Welcome! Here are the website rules, as well as some tips for using this forum.

# Three way mixing valve sizing

Member Posts: 10
I am trying to figure out if there is a consistent way to size three way mixing valves. I would like to size a valve for an air handling unit in a school we are renovating and need some help. I have seen large pipes ran to small valves and looking for some background info. What info do I need to know? How does the Cv rating work into all of this? Is there some good resources I may be able to track down? Any help would be appreciated.

Mike Lundstrom

• Member Posts: 165

When sizing ESBE three way mixing valves we use the following: 1 to 8 GPM = 3/4" valve, 9 to 14 GPM = 1" valve, 15 to 22 GPM = 1.250" valve, 23 to 40 GPM = 1.5" valve, 41 to 60 GPM = 2" valve.
• Member Posts: 131
Valve Sizing

Call me, I can work you through it. 800-321-4739

Tim D.
• Member Posts: 272
Short strokes...

Here's some general stuff to consider which Tim can go into detail when you call him up...

Valves and heat terminal units have performance personalities and much like people are not the same.

In fact for valves, there are roughly five different 'personalities' determined by the shape of its plug . (Quick opening, linear, logarithmic & its cousin equal percentage, and split characteristic)

These valve personalities can be influenced by the speed of the stem stroke, the rangeability of the linkage and control and the minimum controllable flow specification provided by the manufacturer.

Fan/coils can have an infinite # of personalities depending on one of the conditions below:

Constant flow, variable temperature

Variable flow, variable temperature

Constant flow, constant temperature

Variable flow, constant temperature

You've no doubt heard the saying that opposites attract one another?

The same is true with matching a valve to a fan/coil. (or any heat terminal unit for that matter)

For example, a coil designed at 180 deg f, for a 20 degree delta t has postal performance which means it actually reaches close to 70% nominal output with only 17% nominal flow.

To linearize the output, ie: 50% output at 50% flow, one has to select a valve/actuator/control combination which is opposite in performance to the fan/coil(think laid back surfer dude) ie: having 17% flow at 70% of its full opening.

Its the combination of the postal worker and the laid back surfer dude which gives stable control.

A coil designed for a 40 or 50 degree delta t has a more linearized performance and thus requires a different valve.

The Cv tells us how much pressure drop the valve experiences at a given flow rate.

How much pressure drop should the valve experience is the million dollar question.

To have controllability over a system, the control valve should experience at least 30% to 50% of the pressure losses in the circuit it is controlling. The ratio of the head losses in the valve in comparison to the head losses in the circuit is called Valve Authority.

A valve having no authority, no controllability, is too big. Too big is represented by the Cv. If the Cv is too large it means the head losses through it are less than 30% of the circuit losses. It also tells us that in small loads it will get confused not knowing whether to open or close. (which is most of the time for over sized valves)

When the proper Cv is selected the valve has controllability, ie: it is operating in the mid range of its strokecapable of opening up for greater flow or closing for less flow.

If a valve is selected with a small Cv it will have greater authority but it also significantly increases the head losses in the circuit and thus the horsepower of the circulator.

If the right Cv cant be found from the manufacturers data sheets then a balancing valve is required. Its the combination of the head loss through the control valve plus the balancing valve which creates the right conditions for controllability.

Finally a few last things to consider:

Zone valves ARE NOT control valves.

Valve leakage (see manufacturers data sheet) can have a negative impact on system stability. Some valves are simply not designed for fan/coil applications. (ie: rotary/shoe types)

On/off valves (zone valves) on a postal performing heat terminal unit tosses the whole concept of valve authority out the windowone may as well put in gate valves.

The very reason why control manufacturers can take a single valve diameter but make it available with different Cvs and different personalities is because they understand marring valves to coils is a 'psychological' matching process...and is not for the amateur or those who like to use rules of thumbits a engineering design process no different than selecting a heat exchanger, circulator or boiler.

The wrong valve creates inefficiencies and temperature instabilities.

More importantly, inefficient systems with poor control create mad customers which tell two friends who tell their two friends who tell a few women and before ya know it the whole world knows how bad the system isall because of one oversized, incorrectly selected silly valve.
• Member Posts: 272
Some comics for ya...

• Member Posts: 131
Over-dose

I think that Robert over-dosed on writings by Pedijean early in life. He knows way to much about valve authority for it to be healthy. Seriously though, this is what it is all about. You need to get the right valve for the application. Look forward to talking with you.

Tim D.
• Member Posts: 111
3-way valves

are a particular problem. If using manual balance valves, it is necessary to put in two balance valves - one in the return from the coil and one in the line to the bypass to the 3-way valve. Or, you can put one automatic flow limiting valve (Griswold) in the return from the coil.

Properly selecting control valves is not that difficult if you have the right information, but I find that most modulating control valves are selected based on rules of thumb and wind up being over-sized. If the coil is also over-sized, the coil/control valve combination will not be able to provide stable control no matter what kind of control horsepower exists upstream.

If the system is constant flow (all 3-way valves) and is piped reverse-return, selecting a control valve with a head loss roughly equal to the head loss through the coil will usually provide adequate valve authority.
• Member Posts: 6,106
Nice, RB

that's a post all should print and keep! Thanks for taking it to that level. It really notches up the level of info and expertise here, to be able to grasp that info.

I'll put you down for "best post of the month" club. Again

Thanks!

hot rod

• Member Posts: 272
I agree...

Tim knows I have this thing about control valves and balancing...got it from Professor RP himself. ( I think Jerry got rubbed by RP as well) RP is still alive and living in Belgium - a legend in the world of controls.

Jerry is right about the three way valve being a particular problemmanufacturers offer them with ports having either the same (symmetric) or different (asymmetric) personalities and are designed as mixing (constant flow  two inlets  one outlet) or diverting (variable flow  one inlet  two outlets) devices.

and now you know one more reason why over 50% of building occupants are unhappy with their systemswrong valves - wrong combination of personalities- installed backwardsbut I digress.(see upcoming article in The NewsDesigning for Failure  Thanks to John Hall and Dan Holohan).

Since we control flow by controlling pressure and control pressure by the system design and equipment we need to understand where to place either automatic or manual balancing valves.

In constant flow (mixing) secondary applications with negligible influence from the primary circulator, the need for a balancing valve in the bypass line is questionable. However if the primary circulator has significant influence on the secondary side ( ie: the circs are in series when the valve is open) then a balancing valve should be placed between the primary return and the tee fitting connecting the bypass line to the three way valve. A balancing valve in the bypass line is required if its possible to have back pressure created by other three way valves configured as mixing devices but connected directly to a common primary networkthe automatic (Griswold) balancing device is definitely a nice product to have in the tool belt for applications where there is interconnectivity between primary and secondary loops.

In a variable flow (diverting) secondary applications with 100% influence from the primary circulator, a balancing valve in the bypass line is required and should experience the same head loss as the coil circuit. In this manner the bypass valve is in parallel with the coil and the circulator sees the same head loss whether the valve is open or closed.

In all cases there should be a balancing valve in series with the coil.

One of the major things to consider, especially in light of Mike Lundstrom question, is when it comes to control valves, pumps and comfort controls, its best to deal with a firm which understands the equipment, application and assembly. (It applies to 2, 3 and 4 way valves or injection pumps or manual or automatic balancing valves  there are no short cuts!)

As I have said beforecontrol contracting is not pipe fitting anymore than a cabinet maker is a carpenter. They may be touching the same materials but what they do with it is two different games.
• Member Posts: 3,656
Wow!!

So much to know...so little time. Great post. It's all in the details.

• Member Posts: 131
Total Hydronic Balancing

By RP is the most informative piece ever written on this subject. There are very few copies in circulation so if you have one it is a treasure. If you have one that you don't want call me, PLEASE!

Tim D.
• Member Posts: 6,106
I have a nice

piece that Tour Anderson put out years ago about hydronic balancing. A thick paper high gloss piece very pro photos and all.

hot rod