Delta t ...... The obsession

Gordy

Can everyone at least agree that delta t is a design constant? 10, 20, 30, any number in between .



Can everyone also agree that the very same constant used to design a heating system, or emitter becomes variable upon commissioning of said product depending on operating parameters? Say cold start, no setback , setback etc. to say a few.



So,what's all the new obsession of trying to control it, and keep it a constant no matter what?



People are acting as though this is a huge mile stone, and seemingly hydronics has been done wrong for 100 years.

SWEI

Controlling for ∆T

Is a reasonable strategy since it maximizes pumping efficiency.



The ∆T between the emitter and the affected space comes first.  As that drops, the rate of heat transfer drops, which in turn lowers the emitter flow requirement to deliver those BTUs.



It seems to me that Jean-David's system might be well-served by ∆T pump control.

Jean-David Beyer

It seems to me that Jean-David's system might be well-served by ∆T pump control.

Let us say I ignore my convictions and agree (for the sake of argument) to do this.



What delta-T should I set my radiant slab circulator to?

What delta-T should I set my baseboards to?



Should these have the same delta T? It is pretty clear they cannot have the same supply temperatures because the nature of the emitters and the loads are so different.



Since the boiler will change its firing rate to track the outdoor temperature, and since on warm days I need very little heat, what happens when the boiler is putting out  76F water and the thermostat is at 69F if your recommendation above is more than 7F?  It seems to me the circulator would have two choices:



1.) Stop. If it stops, it will never get a higher delta T; in fact the delta T will drop to zero after the pipes all cool down. This will not deliver comfort to that zone.



2.) Slow down a lot, but fail to maintain the specified delta T. In this case, what is the benefit to a delta-T circulator?



If I run two of my 007 circulators all the time (and it is somewhat less than that) and pay $0.20 per KWH (my total cost is almost that high, but the incremental cost is quite a bit less), that figures to cost me $0.036 per hour.. If a constant speed ECM circulator costs 1/10 of that to run, I will save $0.0324 per hour.. If the heating season is 2000 hours, that would save me about $64 per year. It would take probably 10 years to pay off the switch to those circulators unless it heated the house much more efficiently than the present setup does. Now a constant speed circulator would save me only the cost of running the motor. If it is wired up delta-T, we are back to the question of what delta T should it be set for, and what will the pump do in the face of constantly changing supply temperature water, especially at very low supply temperatures and little heat demand.

Eastman

∆∆∆∆∆∆∆∆∆∆∆∆

best thread title Gordy



How is your boiler interconnected to the rest of the system Jean?

Eastman

Detal T control

Hypothetical,



Let's say you have two or three emitters in series in the same zone.  They're sized for design day temperature and design day delta.  How does ODR impact this balance?  Should the delta remain fixed?  Should it adjust on it's own accord?  Is there an optimal delta with respect to supply temperature for such conditions?

Gordy

Jean

I think you have to make the call.



If your willing to eat the cost of two 007 in working order resell on ebay and the cost of the new ECM circs. Then by all means give it a go this really seems to bother you.





If not let it be, the system works does it not? What you gain in efficiency will be very little in my opinion. Will it be technically more efficient yes. Will it be huge no.



You all ready have an oversized boiler so that's a strike which I only know from previous posts you have made.

Gordy

Optimal delta

Is the one that gives the most output with the least flow, and lowest water temp providing the highest comfort. Lots of combinations. This will all depend on emitter type, and how it's used. Meaning say a radiant slab in a home verses a warehouse verses a snowmelt system.



In the end its about pump energy saved as SWEI points out.



What I fear, and see in the future is a delta t or p pump being used to correct a design deficiency verses being designed into a new system properly.



A first aid kit for the tool box. For people who think they know what they are doing, but really only know enough to get them in trouble. Only to find out when it's to late.



You can control your deltas, and design them to be efficient in a system through proper pipe sizing which lends itself to smaller circs. The ECM circ then becomes a bonus in efficiency of the system, and not a tool to correct a design deficiency.

Zman

That's more like it...

Gordy,

Thanks for starting this one. I agree with you that it is a design constant. In the real world it will vary all over. A startup on a cold slab will be a higher delta t.  Out door reset will lower the delta t. Some of this can be corrected using ecm or other controls. I also think that to some extent, who cares? If a system has a delta t of 15 at cold start up on cold design day, then stabilizes to 10 is it a problem?  If the same system drops to 4 when the boiler temp is lowered on a warm day is that bad? I think it is completely normal.



Jean,

A system with a delta t at zero or one is not giving off any heat unless you are pumping the crap out of it. If you loops are circulating 3 gpm at 1 degree you are giving 1,500 btu/hour to the space, not including the heat lost in the piping and circulator. You could turn on 8-60 watt light bulbs or invite 3 people over and generate the same amount of energy. I think you are wasting energy pushing cold water around the house.



Carl

CMadatMe

Jean

A condensing boiler does not change it's firing rate to track the outdoor temperature. Its changes its firing rate tracking the boiler delta-t. That's the aiming device it uses not temperature. Control the boiler delta-t control its modulation rate. Notice, I'm saying the boiler delta-t. You cannot confuse that with the system delta-t when piped pri sec. You can have the boiler or primary side running on a different delta- then the secondary side. A btu is a btu is a btu no matter if I'm moving it at a 5,10,15,20,30,35, 40 degree delta on the boiler side. You can also have an indirect piped in on the primary side using a different delta-t. In essence you could have a pump sized to run an indirect on a 20 and the boiler primary pump on a 30.



The boiler sides job is to move btu/hr to the secondary and that's it period.



Gordy,



The flaw in condensing boilers is the fixed speed boiler pump. Its sized from most mfgs for a 20 degree delta-t to move a flow rate that is only needed when you need the full output of the boiler. How many days is that? What happens with the rest of the gpm the system side doesn't take away from the primary?



I'd love to see a built in relay onto the boiler. Say one that you could tell it each zones flow rate and your design delta-t so it could modulate based off the system need while controlling a variable speed primary pump.

Jean-David Beyer

A system with a delta t at zero or one is not giving off any heat unless you are pumping the crap out of it.

Well, I do not think pumping an estimated 3 gpm is pumping the crap out of my upstairs baseboard zone. And my installing contractor picked a Taco 007-IFC to do it. Trouble is, the heat loss in that zone is only 2500 BTU/hour there when it is 14F below design temperature, so it does not take much in normal operation. I do agree that I could afford to pump it a little slower, but there are not a lot of slower circulators.



So when you say "If you loops are circulating 3 gpm at 1 degree you are giving 1,500

btu/hour to the space, not including the heat lost in the piping and

circulator." I think you are as close to correct as can be without making more accurate measurements than I have. I do not really know the resistance of that zone. I know there is 64 feet of 1/2 inch copper tubing up there, 24 feet of 3/4" copper tubing, 28 feet of 3./4 inch Slant/Fin, and GOK how many 90 degree elbows (but there must be a huge number. Exposed are 12 of them, and the rest are hidden in the floor, but there must be at least a dozen more.). Then there are the tubing near the boiler that applies to just that zone. Some of that is one inch, some is 3/4 inch, some is 1/2 inch. See why I am not sure about the flow rate?  In any case, at design temperature, I am pretty sure 1500 BTU/hour is about the heat load up there. So, sure I could use 8 suitable incandescent light bulbs up there, except one room is my photo darkroom and I would not normally want any lights operating except some 7 1/2 watt safelights that have to be turned off if I am processing film.



Downstairs, the slab needs more heat. About 24,000 BTU/hour when it is 14F below design temperature. So I do not need a whole lot of heat there either on normal days. Now for that zone, the circulators run almost constantly, meaning 12 to 18 hours per day except when it is warm out, when they run less. They would run almost all the time then, too, if the boiler could modulate down farther. But it goes down only to 16,000 BTU/hour. We know my boiler is too large, but an 80,000 BTU/hour (input) was the smallest one offered by my contractor. I see that now, 4 years later, there are a few that are rated around 50,000 BTU/hour, but I am not going to replace a 4 year old boiler.



"A startup on a cold slab ...." I get greater delta Ts when starting up. But I try to run my slab continuously, adjusting the supply temperature with the outdoor reset, so the thermostat never shuts off. Ideally, that would mean one startup in the fall and one shutdown in the late spring. Of course things are not ideal like that, especially with a 2:1 oversized boiler, so it has to go in on-off mode in the warmer seasons.



When the power failed for 6 1/2 days during storm Sandy, my slab sure cooled off. It did not freeze, though. I got high (more than 10F) delta T from there as the return water was about 50F or something, and the supply was something like 90F.  That meant it would take a very long time to bring the house up to temperature. After about 12 hours, IIRC, I overrode the reset curve to increase the supply temperatures by about 10F. And then it took a couple of more days to get the system stabilized and I could run with normal resets again.



Actually measuring the steady state delta T with a slab is pretty tricky, since any change takes at least 4 hours to matter much, and about 24 hours to stabilize. If the outdoor temperature remains constant for a few days, I can get an accurate steady state measurement, but around here it gets a lot colder at night than in the daytime, and it changes a lot from day to day, so I cannot always get an accurate measurement.

CMadatMe

Why Don't You Just

Pipe in a thermostatic with a bypass for that second floor zone. Get the system running on constant circulation. Problem solved..

Jean-David Beyer

A condensing boiler does not change it's firing rate to track the outdoor temperature.

Maybe yours works like that. Mine does not. It measures the supply temperature to the system, and matches that to the reset curve. It does not measure delta-T when controlling the supply temperature.



You have been saying "A condensing boiler does not change it's firing rate to track the

outdoor temperature. Its changes its firing rate tracking the boiler

delta-t. That's the aiming device it uses not temperature. Control the

boiler delta-t control its modulation rate. Notice, I'm saying the

boiler delta-t. You cannot confuse that with the system delta-t when

piped pri sec. You can have the boiler or primary side running on a

different delta- then the secondary side. A btu is a btu is a btu no

matter if I'm moving it at a 5,10,15,20,30,35, 40 degree delta on the

boiler side. You can also have an indirect piped in on the primary side

using a different delta-t. In essence you could have a pump sized to run

an indirect on a 20 and the boiler primary pump on a 30." for months.



But my boiler does exactly that. It measures the supply temperature (not the delta T) and the outdoor temperature and sets the firing rate so that the supply temperature matches what the reset curve demands. I do not know what your boiler may do. What the controller does to set the modulation rate (proportional plus integral plus differential) is to notice the difference between the desired supply temperature and the actual temperature. If the supply is too low, it fires at the "maximum" rate until it gets close to the reset curve, then it gradually lowers the firing rate to minimum as it goes just above the reset curve and if it gets 5F above the curve, it shuts off the burner completely.



Other than for safety reasons, when heating the house, it does not even measure the boiler output temperature. It measures the supply temperature in the secondary (system) loop. Perhaps the delta-T measured at the output and return to the boiler is different from the delta-T measured at the supply and return of the secondary loop (i would not be surprised), but I imagine they are closely related when the large (radiant zone) is the only one running. When both are running, the boiler loop may run at a slightly higher delta-T. But so what?



And I DO run my indirect across the Boiler loop, and the delta-T when doing that is around 20F, although it drops as the indirect comes back up to temperature, so I never really get a steady state reading of the delta-T.

CMadatMe

Believe What You Wish

Water temp and btu/hr output are two different things.. The only way any modulating condensing boiler operates is based off it's delta-t. How else would fan speed adjust? The boiler doesn't modulate off water temp. Yes even your boiler..



I've had this conversation many of time from a friend who worked at Weil and was part of not only your boilers control but also the current control.

Zman

Different strokes

I am not really suggesting you heat with light bulbs, just making a point.

As I indicated earlier, I don't obsess about delta t.

Radiant loops have so much resistance that it take an enormous amount of energy to significantly raise the gpm. I do not think there is a circ out there that has a pump curve that would allow a range of say, .5 to 1.5 gpm on a 200' length of 1/2" tubing. I personally do not think that radiant loops are a good application for delta t circulators. They will work quite nicely on a series baseboard application.

I also don't think that the nearly constant circ setup you have is very efficient, it has nothing to do with the speed of your circulator. You have said that your boiler is oversized, I assume you do not have a buffer tank. I appears that you are running your boiler in a short cycle condition most of the day. If you turn the reset curve up a bit, the boiler and circ will run less hours with longer cycles. I would be surprised if this has any effect on comfort. I believe your overall efficiency will be improved. You will have less hours of boiler heat loss, less electricity used by the circulator and boiler, and longer more efficient cycles. Raising the temp will reduce the amount of condensation slightly, in my opinion not enough to be concerned about.

Carl

Jean-David Beyer

How is your boiler interconnected to the rest of the system Jean?

Eastman: sorry I am replying at the end, but it will not let me reply to your post for some reason.



I am not sure I understand your question, so let me make some guesses.



The boiler is piped primary secondary, using closely spaced Ts. To avoid confusion, I will call the loop with the boiler in it the "boiler loop." And I will call the loop(s) with the heating emitters, the "system loop."



There is an indirect fired domestic hot water heater connected across the boiler loop. There are two heating zones, a radiant slab downstairs, and somewhat oversized baseboard upstairs.



The system has built-in outdoor reset, that I have set up very tight, to supply the lowest possible water temperatures that can make up the heat loss of the building. So in cold weather, the system runs almost constantly. And since the supply temperatures are low, the return temperatures are low. By low, I mean that the largest zone, the radiant zone, has supply temperatures that vary between 76F and 112F depending on outdoor temperature. I heat the building to 69F. With such low temperatures, almost room temperature, the delta T is low. And when I say the supply temperature is 76F, what I mean is the supply temperature in the system loop. And that has a 10 degree differential, so when I say 76F, I mean it is in the range of 71F to 81F. If it happens to be putting out 71F and the house is 69F, you can imagine how low the delta T would be.



If that is not what you were asking, feel free to clarify your question. Do you want to know the pipe size? Pipe material? Where the circulators are located? If there are Flow Check valves, etc.?

Jean-David Beyer

How else would fan speed adjust?

"The only way any modulating condensing boiler operates is based off it's

delta-t. How else would fan speed adjust? The boiler doesn't modulate

off water temp. Yes even your boiler.."



You do not know what you are talking about.



I already said how else the fan speed would (and does) adjust. The controller compares the supply temperature in the system loop with the temperature specified by the reset curve. If the supply temperature is far below the reset curve, the fan runs at maximum speed, putting a maximum of mixed air and gas into the burner. As the supply temperature climbs up towards the reset curve, the fan speed is reduced. When it matches the reset curve, the burner fires at about 20% of maximum. As the supply temperature rises above the set point by 5F, the burner shuts off. Then as the supply temperature drops nothing happens until it gets 5F below the set point. Then the boiler fires and the process repeats.



Actually, it is not quite as simple as this because the error signal (the difference between the supply temperature and the reset curve temperature) goes through a proportional integral, differential circuit first, but that is just fine tuning. I used to design control systems like that, though we used proportional plus integral more often. Adding differential signals tended to make the systems too noisy (electrically, not sound). I also designed some systems that were proportional plus integral plus integral of that. It was tricky to make them stable, but when they were designed right, they would run with zero steady state error. Sometimes that is necessary, though not when running a heating boiler.



"I've had this conversation many of time from a friend who worked at Weil

and was part of not only your boilers control but also the current

control."



As far as I know, mine is the current control. The U-control (must be version two, the one where you could set the pre-pump interval). But in any case, since you obviously do not know how it works, I assume you misunderstood what your friend said, or he oversimplified what was going on.

Jean-David Beyer

As I indicated earlier, I don't obsess about delta t.

I think you are absolutely right not to obsess about that. I would not either, but I originally though I was missing something because some here do seem to obsess about it. After much, mainly useless discussion, I am now of your point of view, that it is nothing to obsess about. I may go farther than you. Design the system. Perhaps have a delta-T in mind when calculating baseboard size, amount of tubing in slab, etc. for a design temperature, and then take what you get.



"I personally do not think that radiant loops are a good application for delta t circulators."



I do not think so either, but since some people think they are the answer to all problems, I wanted to understand how using one in my slab zone would make any difference. I never found out.



"I also don't think that the nearly constant circ setup you have is very

efficient, it has nothing to do with the speed of your circulator. You

have said that your boiler is oversized, I assume you do not have a

buffer tank."



True: no buffer tank. No place to put one.



"I appears that you are running your boiler in a short cycle

condition most of the day."



Well, that depends. The radiant slab zone, the one that uses the most heat, does not short cycle until it gets up to about 60F outside. Otherwise, the boiler just modulates all the way down and runs for hours.



The baseboard zone does "short cycle." I have improved that condition by adjusting the system settings for the baseboard zone much as you suggest below.



1.) In particular, I raised the differential from 10F to 15F.

This increased the time the boiler would fire, although it increased the expansion noises.



2.) I raised the minimum supply temperature from 85F to 110F.

This meant that more of the heat delivered to the baseboards escaped into the room, so the boiler temperature increased more slowly. It means the temperature varies by about two degrees as the thermostat gets satisfied much more quickly, shuts off the heat, the place cools down, and the heat comes back on. But a two degree range is not uncomfortable.



3.) I lowered the maximum firing rate from the default of 94% to 55%.

I did that because the proportional-integral-differential control used too much differential signal (and that is not adjustable) and the boiler temperature increased much too rapidly and hit the upper limit before the proportional integral component could slow it down. With a lower maximum firing rate, the temperature built up more slowly, giving the control more time to reduce the firing rate.



So now, it goes through around 6 cycles per hour instead of 10. I figure that is not too bad compared to my old oil boiler that went through a cycle every two minutes or so.



" If you turn the reset curve up a bit, the

boiler and circ will run less hours with longer cycles. I would be

surprised if this has any effect on comfort. I believe your overall

efficiency will be improved. You will have less hours of boiler heat

loss, less electricity used by the circulator and boiler, and longer

more efficient cycles. Raising the temp will reduce the amount of

condensation slightly, in my opinion not enough to be concerned about."

SWEI

200' of 1/2" PEX flowing 100F water

Has a resistance of 1.9 ft at 1/2 GPM and 14.7 ft at 1.5 GPM according to the Uponor CDAM.  That's a lot of range to cover.

Eastman

Yeah I just wanted to know a few basics

So there are two secondary loops that tie into the boiler loop with closely spaced tees.  The slab uses much more than an upstairs baseboard zone.  Does it run on ODR alone or is there a thermostat for each zone?  Is there a different reset curve for each zone?

Eastman

@Chris

"Water temp and btu/hr output are two different things.. The only way any

modulating condensing boiler operates is based off it's delta-t."



I'm confused, it seems like you're leaving a key piece of information out.  Is there a flow meter in the boiler?  How would the control system compute btu/hr from only the boiler hx delta?  And what number does it want it to be?

Jean-David Beyer

just wanted to know a few basics

"Does it run on ODR alone or is there a thermostat for each zone?"



There is a thermostat for each zone. I wanted it to run such that the thermostats would always call for heat unless the sun load on the building got too hot, in which case, the thermostat would stop calling for heat. But I cannot quite do that, especially for the upstairs zone where I run the supply temperatures a little too hot to reduce cycling rate. Downstairs, the thermostat has been known to call for heat for 18 hours straight, which is pretty good, IMO. Except on very warm days, the boiler does not cycle much at all for the downstairs zone.



 " Is there a different reset curve for each zone?"



Yes. The downstairs zone runs from 76F to 120F as the outdoor temperature runs from 50F to 6F. The upstairs zone runs from 110F to 135F as the outdoor temperature runs from 50F to 6F. It is a coincidence that the two zones use the same range of outdoor temperatures. That is just what it took to get the reset curves to be as low as possible and still heat the house.

Eastman

steady state delta T

Where were you measuring a delta of around 1 degree?  That's an average value?  Is there a typical delta when the boiler is actually running and consuming fuel?

CMadatMe

If 80,000 Btu/hr

Is 100%, off is 0. Boiler mini is 16,000 or 20% so if you were delivering 30,000 btu/hr you'd be at 37% rate of modulation. How do you know your delivering 30,000 btu/hr?



Primary Pump Flow Rate x (Boiler Supply-Boiler Return) x 500...Not system supply and return.



If my memory serves me correctly that 007 on Jean's boiler is moving 7 plus gpm across his primary.



He can believe that the water temp is dictating the firing rate all he wants. Yes the control logic is looking for a target temp but the control uses the boiler delta as its aiming device to get there not the target temp itself. Water temp is just that water temp. Need a temp difference or your just in equilibrium. If it didn't it would just bang off high limit all the time.



As the boiler delta-t closes the boiler backs off, as the boiler delta-t increase, the boiler ramps. Now the control logic may say when you get to a particular water temp to start backing off but that's not a magic number its because of the boiler delta begins to shrink as the emitter begins to satisfy the room.

Eastman

boiler ∆

What is the boiler delta closing on to? Is there a boiler target temp?  What value does the boiler use for this?

Jean-David Beyer

Where were you measuring a delta of around 1 degree?

There are two strap-on temperature sensors that go to the boiler controller. The sensors are in the system loop (not the boiler loop) outside where the supply from and the return to the boiler are connected. Let me see if I can do some ASCII art.



       S                           S

>---------------------------> to system

                |                |

            TO            FROM

          boiler         boiler





I just read out those sensors, indicated by S.So they indicate the supply temperature to the system and the return temperature from the system. (These are not the same as the supply and return to the boiler unless the flow in the two loops is the same. It might be the same sometimes, but surely not when only the little upstairs zone is running.)

SWEI

Strap-on thermistors

are they insulated?  Have you checked their resistance when both are the same temperature (e.g. during summer)?

Jean-David Beyer

Chris keeps saying this, but that is not how the U-control in the Ultra 3 works.

"He can believe that the water temp is dictating the firing rate all he

wants. Yes the control logic is looking for a target temp but the

control uses the boiler delta as its aiming device to get there not the

target temp itself."



This is just not so. The U-control uses the difference between the system supply temperature and the reset curve to determine the firing rate. Nothing to do with delta-T of the boiler. It does not even use the boiler (heat exchanger) sensors to control the firing rate except when running an indirect heater across the boiler loop. The control is not linear. If the difference signal is large and the temperature too low, it fires at maximum rate. As it gets to within about 5F for the set point, it gradually reduces the firing rate. When it gets to the set point, the firing rate is very low. When it gets 5F over, it turns off the burner.



"Water temp is just that water temp. Need a temp

difference or your just in equilibrium. If it didn't it would just bang

off high limit all the time."



Sure you need a temperature difference to run a servo such as the U-control is, but the temperature difference measured and used is not the supply and return to the boiler, it is the difference between the supply from the boiler, measured in the system loop, and the temperature demanded by the outside temperature and the reset curve.



"As the boiler delta-t closes the boiler backs off, as the boiler delta-t

increase, the boiler ramps. Now the control logic may say when you get

to a particular water temp to start backing off but that's not a magic

number its because of the boiler delta begins to shrink as the emitter

begins to satisfy the room."



I suppose you could control a boiler that way, but that is not how the Ultra 3 does it.



Think about it. If it works as I say it does, then when the outdoor temperature changes, the servo acts to change the supply temperature. If the temperature outside goes down, it demands a higher supply temperature in the system loop. The servo increases the firing rate, perhaps a lot if the outside temperature changed a lot. As the supply temperature approaches the desired temperature, the servo reduces the firing rate, and when it gets to the desired temperature, it would stop altogether except it is an integral-proportional-differential one, so it goes a little over. Only if it goes a lot over does it shut off the burner.



Now if it works the way you say, when the outside temperature changes, what happens? The delta T remains the same. Only after the building cools enough so the return temperature goes down, perhaps 4 to 12 hours later if heating a slab, does your servo increase the firing rate. Too late by far. Then the building starts heating up. Only many hours later, when the return water temperature finally goes up a lot (because the supply finally went up too) does it start to lower the firing rate and thus the supply water temperature. So again, way too late, when the slab is seriously overheated, does the firing rate start to go down. The system damping ratio is too low and you would get wild temperature swings. Unsatisfactory, IMO.

Eastman

like this?

see image

Jean-David Beyer

are they insulated?

Yes, but not well.



"Have you checked their resistance when both are the same temperature (e.g. during summer)?"



I never checked their resistance, but since it is 91F outside right now, and 76F inside, it is safe to say the boiler has not run lately and those temperatures should be pretty close to the same. Unless the indirect ran, and some heat leaked past the Flow Check valve and got as far as the sensors.



OK, the indirect has not run and the system supply and return temperature sensors both read 85F. (It is a little cooler in my garage than outside.)

Jean-David Beyer

Close.

The circulator in the boiler loop is in the return, where W-M say to put it.



The system loop (what you called the supply loop) is connected to the boiler loop by closely spaced Ts, but the two zones are in parallel in the system loop. I.e., there are no more closely spaced Ts in the system.

Jean-David Beyer

What is the boiler delta closing on to?

It is not. The boiler delta T is not a controlled value. It exists and could be measured,but since it is not a controlled variable, it is not closing onto anything.



The controlled variable is the supply temperature in the system loop. And the controller uses the difference between what the reset curve specifies and the temperature that actually exists in the supply of the system loop.



If it were otherwise, the value of the outside temperature sensor would not be used; i.e., the system would not have outdoor reset at all.



The error signal that runs the servo is the difference between the desired (reset curve temperature) and the actual (system loop supply temperature). NOT the difference between the boiler supply and return temperatures. If it were the later, the outdoor reset would just not work.

Eastman

this way?

attachment

Rich_49

differing SWT's

are being achieved by what means Jean David ?

Jean-David Beyer

this way?

Yes.

Jean-David Beyer

differing SWT's are being achieved by what means

I am almost embarrassed to say. And it sure would not work for everyone.



The U-control on the Ultra-3 boiler has three thermostat inputs, each with its own reset curve, and three circulator outputs. These thermostat inputs are in priority order so #1 has highest priority and #3 has lowest.



The indirect hot water heater is on #1. Its "reset" is set at a constant 175F; I would have it higher if I needed a faster recovery rate, but I do not need it.



The radiant slab is on #2,



The baseboard zone is on #3,



Now there is some fancy stuff so if some zone requests for too long, it can be interrupted by a lower priority heat demand to prevent hogging.



When #1 calls for heat, its circulator, across the boiler loop, runs and heats up the indirect. The boiler circulator is off if #1 has control.



When #2 calls for heat, the indirect will shut off, either when it is satisfied, or when 1/2 hour expires and the boiler circulator and the radiant zone circulator turn on. The reset curve is the one appropriate for the slab zone.



When #3 calls for heat, if nothing else is calling for heat, the boiler circulator and the baseboard circulator switch on and the reset curve for the baseboard is used.



The only funny thing is that there are not 4 circulator outputs from the U control, so I have a separate Honeywell relay to run that circulator. A consequence of this is that if both the radiant slab zone #2 and the baseboard zone #3 want heat, they both get it, but the reset curve used is the lower temperature one, the one for the slab. Because I do not want to put baseboard heat into the slab. When this condition occurs, downstairs works normally, and upstairs gets some heat, but not enough. But within 1/2 hour it will get enough. This rarely happens.



What usually happens is that 2 or 3 times a day, the indirect runs for 10 minutes or so, and is otherwise off.

Most of the time, the radiant zone runs. Sometimes the circulator to the baseboard runs at the same time, getting some, but not enough heat. And once in a while (perhaps several hours a day, the baseboard zone runs with its preferred reset.



I get away with this, instead of using mixing valves, injecting metering, etc., because the heat load in the baseboard zone is so low, and it is well insulated and with good windows.

Zman

Measuring?

Jean,

'It sounds like you have your shortcycling under control. I suspect there is a flaw in the way you have measured your delta t on the system side. It would be impossible to get the delta t you describe without short cycling.

Delta t should be carefully considered at the design phase. Not obsessed in the real world.

Glad your system is running so well after the tweaks.

Carl

Gordy

Okay

I feel like Huckleberry in Police Academy one, and I missed out on gun play.





Chris has the boiler end circ nailed down. That delta is quite controllable in what I call the primary, or boiler loop.



I was referring to convectors in my statement initially.



There are many deltas in a whole envelope system.



Inside to outside temp delta

Convector to room temp delta.

SWT to pipe wall delta.

Pipe wall to plate, or pipe wall to concrete, what ever radiant detail medium used delta.

The radiant loop delta.

The supply/Return piping to the radiant loop delta.

And finally the boiler loop delta.



In any one of these delta scenarios the narrower the delta the less heating, or cooling transfer "ill throw in cooling because Chris is dabbling in that arena"



Jean pick one delta floors 10 just for comfort, and less stripping, ceilings 15, baseboard 20 boiler 30 the narrower the band the more fluid you have to move and pumping costs go up, but it does not mean that what you end up with in real world.





My house has ceiling radiant, and floors. Floor deltas are a solid 10' but there is a reason, and that's because I have radiant ceiling in those rooms also.



Radiant ceiling loops can be anywhere from 5-15, depending on loop location. Example being a loop that is toward an exterior wall verse a loop that is central to the room, verses a loop again that has radiant floors also.



Now my supply return to the radiant ceilings is always a rock solid 15 no matter the outside temp once heating season is rolling. Exception to this is shoulder season spring, and fall when the house just needs the occasional bump.



So Jean your emitter is in place its worked for a long time. you have to tweak your boiler end like Carl said bump the reset higher it can't hurt.

Jean-David Beyer

I suspect there is a flaw in the way you have measured...

"I suspect there is a flaw in the way you have measured your delta t on

the system side. It would be impossible to get the delta t you describe

without short cycling."



I could accept this. But you know how I have measured it. You know that today, when the heat is surely off, both supply and return read the same temperature (85F I think it was). You know where I measured the supply and return temperatures in the system loop  ( somewhat outside where the two connections to the closely spaced Ts come up from the boiler).



We have one data point. With the boiler essentially off, the temperature sensors seem to be correct. And sometimes with the  radiant slab running, it no doubt runs 85F water. That is not a full calibration of those thermistors but it is something. Nor is it a full calibration of the a-to-d  converter in the U-control that reads it out as temperature in degrees F. It is unlikely to be an error in the U-control, because it read the same with a previous U-control. The factory rep gave me a free new one because the original one went under water due to a mistake by my former contractor when the boiler was installed.



So what is the source of the error? If we assume (doubtful assumption) that the flow in the boiler loop is the same as the flow in the secondary, then there is no backflow between the supply from the boiler to the return to the boiler through the closely spaced Ts. This may be nearly true when only the radiant slab runs. Because the boiler loop and the slab are both served by Taco 007 circulators. It would be true if they had the same head, but I have no reason to believe that. If the slab has a bit more head, then the delta T would be closer to accurate when both zones are running at the same time, but I notice low delta Ts then too. If only the baseboard zone runs, I assume quite a bit of water flows backward through the closely spaced Ts. But it should go down into the return to the boiler, not further back and hit the return temperature sensor about a foot farther back in the system loop. All the piping up there is 1 1/4" copper and W-M specify it should be at least 1-inch. W-M further specify that those Ts be no more than 12 inches apart, and they are about 5 inches apart.



If I look at the tridicator dial, it reads about the same temperature as the digital readout of the system loop supply, but that does not give a temperature difference, just one temperature. And that dial is about one inch diameter, so I cannot possibly read it to one degree of accuracy.



So I do not see where the error is occurring.



Now if I say the boiler fires at 16,000 BTU/hour input when modulated all the way down, and it thinks it is putting out 76F water and that 76 water is returning, perhaps it is putting out 78F water and 74F water is returning, for a delta-T of 4F.. Based on the assumptions that the pipe in the slab are a foot apart (no clue), and so on, ALL UNVERIFIED, I calculated that there is around 3 GPM through there. But if you said I was off two to one, I would not be able to contradict you. So how much heat will 3 gpm carry at a delta-T of 4F?



Q = 500 * f * deltaT

    = 500 * 3 * 4 = 6000 BTU/hr



In other words, it would cycle rapidly since it would not carry away enough heat. On the other hand, 6000 BTU/hour might be enough to heat the house when it is 50F outside, so we may be in the ball park.



If the flow is actually 7 gpm (I do not suppose it is that high, but I have no plans for the tubing in the slab), it could transfer 14,000 BTU/hr. Now if, in addition, those sensors are off 1 F each in the direction of giving greater delta T, I would get 21,000 BTU/hour and no rapid cycling. Even without the last possible error, if it got colder than 50F outside, and I would be running hotter water and getting a greater delta T, ... .



Things seem close enough, that unless I could make very accurate measurements of flow and temperature, I fear this is going to be the best we can do. I have no flow meters, I do not know the head of the piping, those thermistors seem OK but perhaps they are not far enough away from the closely spaced Ts (they are as far as they can be), ...



"Glad your system is running so well after the tweaks."



So am I, of course. The proof of the pudding is in the eating. The comfort level is much higher than with the old GE boiler. There is no overshoot when heating with the slab, and it was considerable before. My gas bill is much less than my oil bill, the boiler does not cycle every 120 seconds like the old one did, and so on. The gas bill does not really prove much because I very much doubt that the heating degree days were the same for the last bunch of years. The bill implies, but does not prove.



If I had the room and money for a buffer tank, I might put one in for the baseboard zone, but I think it would have to be a 4 port tank running as an hydraulic separator. But no room and no money.



Similarly, if one of my zone circulators died, perhaps I would replace it with an ECM one, but I guess a constant speed one would be good enough (if they made one), or one that had a speed control knob. I suspect I could run them with less flow than the 007s to get a greater delta T. But it would be touchy because upstairs is all in series and if I have too much delta T, one room would be a lot colder than the other.



Thank you, Carl, for your efforts at clarifying this.

Jean-David Beyer

I feel like Huckleberry in Police Academy one,

My head hurts too.



"So Jean your emitter is in place its worked for a long time. you have to

tweak your boiler end like Carl said bump the reset higher it can't

hurt."



The emitter downstairs (the copper tubes in the concrete slab) sure have been there a little over 60 years, and unless they start leaking, there they will stay. But that zone is not causing me any problems, so there is no NEED to change it. If I was convinced that picking some delta-T, other than whatever it runs at now, would lower my heating bill enough to get me a payback before I die (in less than 30 years), there seems no need to change anything.



The upstairs zone cycles more often than I would prefer, especially when it is warm out, but with the tweaks Carl suggested (that I, in fact, already did) reduced the cycling rate to about 6 cycles per hour, and I am willing to put up with that. Remember the old boiler cycled about 30 times per hour and lasted over 60 years (except for the burner that was replaced with a Beckett about 30 years ago).

knotgrumpy

Variable Speed ECM Circ

"Similarly, if one of my zone circulators died, perhaps I would replace it with an ECM one, but I guess a constant speed one would be good enough (if they made one), or one that had a speed control knob."



I'd suggest you look at the Xylem (B&G) Ecocirc 19-14 Vario.  Excellent product.  You can easily set the speed you need. 

Mark

Jean-David Beyer

I'd suggest you look at the Xylem (B&G) Ecocirc 19-14 Vario.

Those look really good if you need (adjustable) constant speed or delta-P circulators. I like that they seem to have magnetic dynamic bearings instead of the more usual mechanical ones.



I will wait until my 007 ones seize up, though.