Calculating control valve pressure drop

[IanVG]

Hey everyone, I am trying to understand and then build an excel sheet to help me select control valves (just regular pressure dependent control valves).

The kinds of systems I work with have already been built a long time ago and are fixed flow at the moment. There are several projects coming up where we will begin to add control to allow the pump modulate and add control valves for the reheat coils at terminal units/AHU's to modulate flow based on the demand of the terminal unit. Fairly simple. Now, several sources I have seen/read guide me on selecting a control valve guide me first on finding the pressure immediately upstream of the control valve and immediately downstream of the control valve. My system does not yet have a control valve. This is confusing for me, because if I were to go to the trouble of calculating the pressure drop starting at the outlet of the pump through all the components to where I intend to put the control valve and then the negative pressure drop (pressure gain) starting at the inlet of the pump working backwards to where I will put the control valve- I get the same pressure!

Do these guides just fail to mention that I need to assume/guess a control valve that I will put in the system that causes some arbitrary pressure drop? That delta P just comes out of nowhere in my opinion, without any adequate explanation. One website here, does a fairly good explanation of the topic, but still defines the pressure drop across the control valve, as what the pressure drop across the control will valve. Feels like circular reasoning? Other engineers in my office have told me to just stick to a Cv of roughly half the design flow (which doesn't feel like a robust design method) which means (based on the Cv equation) assuming a pressure drop across the control valve of 4 psi. Why 4 psi? Don't I need to take valve authority into account and find a control valve, which once installed, drops around 25-50% of the total circuit pressure drop? And on the matter of circuits, that is a part of the system that starts out by branching off from a common supply line and ends when it ties into a common return line, correct?

What am I missing here? I can provide more resources, if that will aid anyone.

Edit: Typos & grammar.

 

[nuuvox000]

Just to clarify, are you varying the flow based on the number of reheat coils/AHU's that are calling for heating/cooling? Or are you wanting to vary the flow to each coil based on demand?

Varying the flow to each coil would likely require a motorized control valve that could be balanced. This is not common but sometimes it can make sense.

Varying the flow based on the number of units calling for heating/cooling is much more common. I would suggest using pressure-independent valves if possible. Typically you want to size balancing valves so that you're using the smallest valve you can use and still meet the flows. From there you just look at the worst case scenario (path of greatest pressure drop) and make sure your pumps can handle it.

I've also had controls designers tell me they like to have 3 to 5 psi of a pressure drop on a valve so that it can perform well and give good readings.

If you have an example with numbers that might help to understand where you're getting stuck. If you have a valve model in mind then the manufacturer of the valve will typically have a selection guide.

 

[HVAC-Novice]

I'm confused, how do you control if there currently is no control valve?

Research "Valve authority" to understand why a relatively high pressure drop is required. Based on that you pick a valve based on Cv value. The rules of thumb (i.e. 5 psi, or equaling the branch pressure drop) aren't really based on science, more on experience what "works". Doesn't mean they are or aren't wasteful. Obviously more pressuredrop = more pump energy.
Once you throttle the control valve, flow is reduced and pressuredrop in the remainder of the system drops significantly. So that chart in your link will be totally different once you actually control.

Modern system use PICV (pressure independent control valves) and they often have a 5 psi design pressuredrop (RTFM!). they also don't require manual balancing valves. Just use those and don't worry about all the above, they just work perfectly and don't require too much thinking during design. They also are cheaper if you account for balancing valves and less balancing required.

 

[IanVG]

@HVAC-Novice - the system I am looking at currently does not control. It is constant volume, that's why I am putting in control valves. I've gone down a rabbit hole on the various properties of control valves, and think I have a fairly good understanding (re: see Dunning Kruger Effect) of valve authority and its importance for good control. I am willing to go along with experience on what "works," but at the same time I'd like to understand why I am running with something like 4 or 5 psi, and not let's say always 50% valve authority (which could be 10, 15 or 20 psi, if the circuit it was balancing was 20, 30 or 40 psi respectively).

And, yes I agree, I've looked into pressure independent control valves, and from what I've gathered, I would much rather prefer to use these devices instead. The only significant specifications I can think of from a design standpoint for PICV's are minimum pressure drops required and possibly the turndown. My post was to address the design of pressure dependent control valves, in the instance I was limited to those devices or wished to understand the design of older systems on our campus. I am looking into rewriting some of our campus specifications on control valves, because some of them just don't make any sense to me (e.g. "all control valves shall have 300:1 rangeability/turndown").

@nuuvox000 thank you, I think you made a couple points clear. Yet, I am still confused, let me try and summarize. 1) There are at least two systems in which pressure dependent control valves can be installed to control hydronic flow for reheat coils/AHU's. One is based on the integer number of coils calling for heating/cooling (a coil is either satisfied (0), or is calling for cooling/heating (1) based on demand (delta T)). The other method is based on demand which is a gradient between 0 (no demand) and 1 (full demand) which is based on delta T. 2) The first method (varying flow based on # of coils calling for flow requires pressure dependent control valves somewhere after the pump that control either the entire system or multiple coils and then a balancing valve before each coil. The pump for the first method, is not variable speed. The second method requires pressure dependent control valves before each coil and if each pressure dependent control valve is not three-way, a variable speed pump is needed to vary the flow.

I do not have an example at the moment, but I will try and create a realistic scenario which we can work with later this week. I am trying to understand and address why designers (and specifically other engineers in my office) use 3 to 5 psi of a pressure drop on a valve.

 

[LittleInch]

You really need to sketch this out as it a bit confusing.

In one sentence you say you will add control to allow the pump to modulate ( controlling on what exactly?) and in the next sentence that you want to add control valves for the re-heat coils. Be careful they don't end up fighting each other.

Anything you do will change the network flows and pressures so be careful about messing with only one bit of the system.

You say your system doesn't have a control valve but has "fixed flow". So how does this work? manual balancing valves?

Anyway, the only thing you can do is look at each system and figure out what is the pressure drop if you close the valve, i.e. no flow.

This is your max pressure drop available. However then as you increase flow your piece of equipment will generate its own pressure drop, so work out what this is at your max desired flowrate through your AHU or whatever. Subtract this from your max pressure drop and this becomes your min pressure drop at max flow available to the valve. This will set your max CV, however most control valves only like to operate about 25% to 75% open so your actual valve CV will need to be a bit bigger.

but first you really need to figure out what it is you're ultimately trying to control ( temperature?) and how are you doing that - controlling water flow into the AHU with a fixed fan speed by a control valve or some other means?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[HVAC-Novice]

CV and control valves are not mutually exclusive. CV systems typically use 3-way valve that bypass flow around the coil/device.
The only way CV without valve would work if it is only one zone and the pump cycles or modulates to control. Or you have an application that continually heats/cools fully.

Except pump size and energy, nothing really stops you from using a dp of 20 psi to size your control valve. But the reason to go lower (but not too low) is to save pump power.

300:1 is ridiculous. I bet the actuator isn't able to make such small steps. You are operating on a 0-10 0r 2-10V signal. There is more voltage fluctuation than 1/300. Even if it could, there is no way to see a difference between position 197/300 and 198/300. Most PICV come with actuator and whatever they are able to do is fine. That is like you want to know the best time to plant your potatoes, but you insist on a clock accurate to 1/0000,0000,0000 seconds.

There is nothing stopping you from using PICV on any 2-way application. In fact, I can't think of a situation where they wouldn't be better. Bonus points for the electronic ones where you can monitor actual flow.