∆ T, gpm, output relationship

chakil

Hi,
When Desired Delta narrows we would like to slow a circ , when Delta widens past desired point we would like a circ to speed up .
output= 4gpm*500*20 = 40000 btuh supply t= 180f return= 160
if we slow the circ we would have wider delta t? and less heat output for the emitters?

Zman

You will have less output for the emitters as you go downstream. The first one won't see much effect because it will still be getting the hottest water. The last in line will have a more significant performance drop. A good designer will upsize the downstream emitters to compensate for the designed temp drop.

hot_rod

chakil said:

Hi,
When Desired Delta narrows we would like to slow a circ , when Delta widens past desired point we would like a circ to speed up .
output= 4gpm*500*20 = 40000 btuh supply t= 180f return= 160
if we slow the circ we would have wider delta t? and less heat output for the emitters?

Not really in all cases would you want to slow down flow as delta tightens. Radiant floors for example like a tighter delta compared to panel rads. A tighter delta across a heat emitter, all things being equal (flow rate), would produce a higher average temperature and higher output.

We chose this graph for the cover as it shows how heat output by changing flow rate is very non linear, notice the hockey stick shape to the output as flow increase. At some point the additional pump required to get a small % just isn't worth the squeeze.
A good read on how heat transfers inn hydronics.

https://www.caleffi.com/sites/default/files/coll_attach_file/idronics_23.pdf

chakil

hot_rod said:

chakil said:

Hi,
When Desired Delta narrows we would like to slow a circ , when Delta widens past desired point we would like a circ to speed up .
output= 4gpm*500*20 = 40000 btuh supply t= 180f return= 160
if we slow the circ we would have wider delta t? and less heat output for the emitters?

Not really in all cases would you want to slow down flow as delta tightens. Radiant floors for example like a tighter delta compared to panel rads. A tighter delta across a heat emitter, all things being equal (flow rate), would produce a higher average temperature and higher output.

We chose this graph for the cover as it shows how heat output by changing flow rate is very non linear, notice the hockey stick shape to the output as flow increase. At some point the additional pump required to get a small % just isn't worth the squeeze.
A good read on how heat transfers inn hydronics.

https://www.caleffi.com/sites/default/files/coll_attach_file/idronics_23.pdf

Hi,
From the pdf :

CONSTRAINED ∆T :
When design load is no longer present in one zone, and it turns off, less heat would be removed from the distribution system. This change would reveal itself as anincrease in return water temperature (assuming that the supply water temperature is fixed). The ∆T of the circuit would decrease. A controller that senses this decrease could respond by reducing the speed of a circulator so that the design load ∆T was reestablished for the zones that remain active. This process would repeat when another zone turned off. This method of control reduces circulator energy use during partial load conditions.

Question:
when water flow slows heat emitter releases less heat so return temperature should be higher then ∆T would decrease
so how can we constrain ∆T by reducing the speed of the circulator?

chakil

Zman said:

You will have less output for the emitters as you go downstream. The first one won't see much effect because it will still be getting the hottest water. The last in line will have a more significant performance drop. A good designer will upsize the downstream emitters to compensate for the designed temp drop.

you've assumed a circuit in series

Zman

Is this a hypothetical conversation or do you have a heating system you are trying to dial in?
I did assume baseboard heat in series, perhaps my mistake.
If the question was hypothetical, Idronics 23 should give you plenty to chew on...

chakil

it is hypothetical question

chakil

Youngplumber said:

@chakil going to try and answer your last question.

As you shed load by zone valve shutting, you need less total Btus.

So where you once needed 15k BTUs from three different zones. Two zone valves shut, now you only need 5k BTUs for one emitter. Do the hydronic formula and you come up with less GPM needed. That's assuming the three zones have the same Delta T. 

Change the delta T in any of the zones and the gpm needed for that zone will fluctuate.

I think hot rod was saying that delta T is not linear with gpm not that they aren't related.

Does any of that help?

conventional Boiler doesnt modulate
at first it produces 15k for the design load and ∆T 20 then load become only 5k
the boiler will still produce 15k the supply temperature would increase and ∆T decrease and shortcycling with same gpm?

if we lower gpm, ∆T would increase ? but
if we lower gpm less heat would be relased by the emitter so return water temperature would be higher so ∆T decrease ?

Zman

The answer to your question about the relationship between GPM and delta T is pretty simple. If the supply temp stays the same and you reduce your gpm, your delta T will widen and the emitters will put out less heat. I can't think of a system where this is not the case.

Whether it is a good idea to do so depends on a whole lot of factors: mass of system, output of boiler (modulating or not),laminar vs turbulent, series or parallel to name a few.

It sounds like you are are asking for opinions about a particular system (because you correct people who make assumptions) but don't want to give specifics about the system.

As with anything heating, the answer is "it depends".....

hot_rod

Don't forget the load. The delta on a fixed speed fixed output boiler will drop as the heat load drops. The heat emitters are in charge of the boiler operating condition, always. We put limits and temperature controls on boiler to prevent them from working at unsafe or undesireable conditions.

With a fixed speed circulator and on off burner control, as the load lessens the boiler will short cycle. All a Delta T circulator does is "pulse" the heat into the loop to lessen the on/ off cycling when you have a fixed output boiler, for example. In some cases the gpm drops to a point where the energy being pulsed will not cover the load as the graph shows. So you would need to put a low speed limit as most delta T circs show where the bottom of the curve is truncated.

If the boiler allows, it is much better to lower the SWT, modulate the burner to match the ever changing loads. The loads change as zones kick off, or you are at the point where the thermostat shuts off, load is covered.

As the hydronic formula shows wider delta operation can move or carry more BTUs. The math and basic physics also show that higher temperature emitters, of any type transfer more heat energy with higher AWT. This attached examples shows a constant SWT and 3 different flow rates.

So in my simple way of thinking, design around a delta T, pick one out of thin air, 23.7 for example. You may at some point actually see the system operating at that exact delta. It may be a 30∆ on a cold start, it may drop to a 5∆ just before it satisifies, really no harm in that movement.

If a fixed output boiler short cycles with a low load condition, it's not the boilers fault, the load is the issue. So lower the SWT with ODR if possible. That will help but not completely eliminate short cycles. A buffer tank is another option. The older large mass cast boilers and large volume steel pipe systems were "buffered" by default. Todays cast boilers hold a few gallons and weight 1/2 of the older iron, so that cycling can be more of an issue.

IF we could exactly match the SWT to that ever changing load, it is possible the boiler fires and never shuts down, the output is always exactly matched to the moving loads. With todays deep modulation and ODR, ramp delay and other functions you can get real close to that condition. If the weather cooperates and a load within that modulating windows is present.

A fixed speed, fixed output boiler and fixed speed circulator it will not be easy to get to and stay at a steady state condition, zoning, especially micro zoning makes it worse.

chakil

Youngplumber said:

@chakil going to try and answer your last question.

As you shed load by zone valve shutting, you need less total Btus.

So where you once needed 15k BTUs from three different zones. Two zone valves shut, now you only need 5k BTUs for one emitter. Do the hydronic formula and you come up with less GPM needed. That's assuming the three zones have the same Delta T. 

Change the delta T in any of the zones and the gpm needed for that zone will fluctuate.

I think hot rod was saying that delta T is not linear with gpm not that they aren't related.

Does any of that help?

Hi thanks for the reply
in your example, by reducing GPM we retablished the delta t for the 5k zone
a conventional boiler would still short cycle isnt it? what would be the impact of reducing GPM on the boiler, in your example?

chakil

Hypotheticaly
considering simple system with conventional boiler pump and emitter
boiler set 180 f with 20k btu h 4gpm in design load delta t 20
and two zones each 10k
lets say only 1 zone on certainly the boiler would short cycle
by reducing gpm we would keep delta tee for the zone
what would be the impact of reducing gpm on the boiler, would it keep short cycling ? more short cycling or less?

chakil

i dont understand how can we constrain delta t
let say system with conventional boiler with 20k, 2gpm, delta 10f, load is 10k, certainly the boiler would short cycle
so can we reduce gpm so we increase delta t and then no more short cycling? is that possible?

chakil

how could it be possible, whatever you do with gpm the boiler still give 20k and load take only 10k
where would go the other 10k of the boiler?

EdTheHeaterMan

The other 10K causes the high limit to stop the burner. Oops if the burner stops then the other 10K isn't there.

You can get 40,000 BTU per hour from an 80,000 rated boiler by shutting it off for 30 minutes every hour. or you could use a 40,000 BTU burner for 60 minutes and get the same amount of heat... In a perfect world.

We don't like to do that because there is some added energy lost on start-up and there is also more energy loss on shut down and then there is stand-by loss too. To pump or not to pump... That is the Question!

Jamie Hall

chakil said:

how could it be possible, whatever you do with gpm the boiler still give 20k and load take only 10k
where would go the other 10k of the boiler?

Another way of looking at it besides @EdTheHeaterMan 's comment is that the extra 10K of heat goes into raising the overall temperature of the water leaving the boiler. Now at some point the boiler is going to hit its high limit and shut down. Then the circulating water will gradually cool down. At some point later, the boiler hits the low limit and fires up again.

EdTheHeaterMan

That is a nicer way of sayin' that @Jamie Hall But you always speak better Shakespeare than @Youngplumber and @EdTheHeaterMan

chakil

EdTheHeaterMan said:

The other 10K causes the high limit to stop the burner. Oops if the burner stops then the other 10K isn't there.

You can get 40,000 BTU per hour from an 80,000 rated boiler by shutting it off for 30 minutes every hour. or you could use a 40,000 BTU burner for 60 minutes and get the same amount of heat... In a perfect world.

We don't like to do that because there is some added energy lost on start-up and there is also more energy loss on shut down and then there is stand-by loss too. To pump or not to pump... That is the Question!

so the boiler would run 30min/h straight or it gonna short cycle ?

Canucker

@chakil Like @EdTheHeaterMan said, if the system isn't shedding the load, aka heating a room, it sends that heat back to the place with the fire, aka the boiler, making the return temp climb. If the temp climbs too high the internal safety for the boiler shuts it down until it cools off. Slowing the flow down on a fixed output boiler will speed up the time it takes to thermally shut it off because its flame puts the same amount of BTU into the chamber all the time. The fluid flow is what takes it away and keeps the chamber below its high limit shutdown. Slow down the flow and...

Zman

You are tweaking the wrong variables.
From the boiler side of things there are only 3 things that effect short cycling.

1. Lowest turndown capability.
   
2. On/Off differential.
   
3. System Mass.
   

That's all folks!

chakil

Canucker said:

@chakil Like @EdTheHeaterMan said, if the system isn't shedding the load, aka heating a room, it sends that heat back to the place with the fire, aka the boiler, making the return temp climb. If the temp climbs too high the internal safety for the boiler shuts it down until it cools off. Slowing the flow down on a fixed output boiler will speed up the time it takes to thermally shut it off because its flame puts the same amount of BTU into the chamber all the time. The fluid flow is what takes it away and keeps the chamber below its high limit shutdown. Slow down the flow and...

but from the formula q=500.gpm.delta t
by slowing gpm we would have a bigger delta t which should prevent short cycling?

Jamie Hall

No, it won't. The short cycling comes from the boiler putting out more heat than the emitters can get rid of. Changing the flow rate won't change that.

HomerJSmith

Yes, yes, yes! hot_rod, "The heat emitters are in charge of the boiler operating condition, always." In fact, heat emitters is what it is all about as far as a comfortable environment is concerned. The boiler is there just to satisfy the heat emitters.

One does a whole house heat loss calculation and determines what size boiler to install without taking the heat emitters into consideration may end up as someone stated, "You have a Ferrari up against a tree with your foot on the gas, the wheels are spinning, but you're not going anywhere." The boiler is putting out the heat energy, but it hasn't anywhere to go. Sizing boilers is a juggling act. There are so many variables that one has to consider and heat emitters are just one.

I posted earlier that the BTU transfer from a heat emitter to a space was dependent on flow and SWT. That was simplistic although true. The forgotten element was physics that is the space environment. Heat energy moves to cold. What if the SWT to the heat emitter is 100 deg and the room temp is 100 deg, what would you say the heat energy transfer would be?
Maybe a big goose egg? Maybe a ∆T of 1deg? Environmental factors play a larger part in your planning.

What's so sacred about a ∆T of 20 degs? Wear & tear on the boiler electronics with a tighter ∆T? Boilers have circuit boards that have relays that have points that arc when they close and pit. I can see that. But, HXs that heat up from 70 deg to 180 deg is more stressful, I would think. Better to run the boiler with a tighter ∆T, less stressful. You would still need to put out the goods to the house environment.

Years ago, I was talking to a Glowcore engineer and he said the best operation was for the boiler to run 24/7 and never shutting down. Long, long, long run times while delivering the goods.

It's so complicated, which is why hydronics is so exciting. There are all these puzzles to be worked out and there is a pleasurable satisfaction when everything works out the way you planned. So, what's so sacred about a 20 deg ∆T?

EdTheHeaterMan

Here is some perspective on the points that Homer and Younge and Jamie are trying to make


Now the piping to and from the sections of this much radiation will be different in order to maintain the same ∆ T on the 4GPM loop and the 1GPM loop(s) But it is possible with creative piping design.

This discussion reminds me of a story I heard when I was 14 years old.

There were 3 men who wanted to stay at a hotel but they were on a budget and decided to share the room. The Assistant was on duty and was not very goods with fractions and could not figure out how to divide the $25.00 cost of the room equally among the 3 men. He decided to charge them $30.00... $10.00 each. and all was good in the world.

When the manager returned, he was unhappy with these customers being overcharged and told the assistant to return the $5.00 that he overcharged them. Still uncomfortable with the mathematics the assistant decided to return only $3.00 to the 3 men, $1.00 each, and keep the other $2.00 as a gratuity. The men were happy for the corrected overcharge (the refund) and all was good with the world.

So, as I see it each man paid $10.00 and was refunded $1.00 so they actually paid $9.00 each for the room... Right?
3 men times $9.00 is $27.00... Right?
The assistant has a $2.00 tip... Right?
$27.00 plus $2.00 is $29.00... Right?

So where is the other dollar?

This is an example of using incorrect information that sounds perfectly logical to get an impossible result. Think about it. Where is the extra dollar? is the wrong question based on the information provided. This info improperly presented will never get you a correct answer. Can you see the flaw in the logic of the presentation?

GPM will not get you the answer you are looking for because the end result is not a primary function of GPM but a function of the BTU output of the boiler and the capacity of the emitters to release the heat available. GPM is only a small part of the equation. Like that extra dollar that isn't really gone!

Respectfully submitted.
Mr.Ed

HomerJSmith

I remember that as a kid, it brings back memories. Great!

There weren't no extra dollar, the govament got it! Sneaky!

EdTheHeaterMan

Adding $2.00 is what is wrong. it sounds right but it is not the correct parameter. Each man actually paid $9.00... You either Add $3 t get $30 or subtract $2 to get $25. But you don't add $2.00 to the $27.00.

That is how Uncle Sam gets you to pay for entitlements. LOL tell you you only are paying $3. but actually charging $5.00.

How does that relate to Gallon per Hour and BTU per minute... Did I say that right?


This sounds like the perfect time to bring up Tigerloops and Circulators are not Pumps.

HomerJSmith

We all know that a circulator is not a pump, but pump is a lot easier to type than circulator, besides pump is a common idiom (a language, dialect, or style of speaking peculiar to a people) or is that to peculiar people? hmmm I always thought you guys were peculiar.

EdTheHeaterMan

@HomerJSmith, to quote our illustrious president... "Come-on Man"
if the company that makes it calls it a pump... Then it IS a pump

I guess "That depends on what the definition of IS is!" To quote a former illustrious president.

And where is @STEVEusaPA when you need to talk about Tigerloops anyway?

WOW this has gone way off topic... That is what you get when you have a hypothetical quwey. If the OP just wanted a real problem solved, I guess we could hypothetically help him.

hot_rod

The problem is that the heating load is ever changing, so the boiler would like to modulate along with the changing load. To a point a modulating boiler can follow the load. Here is an example for upstate NY, notice how small of a % you are at design condition. So with multiple zones, especially micro zones even a modulating boiler will cycle. Ideally lowering the SWT will help eliminate short, less than 10 minute on cycle. If the cycling drives you, and your boiler, crazy a buffer tank will extend on/ off cycling.

chakil

hot_rod said:

The problem is that the heating load is ever changing, so the boiler would like to modulate along with the changing load. To a point a modulating boiler can follow the load. Here is an example for upstate NY, notice how small of a % you are at design condition. So with multiple zones, especially micro zones even a modulating boiler will cycle. Ideally lowering the SWT will help eliminate short, less than 10 minute on cycle. If the cycling drives you, and your boiler, crazy a buffer tank will extend on/ off cycling.

Hi,
How could lowering SWT help eliminate short cycling?
my understanding is by lowering swt, emitter would release less heat and that would tighten delta t

Jamie Hall

I haven't really been following this one, so if some of what I say is duplication, forgive me.

You have to start with the emitter -- baseboard or whatever. That emitter will release, to the space, a certain amount of power -- BTUh, which is dependent, given the design and installation of the emitter, on the average water temperature in the emitter. Period. That amount of power will come from cooling the water going through it -- or in it. The power is given by the classic hydronics equation -- BTUh equals 500 times gpm times delta T. You can't change that one. For a given average temperature of the water, you can increase the delta T by decreasing the flow (you'll have to raise the input temperature by half the increase in delta T) or you can decrease the delta T by increasing the flow.

So -- the only way you can increase the heat output of the system as a whole is to increase the input temperature to the system at a constant delta T, or increase the flow through the system at a constant input temperature.

Further, for a given input temperature, there is a very definite limit to the heat output of the system -- delta T can approach zero, but never reach it, no matter how much flow you can cram through the pipe.

Now the average output of the boiler has to equal the average output of your emitters. You can't fight with that one. If the running output of the boiler is greater than the heat output of the emitters, the boier is going to have to cycle on and off. You can change the cycle frequency -- but not the ratio of on to off -- by increasing the stored mass of water, such as with a buffer tank. This will give you longer cycles.

Lowering the source water temperature obviously will result in different cycling, as the boiler output stays the same (unless you have a modulating boiler) but with the lower emitter temperature the mismatch between boiler running output and emitter output is greater. For a fixed mass of water in the system, the on periods will become shorter and the off periods will be longer.

chakil

Thank you for the reply
I have a question:
by adding buffer tank we would have longer cycles
but what would happen when design conditions occurs ? would we have a delta t much wider than designed one?

Zman

Delta T and boiler cycles are not really related.

Boiler cycles are determined by boiler output vs system load. When the output exceeds the load, the boiler will cycle. Whether or not it short cycles depends on the boiler on/off temp differential and the mass of system.

It all comes down to, 1 BTU is the amount of energy required to raise 1 pound of water 1 degree Fahrenheit. I think this is the part you are missing. All of this math, including the universal hydronic formula are derived from this. BTU/hrs = GPM x Delta T x 500. The 500 comes from 8.34 (lbs per gallon of water) x 60 (minutes per hour) x 1 (specific gravity of water).

When you add system mass (buffer tank), you give the system more capacity to store energy and to run longer on and off cycles. If you want to figure out how much difference, figure the weight of that water and how long it will take the boiler to heat it up to the boiler off setpoint and the system to cool it off to the boiler on setpoint.

When you increase the on/off differential you allow the boiler to run longer (store more energy) and take longer breaks (release the stored energy) in between cycles.

When you change the GPM without changing the boiler output or load, it just changes the delta t. The formula is finite. There is no free lunch, the energy cannot be created or destroyed....

ryanwc

To use Ed's analogy, here's where people are putting the $2 in the wrong column. Delta increases, but supply temps increase faster.

If you lower flow from 4 gpm to 2 gpm, the emitters still put out nearly the same amount of energy, but they do it by grabbing more btu's from each gallon that passes through. So in a sense, the delta of return does increase - return water that was coming back 10 degrees cooler will come back 20 degrees cooler.

But meanwhile, you've got a boiler putting out btu's at the same rate. Before, it was adding those btus to 4 gpm, and raising the temp by 10 degrees. Now, it's only seeing 2 gpm, and raising the temp on those by 20 degrees. So it's almost a wash. Except that the emitters ARE slightly less efficient at slower flows, which means the short cycle happens just a little quicker than it would have.

ryanwc

>Delta T and boiler cycles are not really related.

If you treat the emitters as a fixed variable, that's true. So for Chakil, that's true. But the point of adding additional emitter capacity is to give a delta T that is in balance with the lowest capacity of the boiler.

HomerJSmith

Is "fixed variable" a oxymoron? It seems to me that the delta T is a variable. As the flow is fixed, the heat energy emitted by the heat emitters varies depending on the transfer from the emitters to the environment. As the environment temp increases, the heat energy transfer from the heat emitters changes, lowering the delta T. hmmm

hot_rod

This is a pretty good read on how heat transfers.
https://www.caleffi.com/sites/default/files/coll_attach_file/idronics_23.pdf

HomerJSmith

What I'm trying to say is: "Don't forget the load. The delta on a fixed speed fixed output boiler will drop as the heat load drops."--hot_rod. The load is not the heat emitters, but the environment that one wants to put heat energy into.

ryanwc: "Delta T and boiler cycles are not really related." I'm trying to grasp the concept here. It seems to me that they are most definitely related because of the differential integrated into the control mechanism of the boiler. I never thought of it before, but what causes short cycling, a narrow delta T or a wide delta T? What is the mechanism of short cycling? I think I would opt for a wide delta T, but if it is really wide the boiler would never reach shut down depending on the flow.

hot_rod

Maybe another way to explain it is the ∆t is an indication of the heat energy that was transferred through the heat emitters.
The heat emitters are in charge of the boilers operating condition. The heat emitters are responding to the temperature condition in the space they are located.
The thermostat is basically a high limit switch.

SuperJ

chakil said:

Hi,
How could lowering SWT help eliminate short cycling?
my understanding is by lowering swt, emitter would release less heat and that would tighten delta t

The call for heat cycle will be longer (thermostat won't short cycle), but the btu/hr will drop (Causing the boiler to cycle more), this is good for comfort until you start to short cycle the equipment.
That's where ramps, deadbands, buffertanks, combining zones and anti-cycle delays start to be relevant.

It's a common pitfall where people seem to try and impose a delta T on system, you can end up over constraining things, somethings got to give. The delta will often be reflective of the conditions, and shouldn't necessarily be constrained.

Jamie Hall

Perhaps if we sort of start at the end, as it were? The emitters -- the radiation -- can put out a certain amount of power. That depends on the design and size of the emitter, which is more or less fixed, and the difference between the average temperature of the water in the emitter and the environment (which is to @HomerJSmith 's comment that the environment maters). The average temperature of the water, in turn, depends on the input water temperature to the emitters and the flow rate, as the latter times the resulting delta T is the heat output. So one can see that there is a relationship between the power output of the emitters -- which for the rest o the discussion we can assume is fixed -- the input temperature, the flow rate, and the delta T. If the flow rate goes up, the delta T will drop for a fixed input temperature. If the input temperature goes up, the delta T will, paradoxically, drop or a fixed flow rate.

Now what do flow rate and delta T have to do with boiler cycling? Almost nothing at all. If the boiler is fired at a rate which is greater than the power demand by the emitters, it will have to cycle on and off. The average output of the boiler, over the total cycle of the thermostat, has to equal (or actually slightly exceed -- but let's leave losses on startup and shutdown out of this) the power demand of the emitters. If the differential between cutin and cutout temperatures for the boiler is small, the cycle rate will be higher. If the differential is large, the cycling will be slower, with both longer on and off cycles. The frequency of the cycling -- the number of cycles per hour -- will be controlled by the total mass of water in the system (this is why buffer tanks work), but not the relative length of the on and off cycles

In practice, of course, things are somewhat messier, as if we feed the emitters with a lower source temperature, they don't demand as much power, and all else equal the boiler on times will be shorter. If we increase the flow and decrease the delta T for a given constant input temperature, the power demanded by the emitters is greater and the relative boiler on time will be greater.

And one can write some exceedingly messy equations to relate all of this stuff together...