Resistance To Flow Closed Loop System Curiosity

PGB1

Hello All!
I hope this note finds everyone well & enjoying today.

Where I worked, I was often cross-trade assigned to work with the steam-hydronic-chilled water people & pipe fitters, which I loved because I learned a lot. These older men (all of whom now would be well over 120 years of age) always installed long radius elbows or a pair of 45-degree bends in lieu of 90-degree elbows on open systems, such as steam heating, fuel and cooling tower water piping. This is how I piped out domestic water in our home.

For closed loops, such as hydronic (pumped and gravity) and chilled water loops, refrigerant piping, they'd use standard radius 90-degree turns.

The citation was that the resistance to flow (or equivalent straight feet of pipe) did not matter in a closed loop because the fluid that was being pushed was also being pulled by the pump. These systems were always with large pipe, generally 3 inch and above. Some were threaded, some groove joint and most welded.

During some remodeling, I'm going to re-work a portion of our closed loop, pumped hydronic heating system at home and wonder if what the "Dead Men" I knew said makes sense for small piping as well?

In case it is important for the decision:
Ours is a 2-pipe, not-reverse return, mix of cast iron radiators, fin tube baseboard style and a fan coil. Piping ranges from 1" down to 1/2 & mix of copper and black schedule 40. There are many monoflow tees & balancing valves in the system. The heating plant is a converted Bryant steam boiler from 1954 with 100,000 BTU/Hr input.

For my project, using 90's, pairs of 1/4-bends or even PEX-A with bend supports are all equally feasible & presentable. (For fun, I measured radii & found Upnor's PEX-A nylon bend supports had the largest bend radius, followed closely by their metal bend support, then the a copper 45-degree bend with a street 45. Long and standard radius 90's were both far behind the others.

Thanks for sharing your knowledge! It's just something that I've occasionally pondered & now have a chance to implement.

Paul

EBEBRATT-Ed

Flow is flow. Resistance is resistance. It doesn't matter if it's a closed loop or an open loop. In a closed loop the height of the system or building means=0. An open loop is different, the height or lift must be added to the pipe and fitting resistance. 2 45= the same resistance as a 90 in almost any of the different piping materials.

Refrigeration systems should always be long radius (especially the suction line & discharge lines). Pressure drop in the liquid line is more forgiving.

There are time where 2 45s are better than 1 90 but flow resistance is not one of those times.

The most important thing I see is that a lot of designers can't or won't or will not take the time to calculate the system resistance with any accuracy and take a big guess. This usually results in undersized or grossly oversized pumps

But then again I'm not 120 only 68 so I have a lot to learn LOL

Ironman

In a closed loop, the lift in the piping is canceled by “what goes up must come down”(Farris wheel effect), but that doesn’t change the fact of the resistance that the piping and fittings in the system add. Elevation is not a factor in a closed loop; the pump doesn’t “see” that. What it sees is the resistance to flow that the piping creates. Don’t confuse static head with dynamic head.

As long as you don’t use an inordinate amount of fittings, I wouldn’t be concerned about using two 45s vs. one 90. However, using pex crimp fittings should be avoided.

Also, keep in mind that all piping that’s on the same circulator must be proportionate in size: if you have large gravity piping, you can’t add smaller piping that would be sized for modern forced flow and expect to get proper flow in it. Water takes the path of least resistance.

Also, everything on a monoflo loop should be monoflo: you can’t mix piping methods.

Jamie Hall

It is quite true that there is much less resistance to flow in a long radius bend -- say a radius greater than 5 pipe diameters -- than in a standard elbow. But... first, the difference between two 45s and one 90 is minor. Not worth it (or maybe I'm lazy). Second, in most piping layouts the fittings are a relatively minor part of the head loss.

There are situations where a long radius is almost mandatory, however: if you are running close to the NPSH of a pump, and are anywhere close to the inlet -- say within 10 to 20 pipe diameters -- if you can use a long radius so much the better. The same dynamics applies if you want to get even remotely even flow splitting from a T if you are entering from the bull.

PGB1

Thank You All for taking time to reply & to educate. I appreciate your help.

What you explained, Eberatt-ed, gives helpful information & reminders. I appreciate the knowledge.

Way back in the dusty part of my brain, I remember learning about using long radius bends when I was in refrigeration school. When I installed mini duct cooling/humidifying/purifying in my house, I had plenty of room to use Type K with field bends, excepting a pair of tees for the reversing valve's bypass at the TXV.

Refrigeration School? Thanks to opportunities where I worked, I'm cross educated in the trades. Hack... err Jack... of many trades, master of one. I'm a journeyman plumber, but by no means a plumber. After the pipe fitting is done, I'm lost. I worked often with the hydronic & steam guys & have refrigeration licenses, but my only consistent refrigeration experience is with chillers & freeze tunnels using R-22, R-502 & R-717. Replace the compressor on the kitchen refrigerator? All day project for me- if I'm lucky.
In real life, I'm a master electrician & high voltage wireman, spending my career in utility, industrial with some commercial. I could not re-wire a house without wrecking way too much plaster. Those folks are true artists.

Thanks Jamie Hall for the tips & reminders. I vaguely remember learning about elbows & take offs near a pump, but the details have always eluded me so I'd have to ask a real trades person.

Your Ferris Wheel analogy very much will make remembering lift considerations in a closed loop easy. Simple & effective enough for even me to remember.

The resistance to flow comments you made immediately made me think of it this way: If I have a 1" pipe and suddenly reduce it to 1/2", there will be resistance to flow, although velocity will increase if the system were open. It's assumed an elbow will act in a similar way that a size reduction in straight pipe does.

Something occurred to me that might have made my original question void:
The inside diameter of a PEX-A expansion fitting is much smaller than that of a copper fitting. I measured a PEX-A brass, Sweat-to-PEX adapter at 0.370". The I.D. of a 1/2" copper street fitting was 0.534" That is a lot of change in area. I'd imagine the drop-in crimp ring fittings are even smaller, but don't have one to measure.

It seems that the difference between using PEX-A with bend supports & sweat adapters versus all copper, cuts reamed smooth, with pairs of 1/8 bends may be insignificant for my project. Copper tubing's larger inside diameter, but smaller fitting bend radius versus PEX's large radius bends, but smaller fitting and tube I.D..

I suppose if I absolutely had to have the least resistance to flow without increasing pipe diameter, soft copper with field bent elbows would win. My 1/2" copper tubing bender's radius nearly matches the radius of the Wirsbo nylon PEX-A bend support at 2.51" versus 2.48" for the bend support.

Is the small inside diameter and the many bends why in-floor heat with PEX often has it's own circulating pump? When PEX first got introduced in the United States, I had the opportunity to go to Upnor-Wirsbo's classes & learned it there. We received the Design Assistance Manual which I should dust off and re-read.

Ironman, you mentioned not using crimp ring fittings. I use expansion fittings because that's what I learned with. But I'm curious about your comment, as I've seen installations with it on heating systems. Do they work loose from expansion & contraction, similar to how Oetiker clamps do on chilled water installations, such as post-mix soda pop?

Thanks Again everyone for helping. I very much enjoy pipe fitting and look forward to this project.
Paul

Ironman

I said avoid crimp fittings because of their restriction, not that they couldn’t be used.

PGB1

Thanks for taking time to explain Ironman. I appreciate it.
Paul

hot_rod

I've read 8D, eight times the pipe diameter of a formed bend does not count as an ell.

EL charts show what to calculate for fittings.

I like the challenge and "look" of formed tube and pipe. An example of a fitting free, French plumber boiler installs

EBEBRATT-Ed

Anyone who dida job like @hot_rod posted above should win an award as an artist. It beautiful. It's a great job. I think is must have been installed with a micrometer LOL

In the real world customer are complaining about high prices that's why their is so much DIY (and I am not knocking DIY done right) and hack work.

Who could afford to pay someone to do a job like the one posted above???

hot_rod

A handful of guys, and gals posting some nice bending installs on instagram, not quite that elaborate however. It's more common with small refrigeration piping, saves some time, fittings and cost..

PGB1

That sure is beautiful pipe work that Bob posted. The planning alone must have been quite a challenge.

Thanks Bob for posting the chart of straight pipe equivalency. I was surprised to see that a long radius ell has a larger foot-equivalence than a standard radius ell in the half and 3/4" diameters, above which it reverses to less resistance as I expected to see for all sizes.

Jamie Hall

PGB1 said:

That sure is beautiful pipe work that Bob posted. The planning alone must have been quite a challenge.

Thanks Bob for posting the chart of straight pipe equivalency. I was surprised to see that a long radius ell has a larger foot-equivalence than a standard radius ell in the half and 3/4" diameters, above which it reverses to less resistance as I expected to see for all sizes.

That does seem odd -- until you remember that a long radius ell is a good deal longer from end to end than a short radius ell.

PGB1

Thanks Jamie. This makes sense. I wonder why the diameters over 3/4" don't have the problem. It would be interesting to figure out the math behind it and do the calculations.