Boiler modulation question

[MikeG88]

When a boiler modulates I understand it is reducing the firing rate. Is the most obvious evidence of this seeing the boiler drop the supply temperature below the set point? I would assume this happens when the delta drops below a certain number, and the boiler reduces the supply temperature rather than cycle the burner off. I would assume the turn down ratio would be the ratio between max supply temperature and minimum temperature it can drop itself to?

I've witnessed that when the boiler occasionally (short) cycles it will restart and fire up to about 5 or 10 degrees below set point, but I've also noticed that it may drop a significant amount without the burner shutting off, I recently noticed it drop 20 to 30 degrees and slowly start to creep back up but the t-stat was satisfied before it got back up to the set point.

 

[Dana]

Most modulating boiler controls are set to control the output setpoint within a narrow range of temperatures. Some boilers can be programmed to keep the INPUT temperature within a range. (Most Naviens can be programmed by the user to control either the output or the entering water temperature.) I'm not aware of any boilers that control to a delta-T.

Most boilers have a minimum time-out between firings, and on some that is a programmable feature. On a short-cycle once the flame is out, it waits the minimum time-out period, even if the water temp drops well below the set point.

When controlling to an output set point there is a few degrees (sometimes programmable) above the set point that the boiler will allow, and a few degrees below the setpoint before it will re-fire. These are sometimes independently programmable, but on some boilers it's hard-programmed fixed numbers for the setpoint over & under shoots pre-set by the boiler manufacturer, not adjustable by the user.

When the radiation can't emit heat as fast as the boiler is putting heat in at the minimum firing rate, the temperatures will rise above the set point, eventually ending the burn cycle when it hits the overshoot limit, even if the thermostat is not satisfied.

 

[Sponsor]

 

[MikeG88]

Thanks Dana. Last night after a somewhat mild day, I noticed the boiler drop supply temp from around set point by 20 or more degrees. But it didn't look like it shut the flame off like a short cycle, maybe it dropped back a stage? The supply temperature did start to climb but the heat call ended before it got back to set point. I am wondering what this is?

 

[Dana]

More information about the boiler & system might be useful in explaining the observed behavior, such as:

Boiler model #?

Outdoor reset control (Y/N)?

Amount of radiation & radiation type?

The boiler's setpoint when the 20F dip was observed?

Outdoor temperature at which the 20F dip was observed?

Design heat load, and the outdoor design temperature?

 

[NY_Rob]

On a cold start where you rads and pipes are maybe at room temp or below because there has been no call for heat in a while... there is maybe 60-80F temp water entering the boiler's heat exchanger. It will take some time for the boiler to get up to supply water temp setpoint with that cold return water coming in.
Once the now circulating water makes a few cycles through your radiators & boiler it's now is much warmer... you have a narrower delta between supply and return temps and it will help the boiler come closer to hitting setpoint. If you have long runs, and the supply water is at setpoint but the boiler is still firing on high to maintain it- you can get return water that's near setpoint returning to the boiler which will cause the heat exchanger to run at temps higher then setpoint. If that happens, the boiler will try to modulate down to prevent setpoint overshoot- but sometimes the return water is too hot, and the boiler can't modulate down fast enough or low enough so the boiler stops firing instead. Once the water in the loop cools below the setpoint minus differential it will fire up again.

Fin tube radiators are especially prone to this roller coaster phenomenon because they are low mass and can't absorb lots of BTU's quickly like cast iron radiators. One min you have 80F return water the next you have 100F return water, with cast iron rads.. they heat up very slowly so the return water temp rises very slowly over time and there's virtually no chance of overshooting the supply water temp target... and not much chance of cycling due to chasing supply/return water temps.

 

[NY_Rob]

Last night after a somewhat mild day, I noticed the boiler drop supply temp from around set point by 20 or more degrees. But it didn't look like it shut the flame off like a short cycle, maybe it dropped back a stage? The supply temperature did start to climb but the heat call ended before it got back to set point. I am wondering what this is?

I'm south of you on LI, we had the same mild temps yesterday- which means no boiler activity for some time- which mean cold radiators and cold water in the supply and return pipes. So on startup your low supply temps were due to cold return water coming in to the boiler. After a few loops around the circuit the water warmed up, the rads put out enough heat to satisfy the t-stat and that ended the call for heat. Because of the mild temps yesterday- you didn't need to hit setpoint to overcome heatloss.

 

[MikeG88]

Thanks to everybody for your replies.

Dana: here is the information you asked for and anything else that I can think of that might be important.

The boilers is a Navien CH210ASME, No outdoor reset. Single zone plumbed the usual primary/secondary. Boiler in the unheated basement. Circulator pump is a TACO 00R-MSF1-4IFC (set to LOW). Feeding 54' of standard slant fin type baseboard. The house is a modular just over 5 years old (2 x 6 construction and insulated to or exceeding code), single level ranch 24' x 48'. Trying to trace the water circuit around the house on the floor plan I approximate its about 175'. It's 3/4 pex when not in the baseboards. The blueprints spec the heat calcs at a 90 degree differential (probably because it's it modular that could go way up north?) At the 90 degree diff it's 28,196BTU. So for CT a 60 degree differential is probably more accurate - so 2/3 of the 28,196= 19k for rounding.

It was about 35F outside and the boiler was set for 170F. Not sure how much time had elapsed since last heat call- could have been anywhere from 20 minutes to an hour. I was working in basement near the boiler and happened to look at the digital temp readout, It had probably just reached set point (it usually hunts about a bit around set point 2-3 degrees either way for a period and then settles in 1 to 3 degrees below). I think the readout dropped at least to the upper 140's and then started to climb back up (but the t-stat call ended before reaching set point again). The burner light never went off or I would have thought it a short cycle.

I wonder if (as somebody else speculated) if that was the point where water that had been heated more than the rest was re-entering the boiler -or on the converse could much colder water that hadn't been around as much, and the boiler had just settled down at set point but now had to start firing higher again? Not sure how long it takes the water to make the trip once around the circuit...

Thanks again for any thoughts/ideas etc.
Mike

 

[NY_Rob]

Without an outdoor reset sensor- you're working at a fixed supply water temp regardless if it's 50F or -5F outside.

FWIW- you're missing out on a big efficiency boost by not using an outdoor reset sensor.
Exhaust gas will only start condensing with return water at and below 130F... you'll never see that with 170F supply water.
So you have a condensing boiler that will never condense.

 

[MikeG88]

NY_Rob- you are obviously correct. But another user here that I have corresponded with had the outdoor reset on this series of Navien and could never really dial in a good curve (I believe they only have limited presets). Maybe I am basing this on anecdotal evidence, but I didnt see other users chiming in with success stories when he was tinkering with his. I have been told the newer Navien models are much better in all respects including the outdoor resets. I really don't want to invest too much more into this particular boiler, It's a replacement for a previous Navien that was a lemon, and there is a part of me that doesn't really trust this one.

I do reduce the set temp spring and fall, but I like to set and forget for a bit in the winter and not have to remember to increase when going into single digit temps outside. I have also noticed more cycling when running lower set temps (probably due to longer run times, and IIRC hotter water cools quicker than cooler water so you lose delta quicker at lower temps). But then again the wear and tear additional cycling may be tolerable with the less wear and tear and additional efficiency of running condensing temps.

 

[Dana]

The min-fire input on the CH210ASME is 20,000 BTU/hr, ao at 95% efficiency it would be delivering 19,000 BTU/hr, which is approximately your 99% design heat load (70F inside, 10F outside, 60F difference). That means that even without low temperature water it is guaranteed to be cycling rather than modulating with the heat load.

With 54' of SlantFin that would be ~350BTU/hr per foot of baseboard, which would balance perfectly at an average water temperature (AWT) of about 145F without forcing it into a cycling mode. With an out to in temperature difference (delta-T) of 20F the entering water temperature (EWT) would be about 135F, which is above the condensing zone, but not far above the condensing zone. So it's probably running about 86-87% efficiency at the 99% outside design temp if you hold the line at 150-155F.

At an AWT of 130F it would be in the condensing zone, and the baseboard would be delivering 14-15,000 BTU/hr, which is MOST of the output, with ~5000 BTU/hr of "excess" forcing the temperature rise & cycling. The question is only how MUCH cycling.

Without the wired remote, this boiler can be set up to control either the supply temperature OR the return temperature, using the DIP switches on the control board. See p51 of the manual. The remote would overide any of the DIP switch settings, but it's possible, or even likely that it is currently set up for return temperature, rather than output temperature control.

If the boiler was installed with Navien's hydraulic separator manifold or similar, the delta-T at the boiler will be smaller possibly as low as 5F. On a milder day, set the boiler to 125F (in or out) and observe the behavior- time the burns, and count the burns per hour. The temperature overshoot before the it turns the flame off, and the temperature undershoot at which it re-fires are programmable on some boilers, but I couldn't find indications that those parameters are adjustable on this one. Do you have the remote, and the manual for the remote? But let's assume it's 10F between flame-out & re-fire.

With 54' of 3/4" tube and you have about 8 lbs of water in the fin-tube and 27lbs of water in the PEX, or 35lbs total. With 6000 BTU/hr (100 BTU/minute) of excess heat going into the system the temperature is slewing about 3F per minute, and for a 10F swing around the set point you'd get a ~3 minute burn. At 12,000 BTU/hr out (what the 54' of fin-tube emits at an AWT of 125F) it would take about 6 minutes to re-fire, for 9-10 minute cycles, and 6-7 cycles per hour, which would be fine. you'd be in the low 90s for combustion efficiency for most of the heating season, probably 90% AFUE, a 4-5% savings.

With outdoor reset you don't get a lot of programming options with this boiler, but set up to control the return temp using the K-factor curve somewhere between 2.5 and 3.5 would probably do the right thing given your estimated heat load, radiation, and water volume.

 

[MikeG88]

Hi Dana,

The primary loop is copper, I don't have the Navien manufactured manifold.
I do have the remote (its's mounted next to the boiler), and the manual (but it doesn't show any adjustment for difference or delay before flame-out and re-fire). It is set up for supply control but have tried return. When in return (I tried 130 as the setting) the unit would fire up and seemed to increase supply temperature in steps (not sure if they increase based on time or return temperature). The supply would just about be hitting 180 to maybe 185-ish as the return hit 145 (130 setting plus 15 over). (This took a few minutes and the energy saving icon was on most of the time).The flame would shut off and the internal pump and external circulator would continue to circulate the water through out the zone. Usually on a mild day (above freezing) the t-stat would be satisfied before before the water dropped back down to 130 to restart another burn. Often the return would be around 142 degrees when the call ended. Of course as the outdoor temperature dropped the return temperature at the end of call would be lower or even restart the flame for a second burn. The call would often end after the second burn just started but at even colder temperatures the second burn would complete (IIRC in the lower 20's on down).

One plumber told me he suspected this is actually more efficient because by circulating the heated water after the flame is out- more of the BTUs are being squeezed out and not left in the pex buried insulation in the floor joists? I wonder what kind of system would do good in return control, assuming you are trying for a long burn-something with a large delta-T? Wouldn't a room at the end of that zone, with such a large delta, be noticeably cooler that a room at the beginning (unless the amount of emitters was increased to compensate)?

Thanks for the insight!

 

[Dana]

Controlling the firing rate based on the return water temperature is about maximizing combustion efficiency. That can be effective with a low delta-T or a large delta, but the idea is to keep the temperatures in the condensing zone. The raw combustion efficiency is largely a function of the entering water (return) temperature (EWT). With a 40F delta-T you can be delivering 150F water out and still edge into the mid-90s for combustion efficiency, since the EWT at the boiler is 110F. A 5F delta (115F out, 110F return) would only be slightly more efficient.

With a large delta-T on the radiation, yes, there can be room-to-room temperature balance issues. The lousier the insulation & air tightness of the house, the more pronounced that will be. If you're a PassivHaus levels you can heat the place comfortably with point source located about anywhere in the house, but if you're in an uninsulated house the room to room differences will be obvious with a 40F delta-T on a fin-tube loop.

But it's almost impossible to achieve that sort of delta-T and still meet the minimum flow requirements of a modulating condensing boiler, even if pumping direct rather than primary/secondary.

The total amount of heat abandoned in the boiler & plumbing at the end of a burn is pretty small, and still inside the thermal envelope of the house. A minute or so of post-purge flow to even out the temperatures a bit can be "worth it", but not 10 minutes.

 

[MikeG88]

Dana,
I am wondering what you thoughts are as far as running this unit in return? This unit was a straight replacement for another of the same model that was experiencing the error code 12 problem. My builder replaced it (on their dime) but we did try running it in return mode for a month or more before the swap (and never witnessed the error 12 problem during that time). If my memory is right we did have some quite cold days during that time, maybe even single digit or lower.

Setting the return temperature to 140 degrees (the maximum return temp setting) with 54LF of baseboard emitting 300 BTU/HR per foot at that temp = 16,200 BTU/HR. I understand that would be the minimum available as the boiler fires until the return temp reaches 15 degrees over the setting, then continues to circulate the water until it either drops back to the set temperature and refires or the heat call ends. Actually for a while the temperature is in excess of the set point plus 15, then drops to around the set point plus 15 for a while, then continues to drop. Not sure if there is any way to get a good estimate of the extra BTUs supplied by this, except maybe trying plot times and temperatures?

Based on the blueprints, the heat requirement at 60F differential is about 19k, but I am hoping that is a bit on the conservative/safe side. The rule of thub I have been quoted for this area and recent code construction is 15BTU/HR at 0F (1152SF X 15BTU = 17,280) or 13 BTU/HR at 5F (1152SF X 13BTU = 14,976). Based on this estimation the 16,200 (based on 140F return set point without even considering the effect of the 15F allowed over set point), I should have enough available to go 5F and with the extra probably to 0 or lower?

It's obvious that the burner would be firing more times than if it was in supply mode. Would that be a concern, as far as wear and tear, as well as overall efficiency? It seemed that when I did run the unit in return, the energy saving icon was lit for longer, and even though the supplied temperature did climb- it was more gradual -not a hard rapid fire climb. I would think this might save wear and tear and may also be more efficient. I understand the Taco circulator is rated for continuous duty and would assume the internal pump is also similar.

Also- the unit is right under my bedroom and return was quieter without the rapid ramp up!

Thanks for any thoughts...

 

[Dana]

Running it in return mode is probably going to be more efficient, and the right way to go.

A return temp of 140F is going to be at-best 87% efficiency. If you can knock 10015F off to run it at 125-130F return without causing it to short cycle you'll gain a few percent, but may have to bump it back up for the absolute coldest weather.

Count the cycles per hour, and time the burns. Some quick rules of thumb:

A: Five burns or fewer per hour is great, but not more than ten.

B: Three minute burns or longer is totally fine (as long is it's still under 10 burns/hr) but not 100 seconds.

So, 2 minute burns, 2 minutes off is a 50% duty cycle, but it's 15 burns per hour and true short cycling, 3 minutes burns, 3 minutes off would be 10 burns an hour, but probably more cycling than you really want. But 5 on / 5 off it's the same duty cycle, same amount of heat per hour, but only 6 burns per hour, which is better. The lower you can drop the return temp without violating the rules of thumb the higher the combustion efficiency.

If you have an infrared thermometer you can measure the delta-T on the secondary loop by painting or taping the plumbing a foot or so from the hydraulic separator. If it's really tight, say <5 F when the boiler temp is reading 140F the radiation is being a bit over-pumped. If it's in the 10F range @ 140F it's fine (say, 140F out, 130F return). It'll be narrower when the boiler is at 120F. The delta-T on primary loop could be quite a bit higher, even at a narrow delta-T on the radiation.

 

[MikeG88]

Dana thanks for the reply and encouragement to use return, I will have to wait and see if we get any truly cold days during waking non-working times to try to time the burns and count the cycles. If I recall the burns themselves are a bit short in return and the the water continues to circulate until the thermostat is satisfied or maybe a second burn starts during the same cycle, that burn is probably shorter as the water would already be starting at the set point. Not sure on the original Navien unit if I was needing more that 2 burns per cycle on the real cold days (will have to try it) didn't really care because it was going to be replaced.

I call me propane supplier because I'd read that their software to predicts when I need a delivery may be able to provide a K-factor, but all they were able to give me was a degree day factor. They said I usually average between .08 and .11 (although I have peaked much higher once or twice). I tried a quick internet search but could not find a calculator that would allow me to convert this figure to a k-factor? Not sure if thes figure is any good for the purposes here. Also thinking about it - not sure if the allowance the give me for the DHW portion is all to accurate, not sure if they are using assumptions or are able to figure on the tank that seems to last me from spring to fall. (Live alone, minimal cooking, dont usually wait for the water to get hot to wash hands if they are not that bad, etc).

I'm assuming I should continue to run the Taco 00R-MSF1-4IFC on the slow speed in return. With only approximately 175' round trip through the 3/4" pex and baseboard, I bet it's still zipping pretty fast?

Thanks!

 

[Dana]

A K-factor is just HDD per gallon. I'm not sure what "degree-day factor", is.

I suspect it's gallons per degree-day (=1/K). For a fill-up with 0.08 gallons/HDD (probably a heavier space heating load period), at 90% efficiency that's 91,600 BTU/gallon x 0.08 gallons/HDD= 7,328 BTU/HDD, which is (/24=) 305 BTU/degree-hour.

So when it's 10F outside (55F below the presumptive base 65 for the HDD), the load is about 305 x 55= 16,775 BTU/hr.

At 0F (which is below your 99% outside design temp) your load is about 305 x 65F= 19,825 BTU/hr

If the use is actually 0.11, not 0.8, the load @ +10F is about 0.11/0.08 x 16,775= 23,066 BTU/hr, but I would expect that number would be related to a low space heating period fill up, with a large error induced from the domestic hot water use.

If that's what's meant by a degree-day factor, it puts a nice bracket around it, but unless you were on the beach in Belize for a month with the thermostat turned down during that period, trust the 305 BTU/degree-hour number.

Measure the delta-T on the baseboards at a point a foot or two before the return water gets to the hydraulic separator, and a foot or two from the outbound side of the separator. This is most easily done with an infra red thermometer, but if it's copper instead of PEX you'll have to put some hockey stick tape or paint on the plumbing where you're taking the measurements. If it's more than a 25F difference you might want to bump up the pump speed, but if its under 15F don't.

 

[MikeG88]

Dana, thanks for the degree day factor speculation, it seems to make sense so you are probably right on.

A few years ago I did borrow an infrared thermometer and placed tape in several spots. I dont think I ever got a delta over 15F once the boiler was running.

Last night I decided to put the unit into return and the set point was 130F. It was about 40F most of the night but had been at least a few degrees warmer during the day. The cycle would start, and flame for maybe 3 minutes-supply would get up to 180 or so just as the return was hitting the shut off point of 130F plus 15. The burner went off and the internal pump and the taco continued to circulate the water. Watching the return temp it started to waver aroung maybe the higher 150s for a while than start to drop over couple minutes time, when it got to around 145 it seemed to spend some time there and actually went up for a short while. The t-stat call usually ended aroung 141 to 143F (never getting any where near 130F to reignite during the same t-stat call/cycle). At that point I wasn't really watching how often there was a heat call- I was more concerned that a single ignition handled the call. Hours later I woke up and it was still about the same outdoor temperature but I checked on the boiler and noticed the the temperature was dropping to around 133F (just 3F short of reigniting). I bumped the return upto 140F and it actually did drop low enough for reignition before the call cycle ended. The t-stat cycle starts were maybe 25-30 minutes apart at this time. I know I got away with the original unit in return during some pretty cold days (single digit), but I wasn't really concerned with how may reignitions there were. I've run this unit before in return at 130F in the spring/fall when outdoor temps were consistently above freezing and never noticed a reignition being necessary. Even short term drops into the upper 20s it seemed to keep up. I'm using a simple Honeywell "the round" thermostat set for 3cph, I understand the setting is a max and last night it wasn't using the max. Would a digital thermostat with more cph options, maybe set for 5cph, have given me more cycles over the same time, or when it's that warm they cycles might still be limited? I just trying to avoid reignitions- or should that not be a concern?

Importantly to put everything into perspective: Normally running in supply temp control my short cycle issues usually occur shortly after the unit reaches set point, the unit hoovers around 2 or 3 degrees above or below set point and then I soon hear the hum that I think is it modulating. On occasion it doesn't seem to be able to modulate down fast enough and it shuts the burner down for a short and then reignites and the temp comes back up but usually stops about 7-8 degrees below set point. This happens at random times, but more often after a DHW cycle, or on a mild day (it's happened a few times tonight). Should I really be worrying about this? I'm not being super picky about getting more efficiency out of the unit, my real concern is really just not putting a lot of undue wear on the unit. I'm thinking that with the unit otherwise running fine at 3cph, even if it short cycled every time (which it doesn't), that is still only 6 ignitions per hour (and is it really a short cycle because it happens about 3 minutes into the burn).

 

[Dana]

Six cycles per hour isn't boiler-abuse, and three is half that much. Whether cycle-limiting features of the thermostat are necessary to keep the counts that low isn't totally clear, but as long as it can still heat the house during colder periods with them enabled it doesn't really matter.

 

[MikeG88]

Thanks for the sanity check Dana, if the worst case scenario amount of short cycling I may get on a mild day isn't abusing the boiler I won't worry about it so much.

I did see pick up some literature on the Lochinvar Nobles at my propane suppliers showroom (too bad they do have an operational display). I liked it (the NKC110) for it's minimum output, basically half of my Navien CH210. They claim to be very quiet (the Navien is under my bedroom and can become irritating in a real quiet night) I would love to see one in action and witness for myself if really that much quieter.

 

[Dana]

The NKC110's minimum fire output is certainly more attractive from a space heating point of view, but it will be able to support one full-flow shower at a time, but not a whole lot more. That's fine for 1-bathroom houses, but may require some scheduling of laundry/dishwasher use to not conflict with bathing.