Damper Leakage Calculation 3

[TECMSC]

Hello All,
I was trying to figure out the amount (CFM) of outside air leaking through a closed control damper. After some thought it seems the best method is to calcualte the outside air percentage based on temperatures per the following:
x% = outside air percentage (OA%)
Ta = outside air temp
Tb = return air temp
Tc = mixed air temp

x% = (Tc - Tb)/(Ta - Tb)

If I know the total CFM, then multiply total CFM by OA% to get the air leaking through the damper.

I appreciate any thoughts...

 

[marcoh]

First thought would be that damper leakage is dependant on the pressure differential across the damper rather temperature? Most suppliers will provide leakage rates at a given pressure difference.

 

[quark]

Marcoh,

The OP is doing sensible heat balance. It should work if the temperature measurement is precise. However, the formula will be x% = (Tc-Tb)/(Ta-Tc)

 

[MintJulep]

In theory it will work.

In practice....

How good is the mixing? What are the relative temperatures? Are there any other sources of heat gain or loss?

 

[mgmasc]

Sometimes the mixed air temp method for determining the percentage of o/a is not practical due to as Mintjulip suggested i.e improper mixing and/or an insufficient temp spread between o/a and return air. What we have done in the past is measure the delta P across the damper, convert that delta P to a velocity (FPM) using Vp^ x 4005. Measure area (sq ft) of the damper opening as best you can. area x FPM = CFM. This will get you in the ball park.

 

[UtilityLouie]

Another suggestion is to look at damper manufacturers like Ruskin or Greenheck. In their submittal data they have a leakage rate @ differential pressure chart. Even if your damper isn't one of these manufacturers, you can pick a model that's similar. It should get you in the ballpark.

One note: Make sure you know enough about YOUR damper before you do anything. Does it have blade edge seals? Does it have jamb seals? For an OA damper, if the answer is NO to either of these questions, you need to replace it. Depending on your location, the damper should ideally have insulated blades and (in ideal cases) a thermally broken frame.

 

[TECMSC]

Thanks all for the input.

marc/utility/mint -- Damper Specs: I need to dig up the specs but I believe design should be less than 2 cfm/sqft leakage. In this case, we know they leak and are trying to determine if leakage is excessive and could potentially de-rate the unit on a design day e.g 105 degF ambient

Quark
I've read many of your posts and hesitate to disagree, but let me give it a try: Are you sure?) the mixture line on the psych chart defines a percentage or part/whole or (Tmix - Treturn) / (Toutside - Treturn) which is the portion of outside air contributing to the mixed air temp.

mgmasg
this is an intersting approach and will try to add it to the test to confirm the temperature approach (once i figure out the correct formula-with a little help <) another idea is to measure the velocity over the area of the OA damper/louver, average it and plug it into Q=AxV for CFM leakage.

I even came up with this scenario: turn all units to 100% VFD, open the doors(e.g one or two double doors or as needed to get a stable velocity reading) and measure the velocity and then calc the CFM. The room is well sealed, no windows and few doors so the path of least resistence for the air should be out the doors...that's the theory.

 

[quark]

TECMSC,

My formula should give you the OA as percentage of RA. When you do the sensible heat balance omitting the constants,

cfm_OA*t_OA+ cfm_RAt_RA = cfm_MAt_MA
cfm_OAt_OA+cfm_RA*t_RA =([cfm_OA+cfm_RA)*t_MA

Rearranging the terms,

cfm_OA*t_OA-cfm_OA*t_MA = cfm_RA*t_MA-cfm_RAt_RA or,
cfm_OA*(t_OA-t_MA) = cfm_RA*(t_MA-t_RA) and
cfm_OA/cfm_RA = (t_MA-t_RA)/(t_OA-t_MA)

I agree that this is not an accurate method. However, all methods have their disadvantages. Differential pressure method gives you error while trying to use an opening area. Further, there may be errors due to inaccuracy or fluctuation(more significant) in dp measurement. It is OK for a counter check, IMHO.

 

[MintJulep]

With the time and effort spent measuring leakage, why not just fix the dampers up so they don't?

 

[TECMSC]

MintJulep - I agree it is best to fix or replace the dampers. There are multiple 130ton units and the cost would be signifcant and potentially disrupt operations. This calc may assist in quantifying the energy waste etc to present a case to fix, replace or seal them off.

Quark - After I posted, I used your approach with a coworker but ended up with the formula in my original post. That being said, your formula looks correct.

I also had a formula on a cheat sheet from college but not sure how it was derived:
Tma = %oa*Toa + (1-%oa)Tra
and working this out I ended up with the original formula, but the terms were reversed x=(Tra-Tma)/Tra-Toa), although your formula still looks correct.

In this case as long as long as the data is good I believe the accuracy will be adequate.

I'll take some vitamins and look at my fuzzy math again...Thanks

 

[quark]

Don't worry, we both are correct. See the first line in my second post. I am calculating OA as percentage of return air and you are doing it as a percentage of mixed air.

Your formula is derived from simple sensible heat balance, as I wrote above.

Qma*Tma = Qoa*Toa + Qra*Tra
Tma = (Qoa*Toa/Qma)+(Qra*Tra)/Qma, or
Tma = %oa*Toa+[(Qma-Qoa)*Tra]/Qma, or
Tma = %oa*Toa+(1-%oa)*Tra

 

[TECMSC]

quark - DETAILS!! I appreciate your input and follow up.

Thanks all.