Pressure loss in an HVAC system

[MacLaddy]

I'm working on my first engineering project in the real-world and I'm hitting a bit of a snag. It's regarding pressure loss analysis in an HVAC system. We have a system that is capable of providing a maximum of 2.0-in.wg pressure at the air handler, but we are finding very little air movement at the end of the duct. Using the methods that I lay out below, I am finding a final number that is not realistic. I'm hoping somebody can provide me with some expertise. It's very difficult to find any information online that agrees.

The attached PDF file shows a sketch of the ductwork coming off of the air handler. The circles represent sections and the squares represent fittings. I have also attached an Excel Spreadsheet to hammer out most of my calculations. On the PDF there is a datum line perpendicular to section 5 on page 1 and page 2 to show how they connect.

Column L shows some of the assumptions I have made regarding the material. The material is galvanized sheet metal (Spiral circular duct), and I have used the Moody Chart for the friction factor f.

For the Major Losses (Column H), I have used the following formula.

$$ h_{Lmaj} = f\frac{L}{D}\frac{V^2}{2g} $$

Finding the minor losses has proven to be the more complicated part...and the place I believe I am making a mistake. For all of the items listed EXCEPT the 45° Reductions, I have used the values on this webpage to find the K Factor.
https://neutrium.net/fluid_flow/pressure-loss-from-fittings-2k-method/

For the two 45° Reductions, I used step 3.1 on the following webpage and calculated the Square Reduction K value for turbulent flow and then multiplied it by the steps in 3.2. (assuming a 45° angle)
https://neutrium.net/fluid_flow/pressure-loss-from-fittings-expansion-and-reduction-in-pipe-size/

Once the K Values were found I used the following equation to calculate the minor losses.

$$h_{Lmin}=K\frac{V^2}{2g}$$

You can see in the h(Lminor) column that I am getting some very large numbers. I suspect that my error is in the Tee, Run Through method, but I couldn't find any source for this online other than that website above.

If you follow over to Column K you can see the total of all pressure losses in the system. I calculated this at 6-in.wg, which...of course...isn't possible.

I appreciate any help that can be offered. I'm really at a bit of a loss here.

Thanks,
Mac

 

[anorlunda]

ping @russ_watters

 

[Asymptotic]

Caveat: I am far from expert in HVAC, but have had some dealings in such matters.Take what I say with a suitably large boulder of salt.

Where do the CFM values come from?

Do you have a pitot tube/manometer set or an anemometer to perform measurements?

Looking at the sketch, my first thought is much of the flow you are expecting to show up downstream is passing through fittings 5, 6, and 7, all of which are in close proximity to one another, and just upstream from the 90° elbow (fitting #8).

Air density in the worksheet is 0.002329. Units aren't defined, but appear to be slugs/ft³ at NTP (Normal Temperature & Pressure; 20°C & 1 atmosphere). If the actual installation is significantly above sea level then altitude must be factored into account for the loss in air density. However, this is unlikely to account for much of the low flow, "very little air movement" at the end of the duct symptom.

 

[JRMichler]

Start by measuring the static pressure at the discharge of the air handler. Also measure any suction at the inlet side of the blower, and the differential pressure across the blower. A simple water manometer is all you need for this. Compare to the fan curve for the blower.

Estimate the CFM from your measured differential pressure and the fan curve. Recognize that this is a crude estimate, but it's what you have, so run with it. Use this CFM to calculate duct loss and compare to your measured static pressure at the discharge of the blower. I expect you will find that your ductwork is restricting the flow.

Your best source for duct calculations is the ASHRAE Handbook of Fundamentals. You don't need a current copy, a 40 year old copy is just as good for duct loss calculations.

 

[MacLaddy]

Caveat: I am far from expert in HVAC, but have had some dealings in such matters.Take what I say with a suitably large boulder of salt.

Where do the CFM values come from?

Salt taken. The total CFM at the start of the trunk comes from summing the maximum potential of all VAV's down the line. If you start with 5,035 CFM at the start and subtract the VAV's CFM's as you move down, you'll end at zero.

Do you have a pitot tube/manometer set or an anemometer to perform measurements?

No, unfortunately. I'm not even on site. I'm doing analysis from afar... I will have full Test & Balance (TAB) documents in a few days, but I'm trying to verify the original engineering specs.

Looking at the sketch, my first thought is much of the flow you are expecting to show up downstream is passing through fittings 5, 6, and 7, all of which are in close proximity to one another, and just upstream from the 90° elbow (fitting #8).

Do you think this would have a significant effect on losses?

Air density in the worksheet is 0.002329. Units aren't defined, but appear to be slugs/ft³ at NTP (Normal Temperature & Pressure; 20°C & 1 atmosphere). If the actual installation is significantly above sea level then altitude must be factored into account for the loss in air density. However, this is unlikely to account for much of the low flow, "very little air movement" at the end of the duct symptom.

Yes, that is a bit sloppy. Sorry. The units are Slugs/ft^3 at standard conditions. I needed that convention to calculate the Specific Weight.
The property is on the west coast (e.g., sea level), so those assumptions should be close enough.

I realize that this analysis is a bit daunting. I appreciate any help that can be offered. Perhaps you, or someone, knows of a good ASHRAE (or other) Publication that lays out friction and K-factors?

Thanks again,
Mac

 

[MacLaddy]

Start by measuring the static pressure at the discharge of the air handler. Also measure any suction at the inlet side of the blower, and the differential pressure across the blower. A simple water manometer is all you need for this. Compare to the fan curve for the blower.

Estimate the CFM from your measured differential pressure and the fan curve. Recognize that this is a crude estimate, but it's what you have, so run with it. Use this CFM to calculate duct loss and compare to your measured static pressure at the discharge of the blower. I expect you will find that your ductwork is restricting the flow.

Awesome, thank you. Any good reference on finding a fan curve for a Carrier Air Handler?

Your best source for duct calculations is the ASHRAE Handbook of Fundamentals. You don't need a current copy, a 40 year old copy is just as good for duct loss calculations.

This is what I'm looking for. I've been trying to find the standard...as opposed to a plethora of googled sites.

 

[MacLaddy]

The ASHRAE Handbook of Fundamentals references a Duct Fitting Database. I've found it here at this link.

https://www.ashrae.org/resources--publications/bookstore/duct-fitting-database

Do you know if the handbook alone is sufficient to get through a problem like this, or would I also need to subscribe to this database? It seems a bit excessive for my purpose.

 

[JRMichler]

My own copy of the ASHRAE HOF is from 1985. It has 22 pages of fitting loss coefficient information. The HOF is more than enough to handle problems much more complex than yours.

 

[russ_watters]

@MacLaddy are you trying to reverse engineer someone else's work with that spreadsheet or is that the original calcs you are double-checking?

Am I reading your drawing right that you are trying to have 5,000 CFM going through a 14" duct? That's way, way, way too small of a duct for that much air. By a factor of 4!

Is there diversity not being taken into account?

I don't have time right now to dig into more details, but will have a closer look tonight.

I do agree with the others that you should use standard ASHRAE/SMACNA duct fitting loss values and calculations in the spreadsheet (and a ductulator for the straight runs). I'm not sure if you have access to the books or database where you work, but they have a program for it you may be able to download.

 

[MacLaddy]

My own copy of the ASHRAE HOF is from 1985. It has 22 pages of fitting loss coefficient information. The HOF is more than enough to handle problems much more complex than yours.

Thanks. I am working on ordering the 2017 copy as we speak.

 

[MacLaddy]

@MacLaddy are you trying to reverse engineer someone else's work with that spreadsheet or is that the original calcs you are double-checking?

Am I reading your drawing right that you are trying to have 5,000 CFM going through a 14" duct? That's way, way, way too small of a duct for that much air. By a factor of 4!

Is there diversity not being taken into account?

I don't have time right now to dig into more details, but will have a closer look tonight.

I do agree with the others that you should use standard ASHRAE/SMACNA duct fitting loss values and calculations in the spreadsheet (and a ductulator for the straight runs). I'm not sure if you have access to the books or database where you work, but they have a program for it you may be able to download.

I'm trying to reverse engineer the original work. There have been multiple problems in this facility for years regarding airflow. As a newly minted engineer this project is being presented to me as both a learning/training opportunity to see how these systems are originally designed, as well as trying to find the root cause of all the problems that this facility has had over the years. I know that reinventing the wheel isn't usually a good route, but for my continuing education purposes I think it is necessary.

Regarding the CFM, I sketched those drawings right from the original prints. If all of those VAV's are open 100% to their specification, then yes, it would be 5,000 CFM. How do you calculate max airflow for duct size?
I have no idea what you mean by diversity.

 

[MacLaddy]

The first spreadsheet was a bit of a mess. I've cleaned it up some and used K-Factors from the EngineeringToolbox website. So please...ignore that first excel doc. Use this one instead. It is iterative from the start of the run until the end, so the numbers don't match up exactly with my labels on the drawing.

Formula's I used.

$$h_{Lmin}=K\frac{V^2}{2g}$$
$$h_{Lmaj}=f\frac{L}{D}\frac{V^2}{2g}$$

The pressure formula I was using on the original spreadsheet was from Bernoulli's Equation. I realized that wouldn't be correct because it is an energy equation, and that has already been taken into account with the head loss formula's (right?). So this is what I used to calculate pressure from head loss.

$$\Delta{P}=\gamma\sum{h_L}$$

With this calculation my final pressure loss is even larger than before. At 7.5 in.wg, either I am making a large error, or this system has been very poorly designed.

 

[russ_watters]

As a newly minted engineer...

Got it...

How do you calculate max airflow for duct size?

I have no idea what you mean by diversity.

I'm concerned that it sounds like you may be being given no direction at all? Failing a true mentor...

PF is your friend, but google is an even better friend. Google "trane ductulator". I'm still using the one i was given my first day on the job 15 years ago. I was given guidance on pressure drop and velocity to use when sizing ductwork. If you haven't been, that isn't great. Engineering Toolbox has some tables, but for this I'd use a simple 0.1" to 0.15" per 100' upstream of the VAVs and 0.06-0.08" after. Velocities for this ductwork should stay below about 1500 to 2000fpm and get lower as you get smaller.

This type of system probably didn't have a detailed calc done, but your method is roughly correct for it and to fix a problem is a good idea. But find some standard tables; you are going too far back with your calls.

Diversity means variety. Here, It's the fraction of full airflow you ever expect to see. Since an east facing office and west facing office get sunlight at different times, they never peak simultaneously. For an office, you might assume 50-75%. Maybe this facility's engineer (did you say what it is?) assumed less.

Diversity applies to many complex systems, including pumps and even office computers.

 

[anorlunda]

ductulator

Holy makeral! I had no idea such a word existed.

 

[MacLaddy]

Got it...

I'm concerned that it sounds like you may be being given no direction at all? Failing a true mentor...

PF is your friend, but google is an even better friend. Google "trane ductulator". I'm still using the one i was given my first day on the job 15 years ago. I was given guidance on pressure drop and velocity to use when sizing ductwork. If you haven't been, that isn't great. Engineering Toolbox has some tables, but for this I'd use a simple 0.1" to 0.15" per 100' upstream of the VAVs and 0.06-0.08" after. Velocities for this ductwork should stay below about 1500 to 2000fpm and get lower as you get smaller.

This type of system probably didn't have a detailed calc done, but your method is roughly correct for it and to fix a problem is a good idea. But find some standard tables; you are going too far back with your calls.

Diversity means variety. Here, It's the fraction of full airflow you ever expect to see. Since an east facing office and west facing office get sunlight at different times, they never peak simultaneously. For an office, you might assume 50-75%. Maybe this facility's engineer (did you say what it is?) assumed less.

Diversity applies to many complex systems, including pumps and even office computers.

Let's just say that I have been given a unique opportunity for learning. I've been brought in as an entry level mechanical engineer that has a lot of HVAC troubleshooting experience. However, I am the ONLY Mechanical Engineer in the vicinity. I'm surrounded by Civil PE's at every corner.

I have an ASHRAE training seminar scheduled in the near future. I'll be taking the Level 1 and Level 2 course, as seen in the link below. I'm hoping that this may fill some of these gaps (mountains) in my knowledge. It even gives me a "Bonus – Free Copy of ASHRAE Duct Size Calculator (I-P and SI units)."
HVAC design is my thing, and I'll do what it takes to learn it. Retrocommissioning older buildings is a common undertaking with this organization, and I'm sure that I'll put it to good use.

As for Diversity...I'll have to look into it. I was just assuming a worst case scenario 120°F day when all VAV's are open Max. I suppose that isn't the best approach. However, the spreadsheet I have above shows velocities that are WAY higher than your recommendation. I definitely think there is a problem here with duct sizing.
https://www.ashrae.org/education-certification/hvac-design-and-operation-training

 

[MacLaddy]

I added a second page on that worksheet named 50%. It is simply half of the airflow for the entire system. It still calculates a pressure loss of 1.8". Considering that the air handler goes into alarm if it goes above two inches, this definitely implies a problem. Now the real question is...how do I fix it?

 

[JRMichler]

It's way too early to start fixing it. You need to check the design of the entire system.

Start with a load calculation for the entire area served by this HVAC system. Use the 99% outdoor temperature, and get an accurate estimate of the internal heat load (lights, computers, number of people, and ?). Get the specifications for the air handler, including the fan curve. Talk to the system owner - what problems does it have? Does it fail to cool whenever the outside temperature is above XX degrees, does it make noise, cause drafts, trip out, trigger alarms, etc?

Compare the stated problems to your calculations. When you calculate that problems should occur, and your calculated problems agree with the customer stated problems, then you are ready to start fixing it. This will take time. An experienced HVAC engineer could do it in a week or two. It will take you at least two months.

Dive in, get started, and have fun.

 

[Asymptotic]

Looking at the sketch, my first thought is much of the flow you are expecting to show up downstream is passing through fittings 5, 6, and 7, all of which are in close proximity to one another, and just upstream from the 90° elbow (fitting #8).

Do you think this would have a significant effect on losses?

Not losses, but (please forgive the anthropomorphism) where flow "wants" to go. Is the preferential path through a cluster of 4", 7", and 8" VAVs, or the relatively high resistance a 90° turn and ~ 25' run of 10" duct represents?

As you wait for the ASHRAE fundamentals handbook to arrive take a look at 'Duct Design' in the California Energy Commission Advanced Variable Air Volume System Design Guide.

 

[MacLaddy]

It's way too early to start fixing it. You need to check the design of the entire system.

Start with a load calculation for the entire area served by this HVAC system. Use the 99% outdoor temperature, and get an accurate estimate of the internal heat load (lights, computers, number of people, and ?). Get the specifications for the air handler, including the fan curve. Talk to the system owner - what problems does it have? Does it fail to cool whenever the outside temperature is above XX degrees, does it make noise, cause drafts, trip out, trigger alarms, etc?

Compare the stated problems to your calculations. When you calculate that problems should occur, and your calculated problems agree with the customer stated problems, then you are ready to start fixing it. This will take time. An experienced HVAC engineer could do it in a week or two. It will take you at least two months.

Dive in, get started, and have fun.

Thanks. I'll start working on compiling some of that information, and begin learning how to compile the rest. It's definitely an educational process at this point. I fear that I may have to lean on PF a bit for mentoring during the next few months. But I'll definitely have fun. What better challenge than to teach myself how to be an engineer? I can create all my own bad habits and practices instead of learning them from a mentor.

 

[MacLaddy]

Not losses, but (please forgive the anthropomorphism) where flow "wants" to go. Is the preferential path through a cluster of 4", 7", and 8" VAVs, or the relatively high resistance a 90° turn and ~ 25' run of 10" duct represents?

As you wait for the ASHRAE fundamentals handbook to arrive take a look at 'Duct Design' in the California Energy Commission Advanced Variable Air Volume System Design Guide.

Good information, thanks. I'll start looking through it this morning.