To get email notification when someone adds to a thread you're following, click on the star in the thread's header and it will turn yellow; click again to turn it off. To edit your profile, click on the gear.
The Wall has a powerful search engine that will go all the way back to 2002. Use "quotation marks" around multiple-word searches. RIGHT-CLICK on the results and choose Open Link In New Window so you'll be able to get back to your results. Happy searching!
In fairness to all, we don't discuss pricing on the Wall. Thanks for your cooperation.

Advantage of Outdoor Reset?

KestrelKestrel Posts: 102Member
Hey gang, I'm trying to finalize the design of my radiant hydronic system.  I've had LOTS of help here, particularly on the heat loss and sizing issue.  I'm getting very close to buying parts, and need to commit to a design.

To help me understand, is the advantage of modulating boiler output with ODR that it delivers fewer BTUs per unit time when heating needs are less, such that heat delivery over time is close to equal to heat loss over time?

Would the disadvantage of using a constant water temperature across the heating season, say 150' all the time, be that during lower demand times, the emitters get hot and do their job in a shorter time - thus leading to up and down swings in room temp and short boiler cycles (if there is no buffer)?

This whole issue is driven by a new thought about future solar input.  Original plan was TT 110 Solo driving both a TT Smart 60 and a separate heating circuit of rads/floor buffered by a Boiler Buddy 30gal.

New thought would be TT 110 Solo heating a TT SME 80, a tank with a 30gal primary volume, a 70gal domestic volume, and a separate coil for solar input.  This would necessitate keeping the heating and DHW the same temp.

If my worries above about the constant temp system are valid - would a tempering valve like a Taco i-series attached to ODR downstream of the buffer and upstream of the heating emitters solve the problem? 

Alternately, would a DHW tank with a solar coil (like a Heat-Flo or Stiebel-Eltron) in place of the TT Smart 60 in the original plan be a better solution?

· ·


  • GordanGordan Posts: 870Member
    There are comfort and efficiency advantages to ODR

    Comfort, as you've said, is that "modulating" the supply water temperature to what's required keeps emitter output, and therefore ambient temperature, relatively steady compared to bang-bang operation. You can get this with an ODR mixing valve.

    Efficiency gains are pertinent to condensing boilers, solar, heat pumps etc., all of which can operate more efficiently at lower supply (and return) temperatures. You CAN'T get this with an ODR mixing valve - the heat source itself has to take advantage of the fact that the system doesn't require a high supply temperature, by actually supplying lower temperature water. Heating up past the required temp, only to mix down subsequently, misses this benefit.
    · ·
  • Jean-David BeyerJean-David Beyer Posts: 2,435Member ✭✭✭
    Outdoor reset.

    "To help me understand, is the advantage of modulating boiler output with

    ODR that it delivers fewer BTUs per unit time when heating needs are

    less, such that heat delivery over time is close to equal to heat loss

    over time?"

    By itself, that is no advantage. There are comfort advantages to ODR and cost advantages. These are a consequence of being able to vary the delivered hot water temperatures with ODR.

    "Would the disadvantage of using a constant water temperature across the

    heating season, say 150' all the time, be that during lower demand

    times, the emitters get hot and do their job in a shorter time - thus

    leading to up and down swings in room temp and short boiler cycles (if

    there is no buffer)?"

    I used to have a non modulating boiler that delivered most of its heat to a huge concrete slab on the ground with copper tubes embedded in the concrete. Because when firing, the boiler delivered a constant temperature to the slab, its temperature had to be hot enough to heat the house on the design day, and it was actually likely to be even hotter than that. Now with a thermal mass like that, the temperature would swing between about 66F and 75F because when the thermostat called for heat it would take 4 to 8 hours to warm up the slab to deliver the heat, and by then the temperature of the house would drop several degrees below the thermostat set point. When the room finally got hot enough to satisfy the thermostat and turn off the hot water, the slab was so hot that it continued to heat the house for another 4  to 8 hours. With a thermal mass like that, there is no such thing as faster recovery because of higher water temperatures. All there is is wilder temperature swings.

    I switched to a mod-con with ODR built in. It solved the problem of huge temperature swings. It now holds 69F +|- 1F. That is entirely due to the ODR feature.

    Now it happens that the boiler modulates its firing rate depending on the temperature outside and the temperature of the water delivered to the load. When it is warm outside, it delivers lower temperature to the load and rather than running in OFF-ON mode, it reduces the firing rate so as to deliver just the temperature needed to recover the heat loss. This allows the boiler to run longer without the need for a buffer tank. Actually a buffer tank does not make much sense if there is a huge thermal mass (the slab); I would need a buffer tank whose thermal mass was large compared to the thermal mass of the slab to make much difference.

    The fact that the boiler modulates is a big reason it saves money. It allows the boiler to run continuously for longer periods of time than an OFF-ON system. With the old boiler, it would fire for about 30 seconds, and off for about 45 seconds. I must have had a very good aquastat to survive so many cycles. I managed to diddle the aquastat and increase the size of the dead zone, but it was still guilty of perpetual rapid cycling. With a modulating feature, this is pretty much eliminated.

    Condensing also saves money, though this is perhaps 10% or a bit less. And with modulating feature, it condenses more because the amount of condensing is pretty much determined by the temperature of the water returning to the boiler. The lower the supply temperature, the lower the return temperature. I could not do much about the slab, but I have another zone that is heated by finned tube baseboard, and I increased the lengths of the baseboards (one in each of two rooms) from the original 3 feet to 14 feet, so I run a maximum supply up there of 135F instead of the usual 180F. And that 135F is needed only when the outdoor temperature is below the design day temperature.
    · ·
  • KestrelKestrel Posts: 102Member
    Gordan and JDB

    Thanks for the insights!

    Per this and the other thread I posted - I'm coming back to the idea of having distinct DHW and heating circuits.

    As I noted on the other thread, I'm still concerned about whether I need a buffer tank like a Boiler Buddy.  I can get hydraulic separation with any number of devices - I want to avoid short cycling, should one of the smaller rad/TRVs call for heat.

    OTOH, perhaps I should go smaller, like a TT 60 Solo, though this is very near the limit of my design load, and I worry about going too small.
    · ·
  • LarryCLarryC Posts: 331Member
    You have two seperate issues.



    You have two seperate issues.  Domestic HW that needs a heat source that is greater than 140 F, to minimize Legonella bacteria, and space heating.

    The ODR works with the space heating.  The domestic HW should not be affected by the ODR.  I belive the cleanest setup would be a indirect tank for the domestic HW. 

    A second tank would be the buffer tank for space heating with a single coil for domestic HW preheat or future solar expansion.  OR a buffer tank with two coils: one for domestic HW preheat and the second coil for future solar expansion.

    The indirect HW tank gets priority and full fire when it calls for heat.  The buffer tank provides the space heating load and a variable speed pump controls how fast heat is delivered to the space heating loads.  When the buffer tank is too cold to provide the space heating loads, then the boiler fires up and modulates to heat the space.  If the boiler can't go to a low enough burn rate, the excess heat is deposited back into the buffer tank.

    That way the domestic HW tank stays hot enough to remain bacteria free, the buffer tank decouples the boiler from the space heating loads to keep it running as efficient as possible, you have future expansion potential for an external heat source such as solar, and the indirect gets to harvest unused space heating heat by preheating the water.
    · ·
  • GordanGordan Posts: 870Member
    TRVs won't call for heat

    With TRVs, you need one zone (typically your biggest load) to NOT be on a TRV but on a thermostat or some other type of indoor feedback. That device will act as your call for heat. Once heat is being supplied, the TRVs on other zones will throttle the "excess" heat supply in order to keep their zones from overheating. A TRV can't keep a zone from underheating, so make sure that your "master" zone isn't satisfied too long before the others have a chance to get there, and that there's enough flow through that zone for minimal modulation. Sizing emitters to room heat loss will get you most of the way there, and balancing valves on the manifold will do the rest.

    I would not worry much about undersizing the boiler.
    · ·
  • Bill ClintonBill Clinton Posts: 35Member
    Downsides of ODR

    IF your house has only one zone, and IF you leave the heat on all the time during the heating season, ODR is the best thing since sliced bread.

    Around here (San Francisco Bay Area) the norm is fairly large houses, 4,6,8,13,15 zones, etc and the weather is mild enough that it is both fuel saving and practical to turn the heat off when there is no one home. The house can go dead cold. Then, when the occupants return on a day with, say a 62 degree outdoor temperature, the temperature inside the house may be equally cool.  With ODR, it will take forever to warm the house.

    Similarly, if there are one or two unoccupied zones, maybe reserved for guests, they can get quite cool.  Same problem happens--very slow response.  We have had to return to jobs and remove the ODR to solve these problems.

    Bill Clinton, Bay Hydronic, Inc
    · ·
  • Jean-David BeyerJean-David Beyer Posts: 2,435Member ✭✭✭
    I do not live in San Francisco,

    and I am not a heating contractor. But it seems to me that here in New Jersey in September, heating conditions are no doubt much like they would be in San Francisco later on in winter. It is 71F out today, but it went down to 63F last night. I have my thermostats set to 69F all the time downstairs (radiant heat from slab) and I turn down the upstairs at night to 67F.

    As a practical matter, my house is quite well insulated, so its temperature went down to only about 72F. but when it gets a little colder out, the thermostats might call for heat. My ODR is set to pretty much as low as possible consistent with getting enough heat. That means that for my radiant zone, it will put only 75F water into the slab until the outdoor demperature drops to 51F and then it slopes up to 112F when it gets down to the design temperature of 14F.

    I leave the boiler on all year around since it also is attached to an indirect fired domestic hot water heater. But because of the design of the boiler -- cold start for one thing and a water volume of only 3 quarts, I do not think I would save a noticable amount of energy by turning it off during the day. First of all, the house might lose a degree or two, but hardly dead cold. And it might take a while to warm it back up  because it takes a long time to warm up the slab if it had cooled off. If it were at the 62F you used as an example, the boiler might not fire all that day anyway. So where would be the savings?
    · ·
  • KestrelKestrel Posts: 102Member
    Managing Heat Call

    Yes, that is something that was bothering me - it seems the TRVs can get you flow - via "opening a hole" in the pressure that the ECM pump responds to - but how to get the boiler to fire?

    I was thinking that in my original plan, the buffer tank would have an aquastat and that boiler firing would be controlled by the temp of the buffer volume.  That way, the boiler responds, controlled by ODR, to temp in buffer, and separately, the heat is delivered by the ECM downstream of the buffer responding to opening and closing of TRVs.

    The radiant/underfloor on the main floor would be controlled by a t-stat and floor sensor, and if that comes on, then that would control a valve between the ECM pump and the manifold for the tubes in the floor.

    Would that work?
    · ·
  • KestrelKestrel Posts: 102Member

    We are, alas, a bit less temerate than the Bay Area (I hear all the East Coasters guffawing at the distinctions in our West Coast weather!).

    We tend to leave the heat on, as the 40-50'F rainy season seems to last forever.

    I'm coming back around to the notion of DHW maintained at 150'F, and a separate CH managed by ODR.  I'm still wrestling with whether I need a buffer.  I'm wondering, in my 2100 sq ft house with about 55000 BTUH design heat load (20'F), whether I need the TT Solo 110 that has been repeatedly proposed, or if a smaller Lochinvar or Viessmann would be better.

    On that note, what is the difference between the 100 series and 200 series Viessmann?  For my heating need, would a WB2B-26, or a WB1B 10-26, or a Lochinvar 17-85 model be better?  Is there less chance of short cycling with a smaller boiler?
    · ·
  • GordanGordan Posts: 870Member
    That should work beautifully.

    That way, you're truly constant circulation from the buffer to the system (modulated by TRVs and the dP circulator) and part-time circulation from the boiler to the buffer. It's the best of all worlds except one: you need a buffer. :-)
    · ·
  • KestrelKestrel Posts: 102Member
    Right - back to the question of a buffer...

    On the other thread you suggested a way to avoid the need for a buffer - keep the ODR curve 'low' so as to keep the TRVs open most of the time - so that small amounts of heat are being delivered nearly constantly - right? 

    So that avoids the dreaded situation where a small 'zone' or single radiator demands heat and the big boiler turns on and off rapidly - yes?  Because most/many of the radiators are demanding BTUs at any one time?

    It seems the overall goal is to have the boiler firing much of the time, at the lowest end of the modulation.

    Is this what Mark has referred to when he says ModCons "like to work uphill"?  That you want to design a system that has them firing most of the time?

    So if my design heat load at 20' is 55000 BTUH, but most of the time here in rainy Seattle it's 45', or halfway to my indoor target of 68, I'll be needing 28000-30000 BTUH, I'll be near the lower modulation of a TT110 or a Vitodens 200, so that a boiler like that will be firing much of the time at lowest modulation to keep up with the heat loss.  Yes?
    · ·
  • KestrelKestrel Posts: 102Member

    I think with the above set of questions - I'm trying to convince myself that the TT110 is appropriate, and further, that I do not need a buffer tank

    · ·
  • GordanGordan Posts: 870Member
    edited September 2011
    You don't need a buffer

    A buffer renders moot any worries of 1) adequate flow through the boiler and 2) under what circumstances should there be a call for heat. All else equal, you'd have a system that requires less tinkering to get working well enough. But you don't need it, and if you can't afford it, space- or money-wise, don't think that you're somehow doomed; you just have to be more careful with your design.

    The big win was getting you off of the combined DHW/space heating buffer kick. :-) In my mind, the current buffer/no buffer quandary really boils down to how much you want to sweat the details of your emitters, flows, etc. If you're the tinkering kind, as I am, you might enjoy taking the minimalist approach and still getting everything to hum along harmoniously. If you'd rather set it and forget it, get thee a small buffer and find somewhere to put it.

    One more thing: you say above that you'd fire the boiler via aquastat, but that's not going to work with the built-in ODR in the Prestige. An aquastat would have a fixed setpoint. So, you'd need some other way to generate the call for heat. The Vitodens WB2B (and WB2A) have the external sensor option which would be located on their low loss header, but you could locate it on the system outlet of the tank. That would work with the built-in ODR controller. Other boilers may have something similar; I'd look at the new Knight WHN fire tube models.
    Post edited by Gordan on
    · ·
  • KestrelKestrel Posts: 102Member
    Do I like to tinker? Hmm....

    Hmmm, let's see...I'm a homeowner who reads hydronics textbooks and reads/posts here. 

    Yes, I like to tinker - HA! ;)

    So my aquastat idea was to solve the issue of triggering the boiler to fire when the TRV/ECM controlled rads wanted heat.

    The TRV/ECM relationship controls flow, clearly, but where do the BTUs come from?

    I was thinking that a probe on the buffer tank, when it fell below the ODR curve, could be configured to trigger the burner and boiler side pump (internal on the TT 110).  i was calling that an aquastat - perhaps my terminology was wrong.

    Without the buffer, I'm still left with how to trigger the heat.  You say a probe on the emitter side of the hydraulic separator?  That seems sort of like what I was calling an aquastat above.  Conceptually, it seems that the ODR sets a curve for delivery water temp, and if the water in either the buffer tank or the piping downstream of the LLH/closely spaced Ts falls below this, then the boiler fires, right?

    I'm looking at the Knight WHN-085 - the one with the fire tube that modulates between 17-85000 BTUH.  Is that sort of control system built in, or would I need some sort of external device?
    · ·
  • GordanGordan Posts: 870Member
    It's like this...

    Without a buffer tank, you need something to tell you that the indoor temperature is not keeping up; that is, it's falling below some setpoint, but that setpoint is fixed - it's your desired indoor temp. That will fire the boiler. This can be a simple on-off thermostat wired to the T-T terminals.

    With a buffer tank, you need something to tell the boiler that the buffer tank (or: system supply) is falling below some setpoint. But, that setpoint - the desired supply temp - is dependent upon the outdoor temperature (a proxy for heat loss.) So you can't use an aquastat because you're not dealing with a fixed setpoint. You need an ODR controller that can deal with an additional supply temp sensor input (system supply in addition to boiler supply) and the PS will not do this for you.

    There's another arrangement that might work, and that is a constant call for heat. Here, the boiler AND system would continuously circulate through the buffer (except during DHW calls, which take priority) and the boiler would continuously sample the water temp. As the system slowly drains heat from the buffer and the water temperature drops below setpoint minus differential, the burner would kick on and modulate appropriately. You'd have the small extra electrical usage of a boiler circ that's always on vs. mostly on.
    · ·
  • GordanGordan Posts: 870Member
    Btw, from the Knight installation manual, page 61

    Programmable controlling sensor

    The control module is programmed to use the outlet sensor as

    the control sensor by default. If a system supply sensor is

    connected, the control automatically uses it as the control


    The system supply sensor is what you'd put on the system outlet of the buffer tank.
    · ·
  • NRT_RobNRT_Rob Posts: 1,000Member ✭✭
    constant circ/outdoor reset

    typically you just jump out the heat demand so the boiler is constantly "demanded". it then maintains whatever volume of water flows through it at whatever temperature it determines is necessary by ODR. it turns off when it hits "warm weather shutdown" or by user switch perhaps.

    we use a "setback thermostat" in the house so someone can turn it down to, say, 50 while they are gone. that kills the heat demand until the thermostat sees 50. when they get back, turn it up to 90 or some other temp it will never see and constant demand is enabled.

    In San Fran... if the clients have warm weather shutdown on their ODR control, they could just let their own systems determine when to turn on and off. No need for "cold start" and long warmup times. ODR is perfect for "slow start" like that.
    · ·
  • KestrelKestrel Posts: 102Member
    I'm getting there...

    I like the idea of having and ODR and a system sensor to control the boiler firing/boiler pump pumping.

    It seems simple and elegant - the boiler keeps the water temp appropriate based on outdoor conditions, and the system circulates when the individual radiators call for flow.

    Ideally, the water temp only falls when there is flow through the radiators.

    When you say "PS won't do that" you mean Prestige Solo?  That particular boiler does not have the control capability to do this?  While the Lochinvar does?
    · ·
  • GordanGordan Posts: 870Member

    The Prestige only has terminals for a single supply sensor and there's nothing in any documentation for it that I have seen, that would suggest that you can or should hook up an external system supply sensor to it. The Knight documentation is pretty explicit about the use of such a sensor, and their diagrams show it.

    The Knight also has better modulation, and a much more versatile controller. I have - and like - the Prestige Solo 60, but my own setup is pretty bare-bones - as I may have mentioned, I simply pump through the boiler with an ECM circ; no buffer and no primary secondary; so the boiler supply IS the system supply. For my uses, the Prestige's low head loss heat exchanger was more of an asset than a sophisticated controller, so it was a better choice than the non-firetube Knight, which is all that was available at the time. Nowadays, I'd probably choose the firetube Knight, as it seems to have the strengths of both.
    · ·
  • KestrelKestrel Posts: 102Member
    the Lochinvar Knight

    On a thread from several months ago Brad White was HUGELY helpful to me in getting my heat loss calculations correct.  In that discussion he also recommended the Knight, given its lower modulation and 'just right' max output in the 80k range.

    It was only today that I found a local distributor for a WHN (the wall-hung fire-tube models), and as it makes a comparable dent in the budget,  I think that's the way I'll go.  (I do have to buy a circulator, however, unlike the TT)

    My last big unanswered issue with this whole design regards the radiators and design water temperature - though I think I'll start a different thread!

    Gordan, your advice has been invaluable - thank you!
    · ·
  • GordanGordan Posts: 870Member
    You're quite welcome

    One last piece of advice on this topic: don't overpump through the boiler - and it's easy to do, due to the low head loss and pretty low flow requirements. It could shorten your part-load cycles by making the boiler modulate up, to make its setpoint with same delta T and more flow, when a lower modulation with less flow and a longer cycle would have been just fine. On my PS60, the DHW circulator pumps through the boiler and the reverse indirect tank - both of which have comparable pressure drops to what you'd be looking at with the WHN and the buffer - and the Grundfos UPS 15-58 at its lowest setting provides plenty of flow for the PS60 at full output.

    If you think about it, ideally your boiler flow and delta-T would match your system flow and delta-T for any heat load that exceeds the minimal boiler output, so that in those circumstances you're bypassing the buffer - it's not needed - and directly coupling the boiler to the system. That's not going to happen because of three things: 1) you have fixed-speed circulation on the boiler side and variable-speed circulation on the system side, 2) even the fixed speed circulator will be the closest available match and not an exact match for what your calculations tell you you should use, and 3) calculations are one thing and what actually happens is something else. But, you can strive for a good compromise. If you figure on 20 F delta T at your 55 kBTU design load, that's about 5.5 gpm. Calculate the pressure drop through your boiler circuit at 5.5 gpm - probably around 2 ft - and you'll see how little a circulator it takes to get that to happen. A UPS15-10F would do it, with 25W. (Also a UPS15-42F on low speed, but that's 45W.) Those are pretty much the smallest affordable circs that you can find, so it doesn't really make sense to think about lower flows through the boiler circuit and what the tradeoffs may be, but you've got some headroom there in case your head loss turns out to be somewhat higher - you could certainly stand to have a bigger delta-T through the boiler circuit at design conditions, and in fact it may be a good thing.

    I hope I didn't confuse you with too many "it depends"-es...
    · ·
  • KestrelKestrel Posts: 102Member
    Good Points

    I had not been thinking about the boiler circulator specifics until now - as the TT110 I'd been thinking about has a built-in  Grundfos 15-58.

    At the end, when you refer to larger boiler-side delta T - you mean having the return temp be as low as possible, to allow condensing mode to occur as much as possible?  When is there a need for multiple speeds on the boiler pump, and how dose the control work - that is, what parameters control the variable speed, or is it set by the installer/operator and then left there?
    · ·
  • GordanGordan Posts: 870Member
    Don't think of them as multiple-speed circulators

    They're fixed-speed circulators with three selectable speeds. If I were in a snarky mood, I'd say that they're needed when you're not confident in what you're doing or when you don't want to stock too many parts. You're not going to stand in your boiler room to flip the switch as your heating load changes (I hope!) and the selection is not finely-grained enough anyway. Comparing the electrical usage of the two I mentioned for basically the same amount of work, the 15-10 looks like the optimal choice. 25W is piddling.

    As for boiler delta-T... the buffer has a limited capacity. When the heating load is below minimal boiler output, the boiler will be heating up the buffer with water that's at the ODR setpoint or a little above. If there's "too much" flow through the boiler, that means that the boiler has to modulate up in order to bring its supply temp where it needs to be. The result will be that it will recharge the buffer in less time, leading to a shorter on-cycle.

    On the other hand, if there's "too little" flow through the boiler, then at some point the advantage of a slightly lower return temperature will be outweighed by there not being enough turbulence for efficient heat exchange at your max heat load (and corresponding burner output.)

    So, with fixed-speed circulation through the boiler, you want to make sure that you have adequate flow but not too much.
    · ·
This discussion has been closed.


It looks like you're new here. If you want to get involved, click one of these buttons!