Max recommended flow velocities in pex

cabowen

based on all I have read over the years, the recommended flow velocities in pipe for hydronic applications has been 2ft/sec minimum to 4ft/sec max. Recently I have been seeing pex manufacturers, such as Uponor, use 8 ft/sec as a maximum recommended flow velocity. That certainly changes the game when it comes to the BU's any given pipe size can be expected to carry. What gives? What are the experts saying about moving water at 8 ft/sec through pex??? Thanks...

https://uponorpro.com/~/media/Extranet/Files/Heating%20Literature/hePEX_Strts_PG_H457_0213.aspx?sc_lang=en

Ironman

Flow velocity cannot be used as the only factor in pipe sizing. At that velocity, the head would be too high for most residential hydronic applications. If you'll notice, the head for each tube size at 8 fps is over 30'.

Potable water lines are another story.

jumper

Ironman said:

Flow velocity cannot be used as the only factor in pipe sizing. At that velocity, the head would be too high for most residential hydronic applications. If you'll notice, the head for each tube size at 8 fps is over 30'.

Potable water lines are another story.

What length?

Ironman

Per 100'. The chart gives it per foot.

EBEBRATT-Ed

too much velocity =noise and air removal issues + higher head.

Oversized is not an issue but you need enough velocity to move air back to the air seperator

Gordy

Pipe velocity with pex is a different animal than with copper, or iron pipe. The minimum 2fps is to keep air in solution to get back to the air removal device. The 4 FPS is to minimize velocity noise, and pipe erosion.

Pex manufactorers can get away with the higher velocities as the pipe erosion with pex is not as detrimental, and velocity noise will be less in plastic pipe.

What you can’t get around is head loss which dictates pump size. As @Ironman alluded to.

cabowen

I get the head loss concerns, but if you are only talking 20'-30' of pipe, it shouldn't be that big a deal....

Gordy

Probably not for 20-30’.

Zman

It does beg the question, what do you do about the velocity at both ends when it transitions to another material?

Larry Weingarten

Hello, I’ll add that the PEX fittings matter. At higher velocities, erosion could be a problem with brass fittings... plastic fittings, not a problem that I’ve ever heard of.

Yours, Larry

cabowen

Appreciate all the comments.

hot_rod

8 fps comes to a square turn like this? Be interesting to see that demo in a clear plastic fitting. Add some high temperature, high TDS??

Make me wonder if the high FPS is so they can better match copper flow rates, seeing as the ID is much larger in copper, especially 2" and up?

Pipe sizing above 2" tends to look a pressure drop more than fps velocity.

GroundUp

The part about velocity and flow rate that gets me, is that the pump curve is often WAY off of reality. Let's take a small 3 loop radiant slab for example, with 1/2" pex tubing in 250ft lengths. Straight water at 120 degrees. For fun, let's say 1 GPM per loop- my math says 3.8 psi drop or 8.78 ft of head, at only 1.7 FPS velocity. Now let's pump that zone with a 15-58 on speed 2, 8.78ft of head should be dead center of the curve and be moving 6 GPM. Now, we all know even on high we'll never see 1 GPM through a 250ft loop so instead of 6 GPM on speed 2 we're only moving 2 GPM. Why is that? If we want to get anywhere near 4 FPS we're talking over 2 GPM per loop, and there isn't a standard residential circulator I've ever seen that will get anywhere near that. Can someone tell me why? What step am I missing?

hot_rod

If I understand the question, you would need to develop a system curve and lay it over the pump curve to locate the actual operating point OP of the circulator.

So plot a series of scenarios of flow rate and pressure drop from the tube specifications, draw a line with a french curve. then lay that on top of the pump curve.

Grundfos had some transparent overlays for examples of determining OP.

In Idronics 16 we take you through a long hand calculation to plot that curve at various flow rates,

GroundUp

I will try to dig up what hydraulic resistance is and how to calculate it (assuming it's not the same as pressure drop), as formula 4-5 is not listed in the note provided. I have always been under the impression that simply calculating the pressure drop/friction loss in a system and plotting that point on the pump's curve would give approximate flow rate, which in shorter loops has always been spot-on for me. The long radiant loops are the only things that don't work out that way- it's not often we see over 1 GPM even while purging a single loop with a transfer pump, much less a circulator. Thanks for the info Bob