On-Demand Hot Water Circulation

Users who are viewing this thread

John Gayewski

In the Trades
Messages
4,727
Reaction score
1,517
Points
113
Location
Iowa
The numbers differ too much. It's not that. I think i found it, but I gotta confirm something. I think it has to do with the size difference of PEX and copper and iron fittings. Although pex does not have as many fittings it does not make up for it in friction loss. I got to do some more calculating and make a couple calls.

I use a different tool for figuring more precisely which I hadn't consulted in a while. Comparing different pipe types make an extremely large difference. I can't remember where I got the smaller multipler for pex, but it's dead wrong.
 

Kabra

Member
Messages
58
Reaction score
1
Points
8
Location
Colorado
For those that followed this thread I thought I would follow up and let you know where things have landed in my effort to find a sensible sweet between fast hot water delivery and not damaging my plumbing with excessive flow rates and turbulence. If you recall I used two Taco 006e3 pumps in series with great results but it wasn't clear what the flow rates were. Recall that I use this circulation system as an on-demand hot water system as the pumps only run to purge the cold water from the pipes.

I've got it all figured out now and the sweet spot is three of those pumps in series. The velocity is not at all off the charts as some may think. If I were to replace these three pumps with a single pump, I would us a Taco 0013-MSSF2-IFC which has three speeds, and based on the performance curve I would use either the medium or high speed due to the amount of head.

With all three 006e3 pumps set at max I'm getting in the range of 27' to 30' of head with estimated flow in the range of 3.5gpm - 4.5gpm and it's almost all 3/4" plumbing. Note the pressure gauges before and after the pumps so I know the actual dynamic head of the system.

And yes I am aware all those outlets are very near all that plumbing and I plan to do something about that. But also keep in mind that all of those outlets are not hot except when the pumps run which is for 20 - 25 seconds. so 99.8% of the time the outlets have no power. In addition the whole system is of course on a GFI circuit.

Each of the pumps add about 4 1/4 psi of head so right around 13 total. Adding a second pump darn near doubled the hot water delivery time. Adding a third pump improved it by another 20% or so. At the push of a button I get hot water in 15 - 20 seconds depending on the length of the loop it's being delivered to. The hot water supply branches out to four different quadrants.

You may be wondering how I determined the flow rate of 3.5gpm - 4.5gpm. First I put a bucket under a couple of the faucets and measured how much water comes out before it's hot. 1 to 1.5 gallons depending on the loop. The long loop (1.5g) for example takes about 20 seconds to deliver hot water so that's where the 4.5 gpm estimate originates. I also used the Taco SizeRight circulator sizing tool with head from valves and fittings estimates of 25% to 60% which further validates the estimated GPM range.

IMG_1590.jpeg
 

John Gayewski

In the Trades
Messages
4,727
Reaction score
1,517
Points
113
Location
Iowa
For those that followed this thread I thought I would follow up and let you know where things have landed in my effort to find a sensible sweet between fast hot water delivery and not damaging my plumbing with excessive flow rates and turbulence. If you recall I used two Taco 006e3 pumps in series with great results but it wasn't clear what the flow rates were. Recall that I use this circulation system as an on-demand hot water system as the pumps only run to purge the cold water from the pipes.

I've got it all figured out now and the sweet spot is three of those pumps in series. The velocity is not at all off the charts as some may think. If I were to replace these three pumps with a single pump, I would us a Taco 0013-MSSF2-IFC which has three speeds, and based on the performance curve I would use either the medium or high speed due to the amount of head.

With all three 006e3 pumps set at max I'm getting in the range of 27' to 30' of head with estimated flow in the range of 3.5gpm - 4.5gpm and it's almost all 3/4" plumbing. Note the pressure gauges before and after the pumps so I know the actual dynamic head of the system.

And yes I am aware all those outlets are very near all that plumbing and I plan to do something about that. But also keep in mind that all of those outlets are not hot except when the pumps run which is for 20 - 25 seconds. so 99.8% of the time the outlets have no power. In addition the whole system is of course on a GFI circuit.

Each of the pumps add about 4 1/4 psi of head so right around 13 total. Adding a second pump darn near doubled the hot water delivery time. Adding a third pump improved it by another 20% or so. At the push of a button I get hot water in 15 - 20 seconds depending on the length of the loop it's being delivered to. The hot water supply branches out to four different quadrants.

You may be wondering how I determined the flow rate of 3.5gpm - 4.5gpm. First I put a bucket under a couple of the faucets and measured how much water comes out before it's hot. 1 to 1.5 gallons depending on the loop. The long loop (1.5g) for example takes about 20 seconds to deliver hot water so that's where the 4.5 gpm estimate originates. I also used the Taco SizeRight circulator sizing tool with head from valves and fittings estimates of 25% to 60% which further validates the estimated GPM range.

View attachment 97528
Still think you'd have much better results with a different system.

My hot water is so fast that when you use one hand to turn it on and then feel the water it's hot. Zero pump.

The problem is your system is so un-elegant that it borders on absurd (the taco guy's comments not withstanding). It's a silly way to get bad results.

I've built four systems recently in different houses that all work with no pump. We installed a pump any way becuse it's necessary to guarantee a sanitary water system, but they all give instant (and I mean instant) hot water.
 

Kabra

Member
Messages
58
Reaction score
1
Points
8
Location
Colorado
Every time you add a pump reach pump individually gets less results.
That's right because as velocity goes up head increases disproportionately. That's why 3.5 - 4.5 GPM is the sweet spot for 3/4 pex. Great performance and no turbulence concerns. And like I said in my last post if I didn't already have these pumps I would use a Taco 0013 3-speed pump that handles up to 33' of head in the sub 5gpm range.

The problem is your system is so un-elegant that it borders on absurd (the taco guy's comments not withstanding). It's a silly way to get bad results.

I'm chuckling LOL! Sorry you don't like it but the results are superb: Hot water fast, where I want when I want. Admittedly the three pumps is a bit over the top and I only use them because I had them and they do the job well. It's a very simple switch to a single pump if I ever choose to do that. Other than that the hot water branches out to four quadrants of the house so I ran a separate return line to each and that does add a bit more plumbing and logic. I think that's pretty elegant though because I don't circulate hot water all over a 3400sf home to places it's not needed. Just hot water fast, where I want when I want. How is that not elegant?

It's a fantastic solution and you just don't seem to understand or appreciate its applicability and probably not aware of the interest in this type of system. There are quite a few companies that promote this approach and here are just a few that come to mind including Taco:

https://www.tacocomfort.com/documents/FileLibrary/REV3_TacoGenie_Catalog_100-15_072619.pdf

https://smartrecirculationcontrol.com

https://fasterhotwater.com

https://waterquickpro.com

https://gothotwater.com
 

Fitter30

Well-Known Member
Messages
4,864
Reaction score
947
Points
113
Location
Peace valley missouri
Two pumps in series doubles the head same flow
Two pumps in parallel doubles the flow same head
Three pumps time to buy another pump so you can have a pump for each line
 

Kabra

Member
Messages
58
Reaction score
1
Points
8
Location
Colorado
Two pumps in series doubles the head same flow
Two pumps in parallel doubles the flow same head
Three pumps time to buy another pump so you can have a pump for each line
In this application head is what was needed to increase flow and that's why the pumps are in series. One pump per loop is what I had and the 13' of head per pump isn't enough to get the flow that I wanted. At the estimated 4.5gpm the differential pressure is 27' to 30' of head. This setup works beautifully.
 

John Gayewski

In the Trades
Messages
4,727
Reaction score
1,517
Points
113
Location
Iowa
That's right because as velocity goes up head increases disproportionately. That's why 3.5 - 4.5 GPM is the sweet spot for 3/4 pex. Great performance and no turbulence concerns. And like I said in my last post if I didn't already have these pumps I would use a Taco 0013 3-speed pump that handles up to 33' of head in the sub 5gpm range.



I'm chuckling LOL! Sorry you don't like it but the results are superb: Hot water fast, where I want when I want. Admittedly the three pumps is a bit over the top and I only use them because I had them and they do the job well. It's a very simple switch to a single pump if I ever choose to do that. Other than that the hot water branches out to four quadrants of the house so I ran a separate return line to each and that does add a bit more plumbing and logic. I think that's pretty elegant though because I don't circulate hot water all over a 3400sf home to places it's not needed. Just hot water fast, where I want when I want. How is that not elegant?

It's a fantastic solution and you just don't seem to understand or appreciate its applicability and probably not aware of the interest in this type of system. There are quite a few companies that promote this approach and here are just a few that come to mind including Taco:

https://www.tacocomfort.com/documents/FileLibrary/REV3_TacoGenie_Catalog_100-15_072619.pdf

https://smartrecirculationcontrol.com

https://fasterhotwater.com

https://waterquickpro.com

https://gothotwater.com
Elegance is simple beauty. If you like the supermodel look with tons of makeup and a feathered creation atop the head while wearing something ridiculous then sure. You could have hot water much faster with less. Which is more elegant.
 

Kabra

Member
Messages
58
Reaction score
1
Points
8
Location
Colorado
Elegance is simple beauty. If you like the supermodel look with tons of makeup and a feathered creation atop the head while wearing something ridiculous then sure. You could have hot water much faster with less. Which is more elegant.

Seriously? That analogy is vastly more inelegant than my hot water solution, and to use your word, it's downright absurd.

I built a system that met a defined requirement. I did it because I couldn't find a plumber that could do it. I didn't have the knowledge to properly engineer it and I came to this forum to get professional advice and technical knowledge so that I could engineer it correctly. What I got was nonsense that what I was doing couldn't possibly work, I would destroy my pumps and my plumbing, and the numbers were off the chart. Oh and somehow you know what I need and want better than me and I should put in a system that you like better.

I never did get the engineering advice I was hoping for or even some semblance of going through a disciplined process of determining velocity. All I got was it's off the charts even though it's not. Gotta do the math! I was able to extract enough tidbits of info within the nonsense to figure out on my own what I needed to do the math. The system does work and it works beautifully. I'm not destroying my pumps, the numbers are not off the charts, and I'm not ruining my plumbing. With this more complete understanding these changes give me the performance I had originally hoped for and I can call it done.

But now the problem is it's ugly and it's not the kind of system you think I should want. You always know better somehow even though you've been wrong in everything you said from the start.

I do appreciate the time you took and that you tried to be helpful. I hope you think about this though and maybe remove the ego so your efforts to be helpful aren't a waste of time.

Thank you again.
 

John Gayewski

In the Trades
Messages
4,727
Reaction score
1,517
Points
113
Location
Iowa
Getting free of the top of my head advise is about all your gonna get. No plumber would do it because it's ridiculous. When it comes time to properly engineer a job that I'm being paid for I do it. A few advice sir is what it is. Your welcome for guiding you to the system you like. But to say it's elegant is just silly.

Yes un-elegant was used to reflect the issue. I was gonna put the correct word in parentheses, but I thought you get the joke.

In three years when this system is no longer tickling your curiosity you'll wish you have something that won't break.
 
Top