Duct fitting pressure loss factor for industrial fan sizing

[AgsMyDude]

Originally posted in general mechanical but did not receive any replies. This is not an HVAC application but maybe some folks here have some insight...

We are in the final stages of a study and are currently estimating the resources it would take to bring the project from budgetary level of definition to IFC. I am looking for some information on hydraulically modelling ductwork systems to size large fans (approx. 500,000 cfm @ 20” H2O).

For the purposes of the budgetary study design, we used loss factors for specific ductwork fittings(area changes, tees, and bends) from the ASHRAE Duct Fitting Database and input into AFT Arrow. I guess my concern with the ASHRAE database is that it is clearly geared towards smaller HVAC system. Our ductwork will be up to 15'x15' in some places. I confirmed that the loss factors provided by ASHRAE do not change with scale (i.e. loss factor for a 4"x4" bend was same for 8"x8", given the same velocity for the two cases) but I am wondering if there is a better methodology for estimating ductwork dP at this scale. We do have CFD capability but have not typically used CFD to size fans (although we could if that is typical).

My ultimate goal is to scope the effort necessary to build on what we have at present and produce a calculation that could be used to support purchase were the project to proceed.

Thanks in advance

 

[EnergyProfessional]

Every fitting is made and they all will differ a bit. And not all fittings are covered by ASHRAE or other sources. So select one that is close. Even worse for transitions and junctions. You could get the same junction from 2 sheetmeal workers and one will have many more exposed bent metal and screws than the other and consequently will introduce more pressure drop and noise. So what I do in HVAC design is pick a fitting that has a value that is close to the actual fitting and be conservative.

I doubt there were real-world tests on the large fittings you talk about and they likely are very custom-made anyway. I suspect on a very large fitting or duct, the workmanship (overlap, seams, screws etc.) will matter less. Unless someone comes up with a verified source for such large fittings, I'd just use the ones you have. Who knows what fittings were used by ASHRAE when they came up with then umbers and if they applied safety factors. Maybe they had fittings made by the best dcut fitter int he World and real-World fittings have much more pressuredrop. Or they added 20% as a safety factor.

 

[GT-EGR]

In addition to just using the closest ASHRAE equivalent available - I’d see if there is a trending difference as you move from Ashrae’s smallest to largest sizes with the same type of fitting. Maybe you’ll find that as you get larger but keep the same relative geometry and velocity that larger fittings have less loss compared to smaller ones. Then you could consider using the biggest ASHRAE fitting equivalent to be conservative. If you find it’s the other way around then you may have to dig in a little more, but I’d anticipate the former.

 

[Stick.]

Disclaimer: I'm not exactly qualified to answer this, but I've worked with people who are, so I have some second hand knowledge. Hopefully it at least helps your search a bit.

You can size ducts of this magnitude based on loss coefficients, but I don't know what said coefficients are or if they differ from ASHRAE. CFD isn't necessary unless you need increased accuracy or specific details about the flow. I suppose it may be of benefit if you had a particularly complicated section in the duct run, but I would only do CFD on that one portion. For the structural design of ducts like that, I've used the industrial duct construction standards from SMACNA (there's one for rectangular and one for round duct). A quick scan of my copy didn't include anything about flow sizing, but maybe a different SMACNA standard covers that. If nothing else, those standards will be helpful when it comes to designing the duct structure since that is non-trivial at the sizes and pressures you're discussing.

 

[AgsMyDude]

Thanks for the responses all, it is sounding like the approach we took is likely adequate. GT-EGR: I did confirm that the loss factors ASHRAE provided did not change with scale (i.e. loss factor for a 4"x4" bend was same for 8"x8", given the same velocity for the two cases). We did use CFD to validate turning vane design and flow profile to for even flow distribution at equipment downstream of the fan and the dPs calculated for that portion of the system were a bit lower with CFD than that with loss factors so we will likely stick with the more conservative approach.

 

[EnergyProfessional]

I think if you can master CFD, you may have better tools at hand than ASHRAE....

For HVAC design we also always assume that in the field the contractor installs the duct differently to avoid some obstacles or some error in sequence of installation. Like when the sprinkler pipes were installed first, and then there needs to be additional fittings. Often the contractor also cuts corners and installs less efficient fittings. They shouldn't, but often real life installation will be worse. So add some safety margin. So overall, the exact value may matter less than you think.