Math problem

S Ebels

I have a primary loop flowing 8gpm @ 160* AWT. I have a secondary loop flowing 12 GPM into a 150,000 btu radiant slab load.

What temp will the supply for the secondary loop be seeing?

What return water temp will be headed back to the boiler?

My math challenged brain won't do this even with help from Pumping Away.

Jerry Boulanger_2

You should be able

to control the temperature of the water going into the secondary (radiant) loop to whatever your radiant design requires with whatever you are using to mix the temperature down. The temperature drop in the radiant loop should be 150,000/(500*12) = 25 degrees F. And you are taking 150,000 BTUH from a primary flow of 8 gpm, so the temperature drop required in the primary to serve that secondary load will be 150,000/(495*8) = about 38 degrees F. I'm assuming the fluid is water.

S Ebels

MMMMMMMMMMMMMmmmmmmmmm

Where do the 500 and the 495 numbers come from?

BTW, I ran this question past the design guy at the local wholesaler and he said it can't work. He told me the primary pump wouldn't be able to cirulate any water at all. DUH

Dave_22

I come up with a 25 degree temp drop on your secondary based on 150,000 and 12 GPM. It seems to me that you will have 8 GPM@160 of your primary getting mixed with 4 GPM@135 of your secondary return. This would give you a secondary mix temp of 151. The other 8 GPM would flow back to the primary @135 degrees.

Kal Row

did you even try to explain to him...

(had to be a "him" as a "her" would listen rather than insist on stupididy) - that if the primary enters and exits the secondary closer together than six secondary pipe diameters - then the loops can only effect each other temp'wise not flow'wise

caleffei has a nice tradeshow demostrator for this with pinwheel flow indicators - turn on one pump and the other pinwheels dont budge

Mark Eatherton1

Hint...

it's flow times temp... I'll tell more when I get back from the mountains today, unless someone else beats em to the punch. Gotta run.

ME

S Ebels

I've given up

trying to train this particular guy. He's "old school", to put it politely.

"Use the pump that comes with the boiler, run the boiler at 200*, the pump comes already installed why change where it's at, yada yada yada yada............."

Dave_22

Just in case some were wondering:
8gpm X 160 + 4gpm X 135
1280+ 536= 12X
1816 = 12X
151.3= X
Ahhhh- "Prim/Sec Made Easy" was worth it's weight in gold!!!!

Kal Row

bah! - i'm new school, just let tekmar do it, thinking hurts

the dead men had too much time on their hands - no tv, no internet...

S Ebels

But that's where I'm hung up.......

How do you determine the temp in the secondary loop in order to perform that calculation for both the supply and return temps?

Weezbo

*~/:)....

Try to conserve energy any way possible :)its the Now thing to do ......just got home from work what a great day bought a Prestige propane tIIO to day now i need to go reinforce the availability of triangle tubes modulating condensing unit and also ordered a gas model TIIO a little earlier in the day Thru Keller...I have some kinda luck today *~/:) to quote a smart guy ..."What goes Into a "T" must leave a "T".......... so, further de couple the snow melt with a large by pass and mix it down to like 37 or may be 41 degrees and let it circ on troll

DanHolohan

Questions:

What supply temperature do you need in your radiant?

How are you piping this? One-pipe or two-pipe? Whar sort of control valve are you using on the radiant?

Mike T., Swampeast MO

VERY confusing--I agree...

Unless the water between the tees to the secondary loop moves in reverse, I believe it's impossible to have more secondary flow than you have primary flow.

IF that reverse flow does happen, then in your example 4 gpm would be flowing from the secondary return to the secondary supply with the full primary 8 gpm from the primary to the secondary supply.

Again, assuming that goofy reversed flow condition, let's say your primary supply temp (NOT average--the max) is 170°. Your slab load is 150,000 so with 170° supply the delta-t across the loop would be 150,000 / (500 * 12) = 25° for a return temp of 155°. BUT WAIT! PROBLEM! You DON'T HAVE 12 gpm at your max supply temp because 4 gpm of that flow came from the secondary return!

So, your slab load must be satisfied by 8 gpm. 150,000 / (500 * 8) = 37.5° delta-t. Secondary return temp thus equals 170° - 37.5° = 132.5°

BUT WAIT! ANOTHER PROBLEM! Your supply temp is no longer your maximum supply temp because 4 gpm came from the secondary return. Your actual supply temp has dropped to (8 gpm * 170°) + (4 gpm * 132.5°) = (12 gpm * x)

1360 + 530 = 12 * x

1860 = 12 * x

x = 155° supply temp to the secondary.

To cross-check, lets take 12 gpm @ 155° and see if we wind up with a return temp of 132.5°.

Again, with 150,000 btu load on the slab @ 12 gpm the delta-t across the secondary is 25°.

155° - 25° = 130° NOT 132.5°--close, but there's STILL problems. Would have to dig out some old books to remember how to solve properly, but that difference still isn't the BIG problem. The BIG problem is that if you really need 170° water entering the secondary, you have to increase the primary temperature to compensate for the seconary return mixing back into the secondary supply.

--------------------------------------------

Is it actually possible for the water between the tees to reverse flow? I SERIOUSLY doubt it and it certainly won't happen if there's a check valve between. What I believe will happen is that the primary and secondary circulators will interact in a way that just gives 12 gpm through the primary loop!

If that happens, your solution is simple.

Secondary supply temp = primary supply temp.

Temp drop = 150,000 / (500 * 12) = 25°

Secondary return temp = secondary supply temp - 25°

Primary return temp = secondary return temp.

S Ebels

Thanks for chiming in here Dan

Would like to do one pipe or something like it. (see my post to Mike T above) Target temp is 110-120 out to the slab but doesn't have to be a closely regulated for this particular application. We just have to maintain a below grade gutter at anywhere above freezing to facilitate water flow during winter. I haven't decided how to control this yet either. The owner is leaning toward just a manual on/off or a timer to run the circs.

The wierd setup is because the tubing installer ran 1" pex 350 ft (round trip) to the manifold location and then has 12, 350 ft loops of 5/8" pex being fed off that. That being the case, I'm tapping the 1" feed into the high temp side of the system ala mini tube design. I would like to get the GPM's balanced out in such a way that I can use the flow from the loop circ to do my mixing for me by blending some return back into the high temp side.

Am I making any sense at all?

S Ebels

Mike

It is possible because I've done it before. Forget about a typical pri/sec piping arrangement and visualize an apparatus similar to Viessmann's low loss header in the middle of everything. Make sense?

And remember, this application doesn't call for close control of the "slab" temp. If it undershoots we elevate the temp on the main loop, if it overshoots it's just not a big deal. It's a barn. Not a get it perfect the first time situation.

Unknown

It does flow backwards

That's what this question is based on, as I read it.

Look at it this way, Mike. If just the secondary pump is running, it goes "backwards" between the tees.

If both pumps are running, and they flow exactly the same rate, no water flows between the tees.

If the primary pump moves more water than the secondary pump, it flows "frontward" between the tees.

If the primary pump moves less water than the secondary pump, it flows "backward" between the tees.

It has to.

Noel

DanHolohan

Just spent an hour with it

and I'm not able to do it this way, Steve. I think you should include a mixing valve.

Mike T., Swampeast MO

So, you say that this happens predictably and reliably? (Not disagreeing, it just seems strange.)

If so, then the math in my earlier message is reasonably accurate. Will work on the equations to nail it down, but it's been a while since I did that sort of algebra.

Unknown

yes, that's it

That is the confusion factor in determining the supply temperature. It's a blend.

Noel

Mike T., Swampeast MO

Here's the Equation and Solution

(That is if my high-school algebra didn't fail me.)

Desired Supply Temp to Secondary = 115°

12 gpm with 150,000 btu/hr load = 150,000/(500 * 12) = 25° delta-t in the SECONDARY loop.

The simple equation for computing combined flow:

(Port A Flow)*(Port A Temp) + (Port B Flow)*(Port B Temp) = (Port C Flow * Combined Flow Temp)

What we know:

Port A Flow where "A" is the water coming BACK into the secondary supply from the secondary return: 4 gpm

Port B Flow where "B" is the PRIMARY flow: 8 gpm

Port C Flow where "C" is the TOTAL SECONDARY flow: 12 gpm

Combined Flow Temp: 115° (your desired)

What we don't know:

Port A temp

Port B temp

BUT we know the temperature drop across the secondary (combined) flow loop. It is 25° as computed from the flow and load.

SO, Port A temp = Port B temp - 25°

NOW, we have enough information to solve:

4(x - 25) + 8x = 12 * 115

4(x - 25) + 8x = 1,380

[distribute 4 over (x - 25)]

4x - 100 + 8x = 1,380

4x + 8x = 1,380 + 100

4x + 8x = 1,480

12x = 1,480

x = 123.3° PRIMARY (BOILER) SUPPLY TEMPERATURE

SECONDARY SUPPLY TEMP = 115°

PRIMARY AND SECONDARY RETURN TEMP = 98.3°

----------------------------------------------------

Are you concerned with low return temps?

S Ebels

No

The real primary loop back in the mechanical room has a lot of flow and temp in it to absorb 8 gpm of 70* water coming back, if indeed it gets that cold. The system also has a motorized 3 way diverting valve that shuts off some of the system return if needed. This boiler (Rondomat) will live happily at 125 all day long.

Mike T., Swampeast MO

The "proof" of that equation seems to work out.

Using the simple flow formula:

(8gpm * 123.3°) + (4gpm * 98.3°) = (12gpm * x)

986.4 + 393.2 = 12x

1379.6 = 12x

x = 1379.6 / 12

x = 114.96°

Where "X" is the supply temp in the SECONDARY loop (assumed to be 115°).

Perhaps someone can verify that I haven't made a mistake (other than assuming 500# in the delta-t equation).

Jed_2

Now the fun begins

Those flows can certainly be achieved once you have your ^P pinned down. How would you achieve close proximity to those flow relationships? I understand the tolerance isn't tight in Steve's case. But, if it were, a few more calculations would be necessary, right? After all, an estimate of circit length x 1.5 would get you into the L.A. Coliseum. Circuit setters? Great, if you have a meter.

Jed

Mike T., Swampeast MO

Again, don't want to disagree about the flow working that way in such a loop, but I sure "see" it MUCH better with a low-loss header... If normal primary-secondary works fine with higher flow in the secondary why does Viessmann INSIST that the low-loss header MUST be used when system-side flow is not within the specs of the boiler flow?

Should be easy and not too expensive to make something similar from either copper or black iron--I'd just make the "body" pretty large--say 2½" or so???

Mike T., Swampeast MO

Just re-read. See that you're ALREADY assuming a device similar to the low-loss header.

Crazy thing is that I could "see" things better via traditional primary/secondary piping--even if I have my doubts that it could ever work that way.

S Ebels

I see that you see that I see that you see < ; )

> Just re-read. See that you're ALREADY assuming a

> device similar to the low-loss header.

>

> Crazy

> thing is that I could "see" things better via

> traditional primary/secondary piping--even if I

> have my doubts that it could ever work that way.

Floyd_7

This system runs the water backwards...

The two of these zones are heated using a 007 pump, and even if one boiler is just running they have 0011's pushing through them. If both are running there is a major backing up of water. For that reason I have taking to increasing the size of the piping just on either side of the tee's and between the tee's.
I have had no issues with water flow problems after two winters.

Floyd

Jerry Boulanger_2

500 is a constant derived

from the weight of a US gallon of water at 60 F (8.33 lbs) times 60 minutes in an hour. Hot water is less dense, so the weight per gallon and the constant for hot water are slightly lower. Why I used them both in my post I'll never know.

Mark Eatherton1

The answer is....

It DEPENDS! But you really didn't come here to hear that did you.. The return water temp to the boiler will be 123 deg F. The supply water temp to the load is dependent upon the return water temp, but in any case, if the load is a constant, it will always rise 25 degrees F with a pass through the 160 stream.

John Siegenthaler spells this out at figure 11-9 o fhis first edition of MHH.

Essentially, flow 1 times temp 1 PLUS flow 2 times temp 2, and that sum divided by the total of the two flows.

(F1)(T1)+ (F2)(T2)
/
F1+F2

Not sure how this will come out in the wash, but here goes...It didn't work out, so I used a forward slash to signify division by...

ME

S Ebels

Ohhhhhhhh

How I love the Wall. I just really get a kick out of these types of discussions.

Thanks again Dan

Kal Row

what he means is btus per hour against gpm......

it takes one British fellow as dan would say complete with, tweed suit, bowler hat, umbrella and briefcase - to heat one pound of water one degree F in one hour,

being that we use the btus on a gallons per MINUTE basis, but rate system BTU's on an hourly basis, against gallons starting from 60F, we calculate what it take to move that gallon’s temp per pound per minute for an hour to get our standard ratings

so if you impart heat to 1 gallon at the rate of 1 btu per minute per pound, or 8.33*60, after an hour you will have imparted 499.8 btu’s to it – thus we use 500 as our standard hourly btu per gallon factor – however, since 180f water is less dense than 60f water and thus the same gallon volume weighs less, so lots of folks use 495 for an average factor between 60 and 180

if that doesn’t make sense, let me know, there’s tons more bull where that came from

as for the flows through CLOSE-T’s, if t’s are real close, and the secondary is ½” larger, I wouldn’t sweat the primary flow direction, hasn’t made much difference in my experience, of course the system I accidentally reversed had tekmar controls on it so maybe it had auto-adjusted the vari-mix pump to cover my stupidity
but if you are worried, you can use a taco TWIN-TEE which is completely unaffected by direction

http://www.taco-hvac.com/uploads/FileLibrary/LM_TwinTee_Brochure.pdf

Jerry Boulanger_2

Kal

Thanks for providing more detail on the 500/495.

Re: the Taco Twin-Tee, it eliminates the differential pressure between the tees, but it will not make any difference in which way the flow goes between the primary and the secondary. The 'Law of the Tee' will always prevail - what goes into a tee must come out of a tee. If the secondary flow is greater than the primary flow, then some of the secondary flow must be recirculated within the pri-sec connection regardless of the type of connection.

Mark Eatherton1

Check my math...

it was never one of my stronger suits. I check it all the time:-)

ME

Mike T., Swampeast MO

Piping scheme [seems] better than assuming flow in different directions, but the numbers seem wrong.

In particular, don't see how it's possible to have three different temperatures at the 2nd tee when everything is flowing in the same direction.

Here's what I figured using 160° supply from the boiler. Second attachment is for 115° supply to the secondary loop.

IMPORTANT: Am still assuming 25° delta-t in the secondary loop based on 150,000 btu/hr load @ 12 gpm.

Kal Row

oh, and the different flow rates does make sense??...

but that indeed it's a fact of close T's, since the flow rate of one loop doesnt affect the flow rate of the other, you can garantee, that a percentage of water molecules equal to the difference in rate, wont come into contact with one'another and wont transfer heat

ps callife has the pinwheel flowmeter demostrator - where they turn on a pump on one loop and only it's pinwheel moves - the other doesnt budge

Mike T., Swampeast MO

You lost me Kal

Sounds like you're saying that the secondary loop in that situation would just sit and spin without picking up any heat from the primary. 20 gpm - 12 gpm = 8 gpm so those 12 gmp don't come into contact with the 8 gpm from the primary. Won't say I disagree and such would certainly explain why Viessmann insists that the primary and secondary be "decoupled" via a low-loss header....

Kal Row

it's especially true with large pipe size diffs...

and laminar flow, attached is a system diag that i worked out with floyd, i showed it to ziggi regarding his bottleneck elimination article - and he saw nothing obviously wrong with it

best viewd if printed in color - on screen you need to enlarge and pan to see the details

Mike T., Swampeast MO

Will take me a while to digest. 22" monitor at 1600 x 1200 comes in handy for viewing.

Kal Row

still need to do the ....

detail wiring and flow diagrams..
waiting for the arcitect to get his act together so that i can get to this job, i have been working on his autocad stuff cause he could not figure out, how to bear down the roof truss'es snowload over the vaulted foyer...

am thinking of doing radiant around the sklight with the pex tube going up/down the roof drains - and now thery are adding more zones and are talking about me adding hydronic fin coils to the ac system... it never ends

Weezbo

may i ask a dumb question ? What is the advantage of down

sizing the header to the boiler and the indirects...to me if your feeding 1 1/2 X2 and the boilers are tied in what is the advantage of not tying them into "T's " that are sized to two inch as well? with say 2 X1 1/2 X 1 1/2 "T's" for example..

Kal Row

1.5 is all i need for the max flow and btu rate...

i just upsized in the close-T area to make the close t-s effectivly closer - as the rule is, less than 4 pipe diameters of what you are tee-ing into - so with a 2" header i have 8 inches to work with, and the more you upsize the less turbulence induced drag you have,
- the draw back is, that that i need 6 pipe diamters on both sides, that alone is 24" - it's ok as the actual manifold is going to be 3d layers - let's get real, how many people are giving you a 16ft wall to work in