were controlled by set points in the burner firing controller, which led to me shoot the video I included earlier in the post.  The idea was that I could document those events via the information in the controller display panel and then combine that with my field notes to come up with the boiler firing cycle.

Back to Contents

The Boiler Operating Sequence;  Evidence of an Opportunity

By working with all of my field notes and the video, I was able to construct the firing pattern for the boilers in the central plant at the time of my site visit, which is illustrated below.


The graph illustrates why I took the time to document the boiler cycling and interactions;  it turns out that I was on to something.

Specifically, the East Boiler was short cycling.   If you study the timeline above, you will notice that the West Boiler starts with-in a minute of the time that the East Boiler starts.  And, with-in seconds of the time that the West Boiler ignites its burner, the East Boiler cycles back off again.

That means that the East Boiler spends more time in its purge cycle than it does generating heat.  The fact that the West Boiler eventually cycles off and is off for a period of time indicates that the current load condition is less than the capacity of one boiler. In other words, it should be possible to meet the load for the current conditions with out even starting a second boiler.

There-in lies the savings opportunity.  To understand that, you need to understand a bit about what is really going on during the firing cycle for typical boiler like the ones I was working with in Columbus.

A Typical Boiler Firing Cycle

If you study the graphic above, you will notice that there are 8 steps that the boiler goes through in a typical cycle.

  1. When the system supply temperature drops below set point, start the combustion air fan and modulate to full air flow with out lighting the burner.
  2. Hold the full airflow rate for some predetermined period of time – in this case, 1 minute – to ensure that any unburned gas that might be present due to a leaking gas valve or residual gas from the previous cycle is removed (a.k.a purged) from the combustion chamber.
  3. Modulate back to minimum air flow.
  4. Light the pilot at minimum air flow and verify ignition.
  5. Light the burner at minimum air flow and verify ignition.
  6. Modulate the gas and air flow as required to maintain good combustion efficiency while matching the load requirements.
  7. If the minimum firing rate exceeds the demand on the system, then shut down the burner.
  8. Keep the combustion air fan running for some predetermined period of time; in this case, 14 seconds, to purge any unburned fuel and the products of combustion from the boiler.

The energy savings opportunity lies in the context of minimizing the number of purge cycles.

Taking a Look at the Inside of a Water Tube Boiler

As it turned out, one of the boilers at the hotel I was working at had a boiler opened up.  So I took some pictures which will let you see what the inside of a water tube boiler looks like if you have never seen one before.


During a purge cycle, the boiler combustion air fan is basically pumping equipment room air through a heat exchanger that is full of water at the system supply temperature (180 – 190°F in this case) on the other side of the heat transfer surface.  As a result, instead of warming up the water (which is what happens when you burn gas along with the combustion air), the water warms up the air.

That means that during the purge cycle, you are taking relatively cool air from inside the building (which came from outside the building at what ever the ambient temperature was), then you heat it in the boiler and dump it back outside the building.  In other words, the purge cycle removes energy from the water in the boiler vs. what normally happens (the boiler adds energy to the water).

The Energy Conservation Angle

It is important to recognize that purge cycles are very important.

Bottom line, they prevent boiler explosions due to lighting a flame with a combustion chamber full of a combustible, explosive mixture instead of air.  The following image, illustrates what can happen if you do that and was is provided courtesy of Metropolitan Engineering Consulting and Forensics.

The round thing in the picture above is what is left of the boiler after it tried to ignite its burner with a combustion chamber full of fuel and air at the appropriate mix for ignition.  Previously, there was a building surrounding it.  The building is the debris field lying around the remains of the boiler.

I think we can all agree that this sort of thing should be avoided if possible.

And that is exactly what the purge cycle does for us.  Specifically, it blows air through the boiler to make sure that when the burner is ignited, there is no residual fuel from the previous cycle or a leaking valve sitting inside the combustion chamber.

But if:

  • The boiler is oversized, and as a result, cycles more frequently than it would if it could come online and match the load with out cycling off, or
  • If a second boiler, which is not really needed to match the load, comes on line and then drops back off after only operating briefly,

… then the system is spending more time purging than would be required.

In the current example, we have a boiler cycling on when it is not required.  That means that all of the time it is spending on unnecessary the purges cycle is throwing energy away needlessly as compared to what would happen if the boiler staging was adjusted to only bring on a second boiler when it was required to meet a load that could not be handled by the lead boiler operating at full fire.

So bottom line, purge cycles are a desirable thing in terms of operating a boiler safely (see picture above).  But, if your boilers are cycling more frequently than they might need to, then those extra purge cycles represent energy that could be saved if you could reduce the cycles to the minimum number required by the load profile.

Back to Contents

Purge Cycles and Low Cost/No Cost Improvements

If a boiler is short cycling because it it too large for the load it serves, then you probably are not going to be able to make low cost/now cost adjustments to reduce the cycle rate.  Rather, you are probably going to need to add a smaller boiler to the plant or create a thermal flywheel.

I should point out that sometimes, you can create a flywheel by leveraging the thermal capacity of the existing piping network vs. having to add a physical piece of hardware like a flywheel tank.  So if you can leverage the inherent flywheel in the system, then you have a relatively low cost approach open to you.

If you want an example of that, you can take a look at the presentation I did for NCBC in 2013, along with the related blog post.  The example was for a chilled water system but the same concepts would apply to a hot water system.

Having said that, for the hotel I was looking at, it was pretty clear that we had a boiler coming on when it did not need to come on, meaning purge cycles were happening that could be eliminated if we could improve the boiler staging.

The pattern I was observing occurred on a day when the outdoor air temperature ranged from 50-60°F.


The graph above is courtesy of the Weather Underground web site, which is a great resource if you are trying to look back at what was going on for the day you were on site at a particular location.

If you take a look at the bin weather data for Columbus Ohio and assume that at a minimum, the cycling frequency would be the same if it was warmer outside, or maybe even increase, then the potential significance of the unnecessary purge cycles starts to become apparent.



The climate data suggests that “tweaking” what ever was controlling the boiler staging might eliminate the unnecessary cycle from the East Boiler for about 56% of the hours in the year.

Back to Contents

Applying a Data Logger to Assess the Boiler Operating Cycle Over Time

In order to broaden my perspective, I decided to have one of our team members deploy a data logger to monitor the flue temperature for one of the boilers.  My theory was that the flue temperature would provide some insight into the various stages in the firing cycle in addition to providing insight into the cycle frequency.  But the trick there is to use a really fast logging interval.

The reason for the fast logging interval is that in the context of a boiler firing cycle, things can happen really fast.  For instance, if you study the graphic that I presented earlier, you will discover that some of the events have durations of 10-15 seconds. So, if I wanted to capture them, I needed to log faster than those events, otherwise, my data set could be compromised by aliasing.

It was at about this point in the development of this post that I realized I needed to do a separate post on aliasing to keep the current post from becoming even longer than it is.  So I will refer you to that post if you need to understand the details behind how I selected a logging interval for the boiler that is the focus of this post.

The graph below is the raw data from the logger that was deployed to monitor temperature in the boiler flue several weeks after I was on site.


Here is the outdoor temperature data for the same time period.


Clearly, the boiler cycling rate is related to the outdoor temperature.  But before I proceed, I need to stop for a minute to pay tribute to the logger that provided the flue temperature data for us.

Took a Lickin’ and Still Kept On Tickin’

When we deployed the logger, which was strapped to the flue of a boiler that was 22 stories up on a high rise in Columbus, Ohio, we did not fully appreciate how hot the outside of the flue would get.   Our initial installation looked like this and included placing the logger inside a plastic bag to protect it from the weather.


But when we returned to the site several months later, we found the logger blowing around in the 14°F breeze like a little kite, hanging on for dear life to its thermocouple lead which was stuck in the flue and  acting as the kite string.

My guess is that situation evolved when the tape that had anchored the plastic bag and logger to the flue  let go and the bag and logger were directly exposed to the effluent from the flow as the wind blew them around.   In hindsight, that probably happened pretty early on.  So, we figured the little logger was “toast” as they say, if for no other reason that it’s appearance.



Note that this logger (an Onset UX100-14M with a Type K Thermocouple input) should not look like a toasted marsh-mellow (first picture) and that it should be flat, not curved (second picture).

But, when we grabbed it to remove it and hit the little buttons, the little LCD came on.  And sure enough, when we plugged it in, it had data, and the data seemed to be viable and told us what we needed to know for our calculation.  So, as the old Timex commercials would say, it took a lickin and still kept on tickin.  In general, that has been my experience with the Onset product line.  But this really stands out.

I will retire it since I am not sure how reliable it is at this point after all the trauma (I would be totally out of the picture if I went through what it did).  Plus, I am not sure what all of that did to the cold junction compensation, etc.   But I will not throw it out.  It certainly earned the right to relax on the shelf in my office after giving so valiantly to the energy saving cause.

Taking A Closer Look at the Boiler Cycles

If I export the raw data in the previous boiler flue temperature graph to Excel and focus on what the boiler cycle looks like during relatively warm weather, similar to the weather when I made my initial observation, you get a pattern like this.


If I focus in on one cycle, I discover that I can see all of the steps in the cycle via their signature in the flue gas temperature profile.


The graph above is just a different way of looking at the data in the bar graph earlier in the post where I discuss the burner operating sequence. Either way, the data behind the graphics allowed me to quantify the energy going up the flue during the purge cycle.

Back to Contents

Flue Gas Temperature;  Another Clue Pointing to Savings Potential

Before I get into the purge cycle losses, it is worth pointing out that the flue gas temperature profile provided another clue into a potential optimization opportunity.  Specifically, the flue gas temperature is an indication of the efficiency of the combustion process.

To really nail it down, you need to know the percentage of oxygen or carbon monoxide in the flue gas.   But there definitely is a relationship, as can be seen from this table that was extracted from the Department of Energy tip sheet on optimizing combustion efficiency.


The idea behind the table is that if you know the % excess air or % oxygen in the flue gas (most burners target 10% – 15% excess air as a good compromise between peak efficiency and safe operation) and you know the temperature rise across the boiler flue (the difference between the incoming combustion air temperature and the outgoing flue gas temperature), then you can estimate the boiler efficiency.

In a field situation, I think you can work the table the other way to get a sense of the boilers actual combustion efficiency.  Specifically, if you enter the table with your observed combustion air temperature, the manufacturer’s stated boiler efficiency, and an assumed excess air % at the rating point based on best practice, you can also get a sense of where the boiler might actually be operating in terms of combustion efficiency.

Applying this technique to the boiler I was looking at, if you look up the operating specifications ….


… you discover that the rated boiler efficiency is in the range of 82% (output Btu/hr divided by the input Btu/hr).

Using this information as a starting point, and assuming that the burner was set up for 9-10% excess air when it was rated, which would be best practice, interpolated the DOE table data as shown below to estimate what the actual operating efficiency might be based on the steady state flue gas temperature that was achieved after the burner had fired for a while.


My analysis implied that a potential improvement of 3-9% range might be attainable in terms of combustion efficiency if the burner was fine-tuned and/or the heat transfer surfaces were cleaned.  That’s a significant cost savings when you consider the annual boiler gas consumption and that the savings comes directly off of that bottom line.

So, we can add tuning the combustion process and perhaps opening up and inspecting the boiler heat transfer surfaces to our list of potential opportunities.

Back to Contents

Assessing the Purge Losses Using the Logger Data

One of the reasons for logging the flue temperature was to allow me to assess the boiler cycling rate and average efficiency per the discussion up to this point.   But I also hoped to use the data to develop an assessment of the purge losses and potential savings that could be achieved if I could optimize the boiler staging.

I should note that it would also be possible to identify the boiler cycling pattern by logging the burner amps, as illustrated towards the end of my post on Aliasing.  The benefit of monitoring the flue gas in this particular case was:

  1. We were out of CTs, so we had no way to log current with out giving up on something else.
  2. It told us a bit about the combustion efficiency.

Ideally, I might have decided to log both temperature and burner current to have a semi-redundant data set.

Back to Contents

Trying Your Hand at the Purge Loss Calculation

In any case, I will close this post out by outlining the technique I used to assess the purge losses and potential savings that could be achieved by optimizing the boiler staging with out showing the details of the actual calculations.  That way, if you want to try your hand at it, you can.  In my next post, I will provide the details of the calculations in the spreadsheet I used  so you can compare your results  to mine (or just go there if you don’t want to try the calculation on your own first).

I have loaded a Comma Separated Value (CSV) version of the boiler flue temperature logger data at this link for you to use.  If you are not familiar with using that type of file, I have a blog post about working with that type of file and you will find it is very compatible with Excel.

I also included some climate data for Columbus Ohio on the Google Drive since you will also need that information.  But if you wanted to pretend the boilers were in a different climate, then I have a number of blog posts that describe how to find hourly weather data, bin data, etc. on the internet if you don’t already have that sort of data for the location you wanted to consider.

What follows are the steps in the process that I used to identify the purge losses for the existing system as it currently operates.

Step 1 – Associated an Outdoor Air Temperature with Each Boiler Flue Temperature Data Point

As we discussed, I used a fairly fast logging rate for the the flue gas temperature to make sure I picked up the nuances of the cycle.  But I needed to coordinate that with weather data, specifically outdoor air temperature data, which had a much slower sampling rate.  I accomplished this using the VLOOKUP function in Excel.

Step 2 – Detect the Changes in Temperature that Correlate with the Various Steps in the Boiler Cycle

As we observed previously, the steps in the boiler cycle showed up as sharp changes in the flue gas temperature profile.  I have reproduced that image below to make it easier for you to reference it for the current discussion.


For example, when the purge cycle starts, the large increase in air flow through the boiler created by the fan moving equipment room air through the boiler (vs. leakage through the closed intake damper with the fan off) causes a sudden drop in the flue temperature.  Similarly, the heat of the pilot igniting causes a jump in temperature.

We can use these sudden changes to create flags in a column of the spreadsheet associated with an event.  In terms of exactly how to go about doing that, one technique would be similar to what I describe in the blog post titled Assessing Steam Consumption with an Alarm Clock: Step 2 – Detecting a Pump Cycle.  But it is a bit trickier than that because the pump was either on or off; kind of a square-wave type signal.  In contrast, the boiler flue gas temperature signal is “noisier”.

For now, I will leave it up to you to figure out how to resolve that problem.  But I will show how I did it in the next post.

Incidentally, just so you know, there is more than one way to do this.  The approximate number of ways to do this equal to the number of technical people who read this post raised to the power of 3 (because there are three boilers in the central plant) (I needed something to tie the three to).

Step 3 – Determine the Cycle Length for Each Logged Cycle and the Number of Cycles per Hour

Once you have identified the start and stop point for a firing cycle, you can determine the length of each cycle.  If you divide the cycle length by 60 minutes, you end up with the number of cycles per hour.  I did this because I was anticipating that they cycles per hour would very with outdoor temperature so I wanted to be able to establish that relationship.

Step 4 – Develop a Relationship Between Outdoor Air Temperature and Boiler Cycles

At this point, I was able to set up a relationship between the outdoor air temperature and the number of  boiler cycles per hour.  I did this by filtering my data and then, by making a scatter plot of Cycles per Hour vs. Outdoor Temperature, which gave me this chart.


Then, I used Excel’s trend line feature to develop a curve-fit for the data points I had and a mathematical relationship that would let me calculate the number of cycles per hour based on outdoor temperature.


If you are not familiar with how to apply a trend line to a data set, I actually show you how to do that in a little video clip that is included as a part of the blog post titled Using Scatter Plots to Assess Building Performance–Part 2.

If you think about how real boilers might work, you will probably conclude that you need to put some caps on the cycling rate when you do your analysis, at least that is what I concluded.   I’ll let you think about that and draw your own conclusions for now.  But I will show you what I concluded and how I adapted the curve fit equation in the next post, where I show the details behind each of these steps.

So bottom line, our 5 day data set can be used to predict about how many cycles the boilers make in a year, which will let us figure out the purge losses for the year as well as the potential savings if we can eliminate the unnecessary cycles.

Step 5 – Determine the Combustion Air Flow Rate

The number we are really after is how much energy does a purge cycle cost due to the fact that we are taking combustion air from out of doors, heating it up by blowing it through an inactive boiler that is full of hot water, and then tossing it back outside again.  In other words, we are sensibly heating the air and there is a common HVAC equation that lets us calculate the sensible heat that has been added or removed from an air stream based on the flow rate and temperature rise.


To apply the equation, we can figure out the temperature rise by using hourly outdoor air temperature data and our logger data as was discussed under Step 1.  The missing piece of information at this point is the flow rate.

The combustion air fan needs to provide the air flow required for complete combustion during high fire.  So there is a clue.  During a purge cycle, the fan provides this air flow rate with out burning fuel.  So if we could come up with the combustion air flow rate, we would have a pretty good sense of what the air flow rate was during a purge cycle.

I will let you think through how to come up with the combustion air flow, and you can do it with out a field measurement with a bit of theoretical knowledge.  And, there is a clue in the information we looked at previously.  I will show you how I did it in the next blog post.

Step 6 – Determine the Purge Air Flow  Rate for Each Portion of the Purge Cycle

Once you make it through Step 5, you will have a pretty good sense of the peak air flow that will be produced by the combustion air fan.  The problem is that if you study the boiler cycle chart (reproduced below), you will notice that for some of the pre-purge cycle, the combustion air damper is modulating open or closed, meaning the flow rate is varying from minimum to maximum and back again as the combustion air damper modulates.


That means that the combustion air flow rate we estimated likely only applies to the full airflow portion of the cycle.  For the other portions of the pre and post purge cycle, we will need to come up with a different air flow rate.

I will let you contemplate how to do that for now and also, how to turn the cycling rate into energy, but will show you how I did it in the next blog post.

Step 7 –  Determine the Total Purge Losses for a Year Given the Current Operating Pattern

The next step in the process is to put two of the things developed up to this point together and come up with the overall purge cycle cost for a year of operation.  Specifically, you need to:

  • Take the relationship between boiler cycles and outdoor air temperature that was developed in Step 4, and
  • Combine it with the Energy cost per purge cycle information developed in Step 6, and
  • Combine it with some form of climate data like an Typical Meteorological Year file (TMY file), bin weather data, or hourly weather data for a recent year.

The result should be the number of purge cycles that would happen for the year represented by your climate data and the energy loss that is associated with them.

That information can be converted to dollars, which tells you what the purge losses associated with the current boiler cycling pattern are costing.

Step 8 – Determine the Savings Associated with Reducing the Number of Purge Cycles

If you take the calculation you develop for step 8 and adjust it to reflect how you think you can reduce unnecessary boiler cycles, the result will reflect the purge costs (which are necessary at some level if you don’t want to have boiler explosions) for an optimized system.  And, the difference between the two is the potential savings.

I will let you contemplate what the optimized firing cycle pattern might look like, but in the next post, I will share how I visualized it.   The table below presents my results, and maybe a clue about how to think about all of this.


The bottom line is that some fairly simple adjustments could potentially saving at least $800 per year and may deliver almost $2,000 per year  in savings depending on the specifics of the current burner set-up.

Back to Contents

Hopefully, all of this gives you some insight into how to go about assessing the savings associated with what would likely be a relatively low-cost/no-cost adjustment to the boiler control algorithm for a central hot water plant.

And in the broader sense, I hope it shows you how to go about thinking about an energy savings calculation and how something that appears fairly simple can have a lot of nuances associated with it when you get into the details.

But, as they say, God is in the details (or the devil, depending on your perspective).  And from my perspective so is the fun.


David Sellers
Senior Engineer – Facility Dynamics Engineering

This entry was posted in Data Logging, Excel Techniques, HVAC Calculations, HVAC Fundamentals, Operations and Maintenance, Retrocommissioning Findings. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s