piping and flow question

Reply

  #1  
Old 02-06-13, 08:23 PM
B
Member
Thread Starter
Join Date: Jun 2005
Location: ontario canada
Posts: 381
Received 0 Votes on 0 Posts
piping and flow question

here i am again asking you experts for more help

three water to air exchangers in series..at times the return water is a little low in temp

i know each exchager will slow the gpm down and the feed temp to each exchanger after teh first one will be getting lower temps feed..so this is what is happening when all three are using heat..domestic water is a plate eschanger (i don't knwo what size) it draws the temp from 170 to 150 (when the OWB is at its "kick in temp" (low setting) then at teh furnace plenum it drops from 150 to 140 and then at the garage exchanger it drops from 140 to 130 for the return temp (all these are ten degrees higher when the OWB reaches its set point of 180
now if we said each exchanger slowed the flow 2 gpm we would slow the water by 6 gpm (?)

so if i now hooked these three exchangers in paralell i know each one would be recieving water at a lower gpm and take less heat our of the water causing the blowers to run longer but would the gpm in the return line be faster or the same...what i am asking is could i get a higher return temp but the blowers in the home would run a little longer.

or do i simply need to push more gpm (perhaps with another pump hooked in paralell?)

this is not my system but thought i would ask over here where the real plumbing experts reside
 
Sponsored Links
  #2  
Old 02-06-13, 09:18 PM
NJT's Avatar
NJT
NJT is offline
Member
Join Date: Oct 2006
Location: USA
Posts: 23,539
Received 3 Votes on 2 Posts
Air handlers should always be piped in parallel for the exact reason that you say, because the ones 'downstream' will be getting cooler water when piped in series and probably won't perform to specification.

If the flow were increased, one could minimize this problem, but it would probably mean that one would end up with a flow VELOCITY that was too high for the pipe size used.

Heating system flow velocity should be targeted during design at around 4 feet second. For 3/4" pipe this works out to approximately 4 GPM and for 1" about 8 GPM (APPROXIMATELY).

Better to pipe in parallel. The pump would still need to be sized so that it flowed enough for all three air handlers, and the piping feeding the three in parallel would need to also be sized large enough to support the total flow of all three.

For example, if each air handler required a flow of 4 GPM for it's rated output, the main feeding the three in parallel would need to flow 12 GPM. ( 1-1/4" pipe would do it )
 
  #3  
Old 02-06-13, 09:37 PM
B
Member
Thread Starter
Join Date: Jun 2005
Location: ontario canada
Posts: 381
Received 0 Votes on 0 Posts
thanks NJ now one more..if each handler were to say slow the gpm by 2 in series i i owuld guess that as the water slowed from one to the other it would also decrease how much it slowed the velocity becasue the feed into the nex handler is line would be fed slower...but jsut for kicks lets say in weried they all lost 2gpm for a total of 6gpm....now iped in paralell each loop tot he air handler would lose 2 (again i am sure it would be less due to the fact the water is moving less gpm) anyway lets say it is 2gpm each do you still add them up as in series or not (are we still losing 6 or only 2?..i am lost with thi
 
  #4  
Old 02-07-13, 07:40 AM
NJT's Avatar
NJT
NJT is offline
Member
Join Date: Oct 2006
Location: USA
Posts: 23,539
Received 3 Votes on 2 Posts
One thing you have to realize is that the flow does not "...slow from one to the other..."

The flow is the same through ALL of the elements when piped in series. It can't NOT be... how can you have say 6 GPM going in, and only 2 GPM coming out? Where is the other 4 GPM going?

So remember that there is only ONE flow, ONE velocity in a series piped system. It is the TOTAL resistance of all the piping in series that determines what a particular pump is able to flow in a piping system.

Flow being the same through all elements in a series system, any differences in BTU output will be due solely to the cooler temperatures being fed to downstream units.

In a PARALLEL system, the flow CAN be different through each 'branch' of the parallel system, but the TOTAL going in to all of them is the same as the TOTAL coming out of each system.

In parallel system, since they will all be receiving the same temperature water, the BTU output of all of them will be the same... (assuming equal size heat exchanger coils and equal blower speed).



The VELOCITY of that flow depends on the size of the pipe. For example, 4 GPM in a 3/4" pipe will move FASTER than 4 GPM in a 1"
 
  #5  
Old 02-07-13, 04:26 PM
B
Member
Thread Starter
Join Date: Jun 2005
Location: ontario canada
Posts: 381
Received 0 Votes on 0 Posts
i guess what i am really asking is in series then you are telling me all the rads in the series would be getting the same gpm, and if the pump was capable of 8 gpm (with no rads)and each rad (3) slowed the flow by 2 gpm then in fact each appliance would be recieving 2 gpm and at each rad after the first one the water would be cooler so if the total heat loss in this situation was 40 degrees we would have (if starting temps were 180) 140 degree water going back to the OWB at 2 gpm

now if we hooked the three rads in paralell, would the speed of the return water be increased (gpm) and would you still figure in 6 gpm flow reduction or would you figure in only 2?
 
  #6  
Old 02-07-13, 04:54 PM
G
Member
Join Date: Oct 2008
Location: USA
Posts: 2,994
Received 11 Votes on 11 Posts
...now if we hooked the three rads in paralell, would the speed of the return water be increased (gpm) and would you still figure in 6 gpm flow reduction or would you figure in only 2?
With the three rads in parallel, the flow will divide in thirds. Compared to three radiators in series, the total combined flow will increase slightly because the pump will supply a higher flow rate against the lower resistance presented by three rads in parallel.

...would the speed of the return water be increased?
I understand your question, but it is improperly stated and implies a misunderstanding. Flow velocity and flow rate have specific meanings. Flow speed doesn't. Best to google "centrifugal pump curves."
 
  #7  
Old 02-07-13, 05:40 PM
NJT's Avatar
NJT
NJT is offline
Member
Join Date: Oct 2006
Location: USA
Posts: 23,539
Received 3 Votes on 2 Posts
all the rads in the series would be getting the same gpm
Yes, correct. If you put X GPM into one end of a pipe, you must have X GPM coming out the other end.

if the pump was capable of 8 gpm (with no rads)and each rad (3) slowed the flow by 2 gpm then in fact each appliance would be receiving 2 gpm
Yes, sort of... but as Gil has said, you are expressing this improperly.

Each rad does not "slow the flow by 2 GPM".

Each rad ADDS RESISTANCE TO FLOW in an amount that when added up would cause the pump to only flow 2 GPM against that RESISTANCE (properly called HEAD by the way) TO FLOW that the system presents to the hypothetical pump.

at each rad after the first one the water would be cooler so if the total heat loss
To be 'uber-technical' I would say no, not correct, but only because of the terms you are using.

HEAT is measured in BTUs, and is NOT the same term as TEMPERATURE which is measured in degrees.

For your statement to be exactly correct, you would have had to say:

"...at each rad after the first one the water would be cooler so if the total TEMPERATURE loss..."

I realize that this is 'nit-picking' and I'm sorry about that, but I think it is important to use correct terms and concepts.

in this situation was 40 degrees we would have (if starting temps were 180) 140 degree water going back to the OWB at 2 gpm
Yes, exactly... 2 GPM in, 2 GPM out. 40 change in temperature (also called DELTA T) ... 180 in, 140 out.

So, in spite of the semantics, we arrive at the same point.


the flow will divide in thirds
But only if all the piping to each air handler is exactly the same, and each air handler has the same coil. Let's instead say that the flow will divide 'roughly' or 'approximately' into thirds.

now if we hooked the three rads in parallel, would the speed of the return water be increased (gpm) and
Also as Gil has pointed out, you are confusing the terms.

GPM = FLOW VOLUME

SPEED (in feet per second) = FLOW VELOCITY

The answer to the question you are (sort of) asking is:

YES, as long as the pipe leading to the point where the flow splits into three was capable of supplying enough flow, you would have MORE flow in each rad.

It would not be THREE TIMES as much, because the actual relationship is based on some complicated math, but it would be more than three piped in series.

would you still figure in 6 gpm flow reduction or would you figure in only 2?
No, you would not have the same RESISTANCE TO FLOW (aka HEAD) presented to the pump in this case.

The three circuits would appear to the pump as one LARGER pipe, and would present that pump with a LESS RESISTIVE path, thus would be capable of flowing MORE water in the COMMON part of the piping, up to the point where the flow splits into three.

When the flow splits into three, each circuit would take APPROXIMATELY one third of the TOTAL flow.

These concepts are not the easiest thing in the world to grasp, but if you don't think about them too hard, you will get it...
 
  #8  
Old 02-07-13, 05:41 PM
B
Member
Thread Starter
Join Date: Jun 2005
Location: ontario canada
Posts: 381
Received 0 Votes on 0 Posts
let me see if i can workd it differently (im not good at this as you can see)

in the "three in series" i understand that each rad would be getting water at a lower temp and all water would be moving at 2 gpm in the example

what i am asking is if in paralell (i think) each rad would be (as long as the resistance was the same) they would each be gettting one third of the "pressure created by the pump vrs the first example" so i am thinking the velocity of the water is less and so there will be slightly less resistance (head pressure)in the rad and also this will reduce the available btu per hour? i am wondering, now that all three rads are getting the same temp water, will the return water be warmer

the whole i dea of this is when all three rads are calling for heat the return water temps are too low i understand to fix this would be take less btu out or increase the btu delivery. the return temps need to be raised about 5 degrees so i was wondering in going in paralell would deliver water at a lower flow to each rad but all three now would be getting equally hot water and resistance would be lower due the the lower flow (velocity or gpm what ever it is called) would this in fact take alittle less btu per hour and offer hotter return water ?

or is a larger pump with an increased gpm delivery at the same head needed?
 
  #9  
Old 02-07-13, 05:59 PM
G
Member
Join Date: Oct 2008
Location: USA
Posts: 2,994
Received 11 Votes on 11 Posts
what i am asking is if in paralell (i think) each rad would be (as long as the resistance was the same) they would each be gettting one third of the "pressure created by the pump vrs the first example"
There are at least three misunderstandings revealed in your above quote. I recommend that you hit the books and try to learn about the theory of flow of fluid through pipes and pumps. It is very difficult to try to explan all this over the internet. Sorry for being so blunt.
 
  #10  
Old 02-07-13, 06:06 PM
NJT's Avatar
NJT
NJT is offline
Member
Join Date: Oct 2006
Location: USA
Posts: 23,539
Received 3 Votes on 2 Posts
I know what you are thinking, and I'm trying to figure a way to explain it...

I could do this easily in person in front of a white board... not so easy this way!

Any given pump has a 'pump curve' (Gil mentioned this too...)

Of course if you plug the discharge of the pump, you will get no flow.

If you leave the discharge completely open, you will have maximum flow.

At all points between these two endpoints, the pump flow will be different depending on how much HEAD it is pumping into.

It is not a straight line, I think you understand that, but a CURVE.

In any case, starting at the plugged discharge, as you start to allow flow, and decrease the head that the pump is pumping into, the flow will increase along it's curve.

In your SERIES example, the HEAD that the three rads present to the pump is ADDED together, resulting in a HIGH HEAD situation, and the pump will pump a lot less GPM.

The less GPM the pump is moving, the SLOWER the flow in the pipe.

The SLOWER the flow in the pipe, the more TIME the rads have to remove BTU from that flow (because the water remains in the pipe longer), and the COOLER the water becomes.

In your PARALLEL example, even though it isn't TECHNICALLY accurate, we will for this example say that the HEAD will be reduced by a factor of three. Again, it will not be exactly thirds because of all that complicated math, but just for the CONCEPT we will say that it is.

Because by piping in parallel, we are reducing the HEAD by dividing the flow, the OPERATING POINT of the pump will move to the right along it's curve and will be capable of pumping HIGHER GPM.

This will result in each rad receiving MORE FLOW, MORE GPM than in the series example.

With MORE GPM, the water will not cool as much at the exit of the rad because it has not spent as much time IN the rad. The water will be HOTTER all through the rad. This means that each rad will output MORE BTU, and the return water to the boiler will be HOTTER.

Does this help?
 
  #11  
Old 02-07-13, 06:07 PM
NJT's Avatar
NJT
NJT is offline
Member
Join Date: Oct 2006
Location: USA
Posts: 23,539
Received 3 Votes on 2 Posts
How did I do Doug? in 1000 words or less?
 
  #12  
Old 02-07-13, 06:39 PM
B
Member
Thread Starter
Join Date: Jun 2005
Location: ontario canada
Posts: 381
Received 0 Votes on 0 Posts
thanks NJ that is what i was wondering (i know i dont get velocity and btu thing right but i understand your explanation) i thank you all for taking the time to try and understand my question

i think i do understand the velocity thing. 8 gpm is 8 gpm in a 1 inch pipe or a 3/4 inch pipe but the velicity will change to much higher int he 3/4? thats where i get the "slower faster" thing

also i understand that if a pump could pump 8 gpm in a one inch line at its maximum head for that length of pipe it would pump less in a 3/4 line the same length due to "head pressure" and "velocity would increase? (am i on the right track?)

from reading in here and on the net i am getting the hang of it (i think) but i could not find (probably due to my wording) the answer to my question that you ahve been able to answer. and that seems to be that in paralell in this case would rais the return temps to the boiler
 
  #13  
Old 02-08-13, 08:17 AM
NJT's Avatar
NJT
NJT is offline
Member
Join Date: Oct 2006
Location: USA
Posts: 23,539
Received 3 Votes on 2 Posts
Yes, I do believe you have the concepts correct.

I went the extra steps to explain because I believe in teaching a guy to fish rather than handing him a fish sandwich! Enjoy your meal!
 
  #14  
Old 02-08-13, 11:32 AM
B
Member
Thread Starter
Join Date: Jun 2005
Location: ontario canada
Posts: 381
Received 0 Votes on 0 Posts
thanks again and i know you don't want to hear this but....i'lll be back..lol
 
  #15  
Old 02-08-13, 04:59 PM
NJT's Avatar
NJT
NJT is offline
Member
Join Date: Oct 2006
Location: USA
Posts: 23,539
Received 3 Votes on 2 Posts
No worries billie_boy, we are here for ya!
 
  #16  
Old 07-02-13, 04:37 PM
PerryNing's Avatar
Member
Join Date: Jul 2013
Location: United States
Posts: 1
Received 0 Votes on 0 Posts
Hi Billie, If you still thinking about this topic, I'd summarize what NJ said with a little twist:
1) Paralleled is the way to go.
2) Your flow will increase (each rad will be more than the 3 in series).
3) Put a throttle valve on each to reduce the flow as you desired.
4) Your returning water will be at a high temperature than before.
5) You can pass more heat to the radiators if not throttled.
Thanks, Perry
 
Reply
Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Thread Tools
Search this Thread
 
Ask a Question
Question Title:
Description:
Your question will be posted in: