Kv or Cv versus K Coefficient for valves etc

[iainsavage]

Having examined a few manufacturer's families they seem to be using Kv values in the K Coefficient parameter. I'm not convinced that this is correct. There are a few sources on the web which give various methods for relating Kv or Cv to K, mainly based on data in the Crane manual 410M "Flow of Fluids Through Valves, Fitting and Pipe".

From this method K would relate to Cv as:

K=894.01*(d^4)/(Cv^2) with d in inches and Cv in US gpm. (rearrangement of Crane equation 3-16)

If Kv = 0.865*Cv and 1" = 25.4mm then by conversion this implies:

K=0.0016073*(d^4)/(Kv^2) with d in mm

If manufacturers are simply inputting Kv to the K Coefficient parameter then this means the pressure drop calculations would be wrong. For example, using the above equation for a valve with Kv = 4, d=20mm => K=16 i.e. actual pressure drop using correct value of K instead of Kv would be four times greater.

Most of what I have gleaned from the web (and therefore my own interpretation of it) seems to be well-meaning speculation rather than hard facts from Autodesk - could someone from Autodesk, or someone who KNOWS the answer, please validate my assumptions? Thanks.

[Martin__Schmid · Accepted solution]

Indeed, K, Cv, and Kv are different quantities as described here:

https://neutrium.net/fluid_flow/converting-between-cv-kv-and-k/

Revit computes the head loss through a fitting as described here:

https://neutrium.net/fluid_flow/pressure-loss-from-fittings-excess-head-k-method/

However, Revit doesn’t report the loss in head (units of length), it reports it in units of pressure (e.g., PSI).  To convert head to pressure, you need to multiply by the density of the fluid (and throw in 144 for conversion of sq ft to sq in)

v: velocity in ft per second

h: acceleration of gravity, 32.17 ft/sec²

ρ: density of fluid (lb/ft²)

K: unitless factor

ΔP: Pressure drop in PSI

So, yes, if you have content where manufacturers are putting in values for Kv (which should have units of m^3/hr) and populating that in a unitless parameter, and using that parameter in place of K, then things are't going to work out well.

Hope that helps.

[iainsavage]

Thank you very much for your response.

The first link that you provided appears to verify my own calculations derived from the Crane manual (with possibly a small rounding error since Neutrium give K=0.001604 * d^4 / Kv^2 and I got K=0.001607 * d^4 / Kv^2).

In either case this appears to confirm that Kv values should NOT be used directly in the K Coefficient parameter and must be converted first. We should therefore be wary of using manufacturer's valve families for pressure loss analysis unless we check that they have correctly set the K Coefficient parameters.

Thanks again.