calculating air velocity given CFM 3

[PaulKraemer]

Hi,

I have a fan with a rectangular outlet with cross-sectional area 0.89 square feet. The outlet of this fan goes into a transition piece that steps up to a rectangular duct with cross-sectional area 2.65 square feet. This section of rectangular duct contains a modular duct heater with evenly spaced heating elements. On the outbound of this heater duct is a transition piece that steps down to 8 inch diameter round duct (cross-sectional area 0.35 square feet).

I have been told that the maximum required air flow will be 1000 CFM. If I want to calculate my maximum air velocity in each section, is this simply a matter of dividing CFM (which is always 1000) by the cross-sectional area? If this is the case:

Fan Outlet : velocity = 1000 / 0.89 = 1124 fpm
Heater Section : velocity = 1000 / 2.65 = 377 fpm
8 inch dia round : velocity = 1000 / .35 = 2857 fpm

... I feel pretty confident that this is correct, but I just wanted to make sure I am not missing anything. The reason I want to know this is I am trying to determine the expected pressure drop across my duct heater, and the data provided by the duct heater manufacturer indicates that this is dependent on air velocity.

Thanks and best regards,
Paul

 

[LittleInch]

Only if the air pressure is atmospheric or close to it (say within 10%). Those are pretty high velocities so there might be some significant pressure rise. What pressure does the outlet sit at?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[PaulKraemer]

Hi LittleInch,

The blower/heater/duct arrangement I described is part of a more complex system that I am trying to get an understanding of.

Basically, this is part of a machine that applies a liquid coating to a continuous web (which could be paper, foil, or any sort of film). This web is then conveyed through a two zone hot air drying tunnel.

The blower/heater/duct arrangement I described in my original post is part of the system that delivers heated air into the two zone dryer.

Each dryer zone has a supply blower that feeds air into an electric duct heater followed by filters. After the filters, the heated/filtered air is directed into a plenum at the top of the dryer zone. The plenum has multiple air nozzles that direct the heated air at the coated web as it is conveyed through the dryer.

While the dryer has two separate supply plenums with independent blowers/heaters/filters, all of the supplied air is exhausted by a single exhaust blower. The sketch I have attached is my attempt to illustrate the entire system. While the illustration makes it look like the two dryer zones are totally separate from each other, they are actually side by side in a single drying tunnel. The only separation is the plenums used to supply the air for each zone. Once the heated air passes through the air nozzles, there is no physical separation or barrier between the two zones. There is also only a single exhaust blower, whereas the illustration might make it look like there are two separate exhaust blowers.

In the illustration, the point I have labeled "C" is the eight inch diameter duct where I estimate that 1000 CFM would be the maximum air flow we would ever want to move. My thought process when I asked my original question was that if we were moving 1000 CFM here, I suspect that we would be moving 1000 CFM at any other point between the supply blower outlet and point "C" because there is no where between these points where air can escape. If I could assume that we would be moving 1000 CFM through the duct heater, it made sense to me that I might be able to calculate air velocity through the duct heater by dividing 1000 CFM by the duct heater cross-sectional area.

I have not yet measured the static pressure anywhere in this system and I don't know how I would calculate or estimate it. I don't have access to one of these machines at my current location, but I will try to make some measurements when I have an opportunity.

In the meantime, I guess my question is ... if I were moving 1000 CFM at point "C" (in an eight inch diameter duct), would pressure build-up in the system create a condition where I wasn't also moving 1000 CFM through the duct heater and/or where I would not be able to determine air velocity by diving by cross-sectional area?

Just as a note ... we have been building dryers this way for a long time and our customers are able to achieve their desired drying effect. We probably over-size our blowers and ductwork for the air flow we end up using, but using variable speed drives gives us the adjustablity to tune the system by "feel" (is the product getting dried properly). The purpose of my question is an attempt to get a better understanding of the system dynamics from a scientific/engineering perspective.

I appreciate your help.

Best regards,
Paul

 

[LittleInch]

Paul,

I hope I'm not being too simple here, but you need to be very careful when talking about gases (air) when you use CFM as to what conditions it refers to.

There are many threads in this forum on the difference between ACFM, CFM, SCFM, etc. Basicllay saying CFM is virtually meaningless without saying CFM at X pressure and Y temperature.

This is why in scientific terms and when buying gases ( HC or pure gas) people either use energy, mass or volume at an agreed "standard" set of conditions. This is often, though not always, 15 C and 101.4 kPa.

Now in reality many people use SCFM to mean volume of air at the local atmospheric conditions (higher up - less pressure / mass per unit volume)

In your case you will have the same volume flow at any point when converted to the same set of pressure and temperature.

However the Actual CFM and hence ACTUAL velocity at any point will vary with pressure and temperature. So if say the inlet pressure from your blowers (point B) is say twice the local atmospheric pressure, then the velocity will be half what it is if the pressure was atmospheric.

The velocities you have seem very high to me so you could require some high pressures which would affect actual CFM and actual velocity. Hence my guidance that if the pressure at B is more than 10% above inlet pressure into the blowers, you need to start accounting for pressure and temperature.


Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[PaulKraemer]

Hi LittleInch,

After thinking about it, I am not completely sure whether I am talking about SCFM, ACFM, or something different.

On one of of our machines, we installed a pitot tube and a temperature sensor in a straight portion of the round duct leading into our dryer (at point C in my diagram, which I attached again).

From literature on the Dwyer (pitot tube manufacturer) website, I found the following formula to calculate air velocity ....

V = 1096.7 * SQRT(Hv / d)

... where ..
V = air velocity in ft/min
Hv = Velocity pressure measured by pitot tube (differential between total pressure and static pressure)
d = density of air

... I used my measured temperature to calculate d as follows ...
d = 0.075 x 530 / T
... where ...
T = absolute temperature measured in duct C at the location of the pitot tube

... After using the above formulas to calculate V (air velocity), I multiplied V by the cross-sectional area of the duct, and displayed the result as "CFM"

... With the above method of calculating "CFM", I take temperature into account but the only pressure I take into account is velocity pressure measured by my pitot tube. I do not take account of what the static pressure or the total pressure is at the point of measurement. Correct me if I am wrong, but I would suspect that my static pressure (and total pressure) might be high here as I a moving quite a bit of air into and through various restrictions. Would you say that the value I am displaying as CFM is ACFM, SCFM, or neither of the two?

Thanks again for your help - I really appreciate it.

Best regards,
Paul

 

[LittleInch]

As described you are without doubt measuring Actual velocity and hence actual cfm. To convert to standard conditions you will need to know pressure and temperature at your measurement point.

If d is density then you need to correct for pressure as well as temperature....

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[PaulKraemer]

Thanks LittleInch,

Now that I know I am measuring actual CFM at my measurement point, is it a valid assumption that my actual CFM will be the same at all points in my system including (1) my Supply Fan Outlet, (2) through my electric duct heater, (3) through my filters, and (4) where my air nozzles deliver the heated air into the Dryer?

My guess is that this is not a valid assumption for the mere fact the air at my measurement point is heated, and I use the measured temperature in my calculation to determine ACFM. The air exiting my Supply Fan Outlet has not been heated yet, so I suspect that the ACFM here must be different here. The air passing through my electric duct heaters is in the process of getting heated, so I am not sure how this complicates things.

I am wondering if my measured actual velocity / actual CFM of the heated air at my measurement point gives me enough information to calculate the actual velocity through my duct heater. Other than trying to educate myself and get a better understanding of my system, my immediate goal is to determine the expected pressure drop across my duct heater. According to the duct heater manufacturer, the expected pressure drop is dependent on air velocity. (I have attached a table that shows the relationship).

Thank you again - this has been extremely helpful.

Best regards,
Paul

 

[GT-EGR]

In response to your original question

Assuming you are operating somewhere near standard temperature and pressure conditions - yes the manufacturer is looking for duct velocity as you calculated them - and they probably have a chart or flow coefficient that you would plug that velocity into that would vary the pressure drop based on the square of that velocity

Just note that manufacture pressure drops are provided under ideal conditions. So if you are hitting up that fan outlet right into a bunch of transitions one after the other, it is very possible that the pressure drop across the element is actual much higher under your installation.

 

[LittleInch]

Paul,

Actual velocity and actual CFM change around a system depending on the pressure and temperature of the air at any point. In a steady state system the constant at any point is mass flow which you can convert to SCFM.

Now how much you actually need to do this depends on the relative changes in pressure or temperature and how accurate you want to be.

So e.g. if you have a system which doubles the pressure from say 1 bara (atmospheric pressure) to 2 bara (1 barg) then the volume will, within 5%, halve if the temperature doesn't change much. So in that case for the same size duct, the velocity will also halve.

Now for practical purposes, if the pressure doesn't change by more than 10 - 15% or the temperature in C by more than 50%, then you have more or less the same velocity.

You haven't said anywhere what pressure this system operates at in its various parts, but given your duct heater pressure losses are in fractions of an inch of water, you may well fall into the low change in pressure situation outlined above.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[PaulKraemer]

Thanks LittleInch,

I am really interested in measuring the pressure at various points in my system so I can get a better understanding of the dynamics. At the moment, duct work is being installed for the newest machine we have built so I should be able to install whatever measurement instrumentation I think will be useful. If I do this, would you recommend a single static pressure probe at each key location (fan outlet, inbound/outbound of heaters, inbound outbound of filters, the location where I use my pitot tube for air flow measurement, and entrance to dryer)? If I do this, would you recommend that I use a simple differential pressure gauge (or transmitter) with the (+) port connected to the static pressure probe inside the duct and the (-) port connected to atmosphere outside the duct (which is almost always just ambient conditions)? Or should I do something different or more involved?

Thanks again for the help - I really appreciate it.

Best regards,
Paul

 

[lilliput1]

Are you sure the filter is downstream on the coil? Your system would work better if the filter is upstream to catch dirt or lint before it gets to the electric heating coil. Also the filter would assist in evening out the airflow over the coil. See SMACNA HVAC Systems Duct Design for system effects, recommended angle of converging and diverging transition.

 

[PaulKraemer]

Hi Lilliput1,

Thank you for your response. Yes, the filters are downstream of our electric duct heaters. This is the way we were building these systems when I started working for my company and I never questioned it. I honestly don't know if this was a mistake that was made initially and that was repeated over and over, or if there was a specific reason we decided to do it this way.

My only thought regarding having the filters downstream of the heaters is that if there is any chance that over time, metal flakes or particles might chip or come off the heater coils, these would be stopped by the filters.

On the other hand, your point about having the filters before the heaters catching dirt and lint before the heaters and evening out the air flow makes sense to me. This would also save us the trouble of having to make sure we do not reach or exceed the temperature ratings of our filters.

The only thing I should mention that this is a pharmaceutical application that requires the air delivered into our dryer to be filtered a great deal. We install a 95% efficient ASHRAE filter followed by a 99.99% HEPA filter, and our customers routinely (I think once per year) perform some sort of "smoke" test to confirm that the filter are performing within the requirements. I am not sure if this makes a difference regarding the preferred filter location. I should also mention that all of our customers do have pre-filters upstream of our Supply Blowers.

Anyway, that SMACNA HVAC Systems Duct Design you mentioned looks like a very useful read. The $213 price tag will be a little tough on my budget, but I am leaning towards getting it.

Thanks again for your help,
Paul

 

[lilliput1]

You can put the 95% efficient filter upstream then the HEPA filter downstream of the coil. Hospital Surgery Suite AHU require HEPA filter downstream of the AHU and minimum MERV 7 (30% dust spot efficiency) filter upstream of the coil. Other hospital areas would require Merv 7 prefilter and MERV 14-15 (95%) final filter. If you need to maintain constant CFM throughout the filter loading you need to put in volume measurement and control. You can use Air Monitor air flow station & damper or variable fan speed control. Or you can use spring operated constant volume control valve Phoenix or Triatek. Check if fan can handle additional pressure drop.

 

[PaulKraemer]

Thank you Lilliput1,

I will definitely consider putting the 95% filter upstream for our next new build. The system that I am getting ready to startup has already been built, so my goal is just to learn as much as I can about the system dynamics. We do use variable speed drives for our blowers. I am very interested in volume measurement and control, but to this point, we have always just entered blower speed set points in RPM. The previous machine we built was the first time we attempted volume measurement. We used a pitot tube and a temperature sensor to calculate ACFM. This measurement instrumentation was installed by an outside calibration engineer. What I learned from him is that a smaller duct diameter allows you to get a more accurate ACFM measurement at lower air flows. In his opinion, a 14" dia duct will allow you reasonable accuracy at no less than 900 ACFM. A 12" diameter duct will allow you to measure down to 400 ACFM, whereas a 10" diameter duct will allow you to measure down to 250-270 ACFM. Our customers sometimes like to use very gentle air flows, so I would like to be able to accurately measure as low as possible. This tempted me to consider requesting 8" diameter duct. This scared me though because I do not know how to do the system calculations to determine whether going down to 8" diameter would overload our blower motor or cause other problems. I left the decision up to our customer's HVAC contractor and they chose 10" diameter duct for my current project. I would like to learn how to do the necessary calculations myself so that I can make more informed recommendations. It seems like that SMACNA HVAC Systems Duct Design .pdf you mentioned will be a good starting point towards this end.

Now that this current system is now already built and the duct work will soon be completed (at 10 inch diameter), my plan is not only to install the pitot tube and temperature sensors in the 10" diameter duct, but also to install static pressure probes at other strategic points in my system so that I can get a better understanding of what is going on.

I really appreciate all your help with this.

Best regards,
Paul

 

[lilliput1]

We found Ebtron thermal dispersion type airflow measuring devices to be most accurate. Otherwize have a testing and balancing contractor calibrate your system. Have your company purchase SMACNA HVAC Systems Duct Design Manual for their reference library. Learn from it how to calculate duct system losses. You have to add equipment pressure drop then learn to plot the system curve on the fan curve to find stable operating point.

 

[PaulKraemer]

Thanks Lilliput1,

I just purchased the SMACNA HVAC Systems Duct Design Manual, and I will look into the Ebtron thermal dispersion airflow measuring devices for my next project. Just out of curiosity, do you have a feel for what is the lowest air flow you would expect to be able to accurately measure with these devices?

Thanks again,
Paul