Calculating head in parellel piped systems

Andrew Hagen_4

The flow (and dT) in each circuit of parallel piping systems will naturally balance so that the head loss in each circuit is the same. To truly find the flow rate in each loop, you have to write a head loss equation for each and solve them simultaneously.

Since that's not really practical to do by hand, I recommend purchasing software that does it for you. John Siegenthaler's HDS is a great tool.

If you have circuit setters, you can calculate the head loss for the most restrictive loop and then set the others to have the same head loss at the flow rate each requires. The unfortunate part of having one loop that is much more restrictive than the rest is that the circulator is sized based on that loop.

Jim_64

Calculating head in parallel piped systems (Brad White)

Brad, or anyone else I was wondering about the formula to calculate the head in my parallel piped ceiling radiant system. IS it the longest loop plus supply, and return piping plus 50% for fittings? Trying to see if I can down size my B&g 100 HV.

Presently everything works fine with a 15* steady state DT at design temps. My head calculations if done as stated above put me near the cut off for the B&G 100 HV at 11' of head. Obviously that can't be I'm pumping what could I be missing.

Details: Longest radiant loop 250' 3/8" copper tubing

60' 1 1/2" supply, and 60' 1 1/4" return.

50% should be ample for fittings 3 90's 1 45 in the return, and 2 45's in the supply.

1 taco paneltrol mix valve from the 50's

Simple bypass piping at the boiler which is a WM cgm 8

Just looking for a pump to fit the task with a little less power consumption.


Gordy

Brad White_202

Typically

it is the longest circuit, not the sum of all piping, I think you have that, plus fittings, valves, the boiler, strainers and other pressure losses; that is the only real way.

The "shorthand method" of feet of pipe/100 multiplied by 4 then adding 50% for fittings is just a ballpark estimate not based in reality (unless your system actually works out that way). The alternative shorthand method is feet of pipe divided by 100 and multiplied by 6 for the same result.

If you know your actual length circuit and pipe size, fitting count, valve count, presumptive flow rate and other losses such as valves of all types, we can get pretty close.

Write me tomorrow to remind me and I can send you a spreadsheet I developed to speed my work along. All you need do is enter your flow rate per segment and the quantities of fittings and valves and pipe per size and type... I can cover that with you, but tomorrow.

Brad

Mark Hunt_6

Question

How old are the friction charts you use? When was the last time their accuracy was tested?

Brad White_202

Based on

Cameron's Hydraulic Data which is founded in the Darcy-Weissbach formula, older than you, Gil

Start with new pipe and add a factor if old pipe though.

Mark Custis

Head

I always install too many valves and to many "t"s.

I bought an eight point data logger and still can not SEE the flow, Gill's ghost.

Mark Custis

Head again

"t"s allow measuring devices to be installed, (as well as branch circuits, and forgotten future uses).

The valves keep your feet dry.

Simple pressure gauges will tell a bunch about a lot of stuff.

I am in lust with ECM circs. JUST enough every time. Gotta "SEE" the flow though. Ask Gill or Dan.

bob_46

See

It's in your mind's eye. Which assumes a prerequisite : )

Mark Eatherton

What if..

I had 400 feet of 1/2" and 350 feet of 3/8", do I still do the longest run thingy??? :-)

Where does flow fit into the equation?

What about fluid viscosity?

Obviously, you assumed all of Gordy's circuits were of the same bore, which is not always the case, and assumed that the flow through al circuits were the same, which is not always the case. In any case, you choose the highest pressure drop circuit plus distribution piping losses. At least that's what I've always been taught and have taught.

Even more disturbing, what happens in these circuits that are significantly shorter than the highest pressure drop circuit if there is a substantial difference in PD between highest PD circuit and lowest PD circuit?

Thanks for all you do Brad. Just want to make sure we don't misguide some of these young wet heads in the process of iteration we go through in these processes :-)

Gordy, for a more accurate picture of what's going on in your system, if you will put gauges across the circulator, you will be able to see exactly what kind of head the circulator is generating, and using the manufacturers performance chart, be able to predict exactly how many GPM's you are moving and what kind of resistance you have to overcome. Icemaker tap saddle valves can make the task quick and easy. You might be suprised what you "See"... Let us know your findings.

Mr Carlson Ghost, with all due respect, Gil Carlson had nothing to hide. What are you hiding from? I have little respect for people who hide their real identity. No cajones.
ME

Brad White_203

\"The Longest Run\"

Thanks, Mark- I should have been more explicit: The "longest run" should really be expressed as "that which has the largest pressure drop", rather than the Day One Assumption of "all piping in the system".

Naturally, any circuit of many in a system is the candidate for that, as you say a function of bore size, viscosity (fluid mixtures such as glycols and/or temperature).

Finding that most restrictive circuit can be an iterative process as you note. It may well be a larger bore circuit with lower flow than others, but many more fittings thrown in. It could also be a long circuit with few fittings but much flow. It is good that you pointed this out.

Clarity is key and sometimes is the first casualty of brevity

An ideal system is one where all circuits have an equal pressure drop. Such systems are common in places such as Loch Ness, Scotland, Area 51 in Nevada, and Cloud Cuckoo Land for example.

chemeclimber_2

Pressure Drop Calculation Attached

Check out this pressure drop calc. Enter all of your data in the yellow highlighted cells and the pressure drop will appear at the bottom. This is for turbulent flow, however; it will be real close for laminar flow as well. This should get you in the ballpark

chemeclimber_2

Pressure Drop Calculation Attached

Check out this pressure drop calc. Enter all of your data in the yellow highlighted cells and the pressure drop will appear at the bottom. This is for turbulent flow, however; it will be real close for laminar flow as well. This should get you in the ballpark

Gordy

More details

Mark what I have is 22 loops 3 of which are 1/2" tubing the balance is 3/8" tubing. All loops have the old style balance valves. Loop lengths range from 250' to 160' including leaders.

Supply and return piping has equivelent lengths of 133' 1 1/2" and 47' of 1 1/4" a panatrol mix valve ?? on pressure drop.

I have done the pressure gauge on either side of the circ.
except one was right at circ inlet the other was down stream after boiler and mixer on supply. This is were some existing T's lent themselves to the speriment.

I was getting a pressure drop of 5 psi which indicates 11.55 feet of head when multiplied by 2.31. Whether this is accurate I don't know but the other version of calcs about the same 11 feet of head range.

This puts me at around 15 gpm on the B&G 100 HV graph at the upper limits. 15gpm kinda tells me I'm over pumping for the btu's the system needs for the heat loss of my house which is 67,000 btus. T

The boiler however is a WM cgm7 which has an IBR of 150'000 btu's. So all indications say that the guy who installed the boiler before I inherited the system got the pumping right for that size boiler but missed the boiler sizing by double.

As for Andrews comment on the parallel piping arrangement being self balancing I'm not certain. If I take the 15gpm the system SEES, and divide by the 22 loops then each loop would theoretically see .68 gpm not out of line. But is it true? Maybe if the return were piped in reverse return fashion this could be true. But my system is not piped reverse return.

Hey all is well the system works beautifully except for a boiler double the size needed, and a little excessive pumping from a 200 watt circ. Which I would like to tone down so I don't wear out the pipes. I can here a little velocity noise in the bedroom at night.

Just trying to make sure what I think I'm learning is right, and I could not think of a better place to find out.


Gordy

Andrew Hagen_4

Balancing

What I was trying to say was that the head loss is equal in each loop. For example, say they balance to 1-ft of head loss. At 1-ft of head loss, a 250-foot 3/8" loop will have a very low flow rate compared to a 150-foot 1/2" loop.

The head loss is equal in each loop, but the flow is different.

Jamie_5

Equal pressure drop

What Andrew said is correct, the pressure drop across each circuit will be the same. This does not mean the flow will be the same. Pressure drop in this case is the analogue of voltage drop in a parallel electrical circuit; both are measures of energy used moving the substance across the branch, fluid and electrons, respectively. Flow is analogous to current, a measure of the amount of fluid or electrons moved per unit of time.

Brad White_202

Andrew

is correct, I agree. Things will settle to a common pressure drop but at different flows. Such is the goal of "self-balancing" systems and why loop lengths of a given diameter on a common manifold want to be the same lengths or at least within 10 percent.

I knew what you meant, Andrew!

Mark Custis

Bob

I had a customer who said she could hear elctrons moving through thermostats.

I like using an electric meter to see electrons. Andrews point about delta P, is reinforced with Mark's tap valve quick look idea.

I guess I am getting old or growing up as I have given up guessing, I want to know.

Gordy, I did gloss over your question at first glance. If I had Mark's wisdom I would have sent a differant picture. Thanks for the note. BTW