## Turning Vanes and Duct Elbows, Part 1

Recently, someone asked me if it made sense to add turning vanes to a radiused elbow.  That’s one of those questions where there is no exactly a right answer that applies for every situation. The impact of the turning vanes has a lot to do with both the geometry of the fitting and the velocity of the air moving through it. To gain some insight into this, lets consider a 12” square elbow with a 12” centerline radius (i.e. the centerline radius is equal to the duct width) like the one pictured below. At 1,500 fpm, the loss through such an elbow without vanes is about 0.03 in.w.c.; virtually nothing. If we add one turning vane, the loss drops to .01 in.w.c.; with three vanes, we end up with a calculated loss of 0.00 in.w.c. The parameters associated with these configurations are summarized in the table below. To me, these results indicate that in a low velocity application, all three elbows are pretty close to the same and its probably not worth the added cost for the vanes. But, if you double the airflow, the losses are much more significant. Now, the unvaned elbow has a loss of 0.12 inches w.c. while the elbow with three vanes calculates out to 0.01 in.w.c. In other words the elbow with out vanes has a measurable loss while the elbow with vanes is nearly inconsequential. Here is a table summarizing the parameters associated with the elbow operating at a higher velocity. To me, these results indicate that the vanes may be worth something in a high velocity application.

If you take the three vane elbow at a high velocity and start to change the geometry of the elbow, the loss starts to increase. For instance, if you reduce the center line radius from 12” to 8”, the loss goes up to 0.02 in.w.c. all other things being equal. If you start to flatten out the elbow, the same thing happens; a 7” x 21” elbow has a loss of 0.5 in.w.c. all other things being equal (those are funny duct dimensions, but I did it to illustrate the effect  while holding the velocity relatively constant). Here is a table summarizing the parameters
associated with the elbow operating at a higher velocity. So, my point is that you have to look at the vane decision in the context of the application; the turning vanes will be more important and worth more in high velocity applications and in applications where turn radius are tight and aspect ratios are high (thin, flat ducts).

In the next post, I’ll continue this discussion and look at what happens when you increase the size of the duct. David Sellers
Senior Engineer – Facility Dynamics Engineering This entry was posted in Uncategorized. Bookmark the permalink.

### 1 Response to Turning Vanes and Duct Elbows, Part 1

1. David Sellers says:

Sorry to not have replied sooner. I missed this comment when it came in. I also realize that this is one of the posts that I need to fix since a lot of the graphics seem to have been lost when I moved stuff from the CSE web site, so I will take care of that next.

The answer to your question is like the answer to many HVAC questions; it depends. The primary factors that will come into play are the geometry of the elbow and the velocity through it.

For instances, a high aspect ratio (flat and thin) elbow will tend to have higher losses than a square one all other things being equal. And even if the loss coefficient is 1.0 (which is a pretty high loss coefficient), if the velocity is only 500 fpm, then the velocity pressure will be 0.0156 in.w.c. and thus the loss will be 0.0156 in.w.c. In contrast, if the velocity was 2,000 fpm through the same elbow, then the velocity pressure will be 0.2500 in.w.c. and the loss will be 0.2500 in.w.c. As you can see, the velocity has a major and non-linear impact; an velcoity increase of a factor of 4 resulted in the loss increasing by a factor of 16, a manifestation of the “square law”.

So to really understand the value (or not) of improvements like turning vanes, you probalby need to do the analysis, but it will certainly be more significant for higher velocity ducts. The analysis is not that difficult once you do it a couple of times, espeically if obtain some of the resources listed in the posts, like the ASHRAE loss coefficient tables.

David