The Hydronic Formula and it's limitations

hot_rod

What the formula "predicts" and what is actually taking place inside the system, "your mileage may vary" Good food for thought in this PM Mag article.





<a href="http://www.pmmag.com/articles/95877-nature-vs-math">http://www.pmmag.com/articles/95877-nature-vs-math</a>

SWEI

A series of differential equations

that can make you cross-eyed.

CMadatMe

Interesting

But he is only comparing 1 piece of baseboard not an entire zone. The other thing I found a little odd was that he didn't go back and reference what he has in the past in his articles on fin-tube. If I remember correctly, he has stated in the past that the published ratings provided by the manufacture carry a 15% heating effect factor and that in reality should also not be used when looking at the btu/hr rating per foot.



Would have been nice if he reference the below article as a refresher before reading the current one. They go hand in hand in my opinion.



http://www.pmmag.com/articles/85001-sizing-baseboard



Lastly, there's no mention of a circulator pump and this sounds, reading in between the lines he's indirectly talking about the operation of a Delta-T pump..

Rich_49

I see

another Delta T / proper design discussion arising here . I am ready for this one . I know several others are also . This is a pivotal time for hydronics with new equipment available to us .

CMadatMe

I'm Interested

In seeing where this goes. His next article in on the number 500. I already see that, it has to do with water temp and its effect on the number 1 which effects delta-t.



Maybe someone should write a column on all the heat loss software being used on the market today and its inaccuracies.



A heat loss is nothing more then an educated guess and delta-t is an aiming device used to calculate flow rates in one part of selecting a circulator pump. For the most part a fixed speed pump is over pumping (increasing flow) 99.9% of the time.



By the way, nature means natural and there is nothing natural in a hydronic system. There is always something imposing itself on the operation and the main villain is the circulator pump!

hot_rod

let's discuss

would you agree with any of the concepts, math or findings in the article?



Regardless of what method or device you chose to change the flow rate this seems to be the main point from the article.



"The notion that a system “wants to” or even can remain at a fixed temperature drop as the flow rate changes is not supported by these results." credit PMmag 9-2013 Nature vs Math, John Siegenthaler



Some sensors and datalogging in a controlled condition could confirm these observations and findings. I suspect heat transfer manufacturers like, baseboard, fan coil, HX, or the speciality HX tube folks, have already done this



Being from the Show Me State I decided to build a visual for this. I like to "see" hydronics work



In this clear piping display with VS pumps attached I can modulate the speed and watch how the fluid reacts inside.



It is very clear at typical design flowrates of 2-4 fps the fluid twists, turns, and tumbles as it circulates. You can even see how an ell or fitting causes changes in the fluid flow.



As I crank down the flow rate you arrive at a point where the fluid seems to stop, visually, yet the flow meter still reads flow. Any air I inject to the system will just stand still. In fact any small air bubbles in the vertical piping will slowly rise and trap at a high point even with flow in the pipe. The low flow rate can no longer push the air thru the piping.



This proves to me what we accept as a known about heat transfer, that when the flow is no longer turbulent, the ability to transfer energy from that fluid stream drops considerably. The engery contained in the fluid loses it's "connection" to the wall of the tube where it has to transfer the energy to the load.



Think of Radiant Engineerings tight fit transfer plates, compared to suspended tube heat transfer, for example.



Granted I do not have lab quality instrumentation or conditions, many manufacturers do however and could test these findings. Probably for us all to agree a third party testing would be best.



Just knowing or suspecting that at some point when you reduce the flow and if you induce laminar conditions, the ACTUAL output is not as predictable as numbers in a formula.

Mark Eatherton

Bravo!!!

In the36 years I've been crawling around, through Andover mechanical rooms, I have personally witnessed a system eating at the theoretical perfect 20 degree F delta T ONCE. That was on a system that we had just finished, and the GC had yanked the space heaters in anticipation of starting the heating system the next day. The house was down to 35 degrees F, it was 10 degrees F outside. RARE conditions.



Anyone else ever seen the elusive ideal 20 degree drop in the field under normal operating conditions?



While we are on the subject, how many people have seen a properly sized bang bang boiler operating at 100% of capacity at design condition?



I think they are both directly related and are reality staring us in the face:-)



Buehler? Beulher? Buehler????



ME

Robert O'Brien

That would be never!

"

Anyone else ever seen the elusive ideal 20 degree drop in the field under normal operating conditions?



While we are on the subject, how many people have seen a properly sized bang bang boiler operating at 100% of capacity at design condition?



I think they are both directly related and are reality staring us in the face:-)"

CMadatMe

I Would

Not at all disagree with any of what is said in the article. It's common sense. Everything we use in calculation from a structures heat loss, circulator sizing, etc all are nothing more then educated guesses and effect your systems designed delta-t.



Designing around a chosen delta-t is just like using the educated guess heat loss provided to us from our favorite software. We use the heat loss as the aiming device to size our proper emitters and heating plant. We use delta-t as our educated guess as to the flow we need or want. In both cases we do our best to design our systems using the parameters we decided to design around to provide comfort to the consumer while

maintaining the best system efficiency we can.



We are not building Space Shuttle's or Rockets, they're heating systems.

hot_rod

yes and maybe

"Maybe someone should write a column on all the heat loss software being used on the market today and its inaccuracies".



The RPA actually did compare various heat loss programs back in the 90's. And yes the differences were eye opening, considering they were all based on the Manual J.



I'd bet they are more inline with one another now. Manufacturers may some times "tweak" numbers to gain and edge. I prefer to use a load program that is not spitting out brand specific data. Plenty on non radiant load programs out there.



"A heat loss is nothing more then an educated guess and delta-t is an aiming device used to calculate flow rates in one part of selecting a circulator pump. For the most part a fixed speed pump is over pumping (increasing flow) 99.9% of the time."



99.9% would be a high guestimate That would mean every pump you sized is wrong also. In a simple one zone hydronic application getting the pump sized right is not too hard. Heck some pump manufacturers still shave impellers to dial the performance in to an exact sweet spot.

I agree with todays technology we should modulate the flow based on ever changing conditions, as long as it is appropriate to the task.



I don't agree that fixed delta T is ALWAYS the best way to control, knowing what you just stated the the conditions are ever-changing on a hydronic system.



Could not extreme low flow conditions could cause a non modulating, low mass copper for example, boiler to short cycle if the heat emitters cannot transfer the energy to the load?

Isn't this why we started down the modulating burner, or buffer tank path. So adding a "smart" circ to a dumb system may not always present the results you expect. You have stated this before, and I agree.. look at the entire system not one component.



Viessmann and the uber engineered manufacturers "get" this. the pump and the boiler communicate and understand the load and everchanging needs.



The question has, and always will be Is the US market willing to pay for this much engineering. Looking at the graph at Think Hydronics makes one wonder how much time and money a manufacturer is willing to spend developing high tech for our market, for what return? Building 120V or NPT components for a very small % of the market takes some thought and commitment.



I understand your position also, you probably sell more then one brand of any hydronic component, and there are buyers for all.

hot_rod

Exactly, Chris

then would you install a component that "locks in" a certain ∆T for the heat emitters, or even the boiler? Keeping the Vitoden pump in mind

CMadatMe

Yes To The Boiler

As long as the control logic of the boiler is modulating the pump to match the boiler modulation rate, yes I would. Isn't that an option from your friends at Loch in the WH?



Viessmann, since you used them as a boiler example does on the other side of the pond. Not only is the boiler pump controlled via the control logic, there is coding for the system pump as well. Those systems as a whole over there are entirely differently designed. Constant circulation with TRV's for the most part and we really cannot use them as comparison to ours.



We would all be fooling ourselves if we don't know where all this is going. It's the

delta-t vs delta-p debate and one that rightly deserves conversation and is great debate. I think it shares knowledge that is needed in our industry but most don't absorb nor care about it.



For the system side yes I want to lock in a specific delta-t target. We cannot have any debate on this without bring the circulator pump into the conversation. In the end that is going to determine how and where we operate maybe even more so then the emitter.



The emitter is going to pull out the energy being delivered in the distribution system based on it's need. It's our job to do the best we can in delivering the btu/hr/gpm to it under changing conditions. Without using delta-t as our aiming device, how the heck else can we achieve delivery of btu/hr with any comfort of knowing we chose the right pump? Is there another method for calculating gpm requirement that I should know? Please share it or I at least hope John shows us a different way as he gets through his columns.

hot_rod

No it's not

about ∆T vs ∆P circs in my mind.



I think many know and agree what ∆P pumps can and cannot do. Plenty of articles webinars and traininig on that technology. To my knowledge every major pump manufactures has ∆T pump options on the shelves.



Nor is it about a brand, person, or color of product. It's all about heat transfer what works and what doesn't. And how an installer or designer should use this information.



The question is, or should be will locking a system to a fixed ∆T perform as needed. Granted a more efficient motor will consume less electricity. I'm asking about system performance, heat output, and efficiency gains if any.



It's about getting the known, understood, and accepted facts and information out there. I think many will take the time to understand and keep asking questions when it is not clear. We are all constantly learning, education breaks down when you don't ask questions about concepts that are not clear or understood.



Thanks for the civil discourse by the way.



You said:



We would all be fooling ourselves if we don't know where all this is going. It's the

delta-t vs delta-p debate and one that rightly deserves conversation and is great debate. I think it shares knowledge that is needed in our industry but most don't absorb nor care about it.



For the system side yes I want to lock in a specific delta-t target. We cannot have any debate on this without bring the circulator pump into the conversation. In the end that is going to determine how and where we operate maybe even more so then the emitter.



The emitter is going to pull out the energy being delivered in the distribution system based on it's need. It's our job to do the best we can in delivering the btu/hr/gpm to it under changing conditions. Without using delta-t as our aiming device, how the heck else can we achieve delivery of btu/hr with any comfort of knowing we chose the right pump? Is there another method for calculating gpm requirement that I should know? Please share it or I at least hope John shows us a different way as he gets through his columns.



I would propose the emitter is in charge. The emitter has the ability to determine the boiler operating condition also. The pump responds to the "call" for heat the boiler to emitter relationship will determine the temperature conditions. See the chapter on Thermal Equilibrium in Modern Hydronic Heating.



You say the emitter will pull what it needs in one sentence, but that you would like to lock in a fixed operating condition via ∆T. How can it be both ways? If the emitter needs or wants more flow or temperature which is really what the flow is transporting, , but a device in the system prevents this???



For this example the ∆T wants and should be able to move and adjust to what the emitter needs to cover the load. If you drop flow rate via ∆T control and the energy can not transfer due to flow type , turbulent vs laminar how is the load to be covered? The emitter wants and needs the flow but a device is saying no to the flow needed??

HDE_2

Beautiful

Next article should be how unknowing self-made hydronic experts misunderstand boiler flow rate and btu firing rates and heat transfer attempting to apply the basic hydronic formula to mod con operation.



That 20 DT bang-bang operation is not how many 5 to 10 -1 turndowns actually function.



Time to reconsider heat emitter and DOE outputs.



Perhaps resurrecting and developing a redesigned "IBR" is needed?

Rich_49

Heat emitters

are forward observers and relayers of information . They are telling the device what has been taken out .  The emitters have decided what has been transferred and relay the information to the device via an amazing thing we shall call our heat transfer medium or fluid . If it went out at 140 and comes back at 110 the device will slow down the flow if the 140 came back at 130 it would slow down the flow .  In either case I cannot understand the discord here . Heat has been lost somewhere in the system , was it through the S & R pipng which here in my projects is insulated ?  Are we disputing that the pump reacts to the information it is receiving in real time from the system .  What if a designer employed Outdoor reset and used a Delta t pump such as what I do . What then ? I'll tell you what the water temps would vary depending on conditions and the Delta T circ would make slight changes to flow , something the fixed speed pumps can't and delta P variety can but not as well . My earlier jobs do not perform as well as the ones I have installed since choosing  the better alternative . It's no wonder our entire industry and market share are where they are . Too many damned geniuses spouting too much different information to too many people installing systems poorly to begin with .  

I think it is more desirable for the piping and emitters to have a sweet spot than for the pump to have a sweet spot . In this way all the rooms (you know , the places where the recipients of the heat are) will all experience some level of comfort as opposed to the circulator being happy in it's nice little warm spot . 

STFU .  We are now designing systems that are as about as efficient as we are gonna make them , people are paying less to heat their homes than ever before , and manufacturers all have their share of the pie .

 Unless someone is gonna come up with a new Universal Hydronics Formula and change every standard we now use and explain to us why this is being done . Be quiet .

 Sorry for the rant .

One last thing .  Delta T is the by product of what has been delivered under any flow , laminar or turbulent , if no transfer took place there would be 0 Delta T . Won't Jean david Beyer be thrilled .  

Gordy

Emitters in control

To me its quite simple. Trvs are the throttle for the loop, responding to room temp feed back. We need to adapt these to radiant loops, and or zones radiantly speaking of course. Room temp is what the emitter has to satisfy. The delta T will take care of itself with the right circulator.  This is how Europe does it with rads. Why not with radiant. its really elegantly simple no?

CMadatMe

Thanks HR

Likewise, until I get to that post above Gordy then civil discourse will have ended, well at least in that conversation.



Will you always maintain a fixed delta-t. Nope. Can't. No matter the type of pump it still rides on a pump curve and at some point you will be over pumping.



A emitter/zone is designed based on a heat loss calculation and the flow you design for to overcome that heat loss under design conditions. If that emitter can overcome the heat loss at the designed flow rate and delta-t, then it should always overcome the heat loss if I maintain my delta-t while varying my flow. I will never get to the .04 gpm point because the pump won't let me. I'm limited but it's still a better way then fixed speed or "p".



I have a hell of a better chance in my opinion of making the best comfortable system sticking with varying my flow based on a delta-t then I do by just flying at the seat of my pants as a "P" type pump does and forget a fixed speed pump we both know it is not in the conversation.



I like what he wrote, its an education. Here's the question, how do you apply it to pump selection? How do you apply it at all? What is its purpose other then for knowledge if there isn't a product to put this to use? I'm left with these questions.

hot_rod

KISS

a toast to TRVs, Gordy.



I'm not sure why this super simple control is not gaining the respect it deserves in the US market? Is it the"look" of the control? Too inexpensive Un-complicated design and assembly not manly enough?



No wires required, weather and indoor temperature responsive, proportional modulation control, no complicated programing directions, or complicated I&O documents. I have seen some brands that include nothing in the box but the TRV. No motors, gears, switches to cycle out.



Yes a water seal and a wax pill that may fail. Generally harsh fluid conditions are the cause of seal failure and off season sticking. Both can be designed out of them.



Match them with a pump that adapts and learns their good habits and let them be.

SWEI

Proportional zone valves

are the answer IMO -- at least until we get variable speed micro-circulators.  I consider them essential for use with ODR -- a bang/bang zone valve has a hard time effecting control when the fluid temp is only a few degrees from the space temp.



TRVs are one form of proportional zone valve.



Belimo CCVs work marvelously for this, but you need a zone control with 2-10V outputs (which for some reason is not an off-the-shelf item here in the US.)  We use DDC for this on commercial jobs.  I'm working on something for smaller systems.

CMadatMe

I Guess HDE

All the boiler mfgs should start rewriting their manuals because they all do it in those nice pump selection charts. If it's good for high fire, it's good for the rest of the modulation rate.



Your a Navien guy. How do you propose your getting the published 130,000 btu/hr in their lit for that manifold kit? Please educate us. That is a fixed speed pump on that unit correct?



This was a nice conversation of debate and sharing until you showed up with throwing stones. Why do you hide behind initials anyway?

Gordy

Low tech

Seems to get a bad rap. If its not complicated it can't be superior is the North American mentality.



Someone could make them pleasing to the eye. I'm all for KISS. Europe has a habit of making kiss elegant. So why can't we?

CMadatMe

Nice Gordy

I'm surprised we don't hear more from Danfoss and other mfgs of them like we do from Caleffi, Taco and others who promote the heck out of educational webinars. The Danfoss' and companies that offer them need to get the word out on how effective they can be. I don't think most contractors actually know how to deploy them in residential applications.



We see them more in apartment complexes and other commercial type constant circulation systems.

hot_rod

you see my take would be

You said:

"One last thing . Delta T is the by product of what has been delivered under any flow , laminar or turbulent , if no transfer took place there would be 0 Delta T . Won't Jean david Beyer be thrilled "



I see it as:



The delta T is the driver of the heat transfer either out of the emitter or into the system at the boiler or heat source.



The larger the delta the more energy transfer, and vice versa.

If there is no delta T then there is no energy transfer. And if this happens when the load calls or needs heat, you might have a flow condition present that is preventing it.



"The pump won't work because the vandals stole the handles" Bob Dylan's take



Or you have reduced the flow to a point where the energy is no longer able to be transfered from the fluid flow.



In the GEO industry this flow rate and type of flow is critical. You need to move the energy transfer from the fluid to the earth or water. Lots of attention is given to that Reynolds number in loop designs.

Rich_49

Hot Rod wrote

The delta T is the driver of the heat transfer either out of the emitter or into the system at the boiler or heat source.



The larger the delta the more energy transfer, and vice versa.

If there is no delta T then there is no energy transfer. And if this happens when the load calls or needs heat, you might have a flow condition present that is preventing it.

  No need for drunk drivers though Bob , all over the road , smashing stuff up and killing system efficiency .  Think of these pumps that CAN keep your DESIGNED FOR diff as a drive by wire mechanism , if the drunk driver veers left or right it is corrected by others before disaster can happen .

     "The larger the delta the more energy transfer, and vice versa.

If there is no delta T then there is no energy transfer. And if this happens when the load calls or needs heat, you might have a flow condition present that is preventing it. "

  I agree , I also don't concern myself with this nonsense . If there is no Delta T in a system the zone better be satisfied and if there is no heat transfer when there is a call I did something horribly wrong . This condition should not happen . And if it does I will find another profession .

   About Geo Thermal equipment . I would love to discuss my thoughts on that in another discussion but not here , the reason for this is that you only require a pump and a source of heated fluid or earth for Geo Thermal , a ground source heat pump is another animal completely .  I repeat , Geo Thermal requires no mechanical assistance other than to move naturally heated fluid to your system .

bob_46

Siggy's article

My take on Siggy's article is that he is trying to illustrate that trying to control emitter output

by controlling flow rate is a fools game whether you do it with pumps or valves.

Jean-David Beyer

differential equations

I did not see even one differential equation in that article. I know my eyes are weaker than when I got a degree in math over 50 years ago, but I still think I would recognize one when I saw one.



And my vision problems are myopia and presbyopia. A pain to have both. But I am not cross eyed.

Zman

Great post

First off, thank you all for what is hands down the best post of the year.(I really mean it)

Thank you Jean David for for your expert mathematical opinion on differential equations. It is just so helpful and relevant to the subject. (I really don't)

Gordy

Not a fools game its finding the sweet spot.

Look at the charts in the article:



Going from .4 to 1 gpm delivers an additional 500 btus output.  But to gain another additional 400 btus output the flow rate must be increased 4 fold going from 1 to 4 gpm.

In the end my interpretation of all three charts is that flow rate of .4 to a little over 1 gpm reeps the highest return in energy used.



I think it enlightens the fact of good design. Tube dia. , loop lengths, keeping circulator as small as possible to achieve the most output with the lowest energy used to do it.



The math works but there becomes a point of diminishing returns so one needs to apply it with thought.



Like tube spacing as you tighten tube spacing water temps decrease, and output increases to a point where the additional material, and labor costs are prohibitive. To the net gain in the decreased temps, and increase in output.

Mark Eatherton

Interesting observations HR...

And I completely understand what it is that you are saying and conveying as it pertains to the Reynolds numbers. In applying this to DHW CR systems, we SHOULD induce laminar flow, thereby reducing exterior pipe loss, AND reducing the potential of hydraulic erosion corrosion.



Variable speed pumping on DHW CR makes as much if not more sense than space heating applications. The "load" is a constant (fairly much so) hence the need for speed is not there. Simple timer and aquastat on the return with a variable speed circulator, and viola, boo coo energy savings AND minimal wear and tear to your piping systems. Maybe THIS is why all these pump makers are going to non ferrous pump bodies...



Great idear there Mr Show Me :-)



ME

JStar

Flow

These are the things that keep Quantum Physicicts awake at night.



I think the point of the article was to shine light on the fact that not all paper-theories translate into real world results. No big surprise there. We don't have enough test equipment, and not nearly enough time to really research this stuff. But, we do our best to get close to perfect.



I can't help but think that Delta-T and Delta-P are two sides of the same coin. In the end, you are still controlling flow. Think about a circulator not running. It has a Delta-P of zero. No pressure difference = no flow. Increase pressure difference = increase flow. With Delta-T...too high = low flow. Too low = high flow. Still controlling flow.



Maybe what we really need is a circulator that can calculate BTUH output on its own. Isn't that the calculation we use to determine outdoor-reset curves? But what defines BTUH? Average temperature. Not just temperature difference. Example...



10,000 = 1 GPM @ 20F Delta-T



However, 20F Delta-T could be anything from (180-160) to (800-780F). So, how can output be defined only by Delta-T? There is the "flaw" in the hydronic formula.



And what about the room temperature's affect? A 50F home will produce a larger Delta-T than an 80F room, while delivering the same average supply temperature. So what does that 50F room need? More or less flow? More, right? More flow = more BTUH's delivered. How do we increase flow? Increase pressure or lower resistance. A lot easier to increase pressure, which in turn increases flow, and decreases Delta-T. How are they not the same?



Maybe the answer is that we need a Delta-EVERYTHING circulator. Pressure, temperature, flow, BTUH, etc. Or, maybe all modulating circulators already are Delta-E. Maybe we need modulating emitters.



Tangent: end.

Rich_49

The Hydronic Formula

has no limitations except people who doubt it's relevance , do not use it , believe it has a cut off point .

knotgrumpy

TRV's are wonderful

Gordy said: "To me its quite simple. Trvs are the throttle for the loop, responding to room temp feed back. "

What a wonderful thread this is!  Good stuff.

I want to thank Gordy for leading me towards TRVs last winter.  I've spent the best part of the summer ripping out my iron monoflow piping, replacing it with a reverse return 2 pipe system with TRVs.

Awesomeness.  The TRVs do exactly what Gordy is saying and are almost spooky in keeping the room comfortable and at a set temp.  I take that back - they are spooky.  Steady heat and not the cool down, heat up, cool down, heat up.

I'm using a B&G Ecocirc Auto circulator on the secondary side of the system (thanks SWEI) and it seems pretty happy.

Put the Vario model of that pump on primary loop.  Both circs drawing less than 50W.  Both circs run continuously.

I'm close to happy with my primary loop, but it will need a few mods.

I'll post some pictures later if you'd like. 

Gordy

Glad to hear it!

While sometimes painful to incorporate into an existing system it's out come is well worth it. I'm sure the Eco circs helped bunches also. Along with reverse return piping.