oversizing mod/cons YES!!!!

jp_2

with some modifications, this is the better approach for staying in the higher efficiency areas of the graph.  see graph below.    the graph clearly shows in worst case 160F supply temp, oversizing by 30% would be beneficial.  of course if the system can be ran at supply temps below 130F this wouldn't applied.



the error lies in the control design of these boilers and the famous tekmar's.   they are designed with inherent flaws that force you to use smaller boilers that need to be driven harder and less efficient.

Robert O'Brien

What

does the size of the boiler have to do with the temps? Oversize the radiation is more like it!

CMadatMe

Modulation Limitation

 It's use is to be able to limit the heating circuit rate of modulation separate from the domestic load in VItodens. Low heat loss but need to supply domestic demand on a single boiler system.

NH03865

Why??

Why would you ever want to operate with a 160F water temperature.  That is using a mod/con in a non-condensing mode?

If you want to use 160F water use a simple CI boiler.  Much simpler to operate and less maintenance.  If you want to use a mod/con, at least use it in a condensing mode and as your chart shows the lower the water temperature the more efficient the system. 

As Robert has said, increase the emitter to lower the water temps to as low as practical.

CMadatMe

NH, Not Necessarily True

Could design using a 30 degree delta-t. In the case of Vitodens does water temp matter if the limitation of the rate of modulation gets us into 90%? We can limit a heating circuit rate separate from another circuit or domestic.  Even using 160 at full load could still condense 90% of the heating season. Only need the 160 on design day...

Have to step back and look at the total system design.

scott markle_2

good point

l have begun to rethink this as well. I have made a concerted effort in the past to size on the small side and try and match the boiler to the load as best as possible. It's often impossible as even the smallest boilers are often "oversized" for a modestly sized well insulated house.



Looking at charts that show efficiency as related to boiler output it's clear that we can achieve condensing efficiency at higher temperatures when we are at partial load.

Your on to something about leveraging efficiency by keeping modulation in the low range as much as possible. Unfortunately most controls are not so good at managing an oversized modulating boiler. Lets take a basic onboard boiler ODR as an example. Most installers are going to set-up a curve thats higher than it needs to be to keep the structure in its comfort range, Why? for one thing the best heat-loss assessment is just a starting point and in a competitive market how many installers are willing to make multiple visits to a site to tweak a reset curve and establish constant circulation? Secondly if a curve is set aggressively low such that it can only maintain 68-70 what happens if zones are imbalanced or grandma is visiting and needs 75, or for that matter how will people feel about very sluggish recovery from setback.



So take a typical mod con with a higher than needed reset curve and regulate room temp with a bing bang stat. This is probably 4 out 5 of the units that are in service, Now lets suppose a really well insulated new structure with a 30k design load and a 30-90k mod con. lets say it's an average cold day and the thermostat is averaging a 70 percent on time, and it's cycling 6 times an hour, what happens when the boiler first kicks on and its targeting 130. Well it's going to encounter a lot of room temperature water as it begins to warm up the emitters and even though the actual loads on the structure are maybe only half of the boiler minimum modulation and the target temperatures are relatively low it's going to ramp up to a high modulation at first and then ramp down as the emitters come up to temp. My point is that it never needed to come above min modulation and it would have given it's best possible performance if it had never ramped up. This is especial true for a high mass situation, we may think that mass and even a buffer is a good thing but if the boiler is operating based on desired supply target this mass is going to be seen by the boiler as a load and it will unnecessarily go into a higher firing rate than it actual needs to.



My own solution(at home) was to cap the modulation rate (via a potentiometer on the vitodens lgm board) and run my system on an ODR interval timer (tekmar steam control), It works really well, the combination of minimum firing rate and cycling of a low temp high mass emitter (slab) makes for the highest possible efficiency.



It's a bit quirky and leaves something to be desired in terms of the user interface but I believe this sort of approach should be explored in a more commercialized way

scott markle_2

good point

l have begun to rethink this as well. I have made a concerted effort in the past to size on the small side and try and match the boiler to the load as best as possible. It's often impossible as even the smallest boilers are often "oversized" for a modestly sized well insulated house.



Looking at charts that show efficiency as related to boiler output it's clear that we can achieve condensing efficiency at higher temperatures when we are at partial load.

Your on to something about leveraging efficiency by keeping modulation in the low range as much as possible. Unfortunately most controls are not so good at managing an oversized modulating boiler. Lets take a basic onboard boiler ODR as an example. Most installers are going to set-up a curve thats higher than it needs to be to keep the structure in its comfort range, Why? for one thing the best heat-loss assessment is just a starting point and in a competitive market how many installers are willing to make multiple visits to a site to tweak a reset curve and establish constant circulation? Secondly if a curve is set aggressively low such that it can only maintain 68-70 what happens if zones are imbalanced or grandma is visiting and needs 75, or for that matter how will people feel about very sluggish recovery from setback.



So take a typical mod con with a higher than needed reset curve and regulate room temp with a bing bang stat. This is probably 4 out 5 of the units that are in service, Now lets suppose a really well insulated new structure with a 30k design load and a 30-90k mod con. lets say it's an average cold day and the thermostat is averaging a 70 percent on time, and it's cycling 6 times an hour, what happens when the boiler first kicks on and its targeting 130. Well it's going to encounter a lot of room temperature water as it begins to warm up the emitters and even though the actual loads on the structure are maybe only half of the boiler minimum modulation and the target temperatures are relatively low it's going to ramp up to a high modulation at first and then ramp down as the emitters come up to temp. My point is that it never needed to come above min modulation and it would have given it's best possible performance if it had never ramped up. This is especial true for a high mass situation, we may think that mass and even a buffer is a good thing but if the boiler is operating based on desired supply target this mass is going to be seen by the boiler as a load and it will unnecessarily go into a higher firing rate than it actual needs to.



My own solution(at home) was to cap the modulation rate (via a potentiometer on the vitodens lgm board) and run my system on an ODR interval timer (tekmar steam control), It works really well, the combination of minimum firing rate and cycling of a low temp high mass emitter (slab) makes for the highest possible efficiency.



It's a bit quirky and leaves something to be desired in terms of the user interface but I believe this sort of approach should be explored in a more commercialized way

jp_2

boiler size- water temp

boiler size,  water temp no connection.



as per the graph, boiler load is inversely proportional to efficiency.   harder you drive the boiler, the lower the gas mileage. 

Jason_13

Huh!!!

I can't agree with this. Sounds like you are saying that a 30k heat loss home and a boiler rated at 30k to 90k. You must be assumming that it is a one zone home. If there are two zones and in theory they are each 15k what happens then?

It would seem that an 80k which is also oversized but will modulate to say 16k which will closely match minimum firing rate.

What am I missing with your theory.

Jean-David Beyer

MY theory.

My house is two zones, not one. I calculated the heat loss two different ways. The following numbers are probably the most accurate; I calculated them with the Slant/Fin program. The numbers may be very slightly low because I suspect there is a little more air infiltration that assumed in the calculator.



Downstairs zone is 22,761 BTU/hour when it is OF outside and I am trying for 70F inside. Radiant slab at grade.

Upstairs zone is 6,473 BTU/hour under the same conditions. Oversize Slant/Fin baseboard heat, oversized so I need not run more than 135F water through them.

My boiler is rated at 16K to 80K  CSA input;  14.2K to 71K CSA Output; 12.4K to 62K I=B=R output.



So you are missing that the two zones have very different heat loads. The maximum heat demand for the upstairs zone is way below the minimum output even when it is 0F outside, and in fact, I have never seen it that cold here. Design day is 14F. I have seen it get below 10F for an hour or so at night about once or twice a heating season. It is safe to say that the upstairs zone never can take the minimum boiler output. So it must operate in bang-bang mode when only the upstairs zone is calling for heat. The downstairs zone can take the minimum output by modulating all the way down, and if both zones call for heat, that is OK too. The boiler cannot modulate down to 6,500 BTU/hour, and if it is warmer than 0F outside, it has to modulate down even further than that. If W-M made an Ultra 3 40,000 BTU/hour boiler, I should have gotten that one. And if it would modulate 10: instead of 5:1, that is what I should have.

NRT_Rob

you are forgetting

design day is .04% of the year.



If you oversize, you raise the min mod and you are in cycling mode, which is less efficient, far more often than that. Even if the boiler is perfectly sized, with a 5 to 1 turndown that's 20% of your heat load range you're in "cycling" mode for.



Far better to size properly. Then you are only "pushing it" during the coldest periods, and you minimize cycling. This should also extend the life of your equipment. If you check out some bin data you should find that 30% load condition is something like most of the BTUs you'll need in a winter as well, I think, though I haven't studied that rigorously yet.

Jean-David Beyer

I am not forgetting.

"design day is .04% of the year." True, by definition. For the last 6 months,

I get the following number of degree-days using a base temperature of 14F, the design day temperature. Only 8 days go below this for long enough that their observations can detect it. I think their data are once an hour.



Description: Fahrenheit-based heating degree days for a base temperature of 14F



Station: Airport: Belmar-Farmingdale, NJ, US (74.12W,40.19N)



Month starting   HDD



10/01/10            0

11/01/10            0

12/01/10            0

01/01/11            8

02/01/11            0

03/01/11            0

NRT_Rob

sorry, not clear

I was responding to original poster, not you.

CMadatMe

Pick the Poision

Lets face it. At some point all boilers are over sized and short cycle. You just have to find the less intrusive poison to limit it as much as possible.



We also have limited product from our European friends. Here is the product avail from ACV who in our world is Triangle

http://www.acv.com/int-en/03_10/17/app.rvb?prodfam=19



Could you cut down on short cycle with one of these...JB's heat loss would dictate the little 2.2kw - 18kw.. That low end is 7,500 btu's..



Another example is Viessmann. This is the list of the mod/cons available across the pond. The boilers we have here low end start at 9KW not the 3 and 4kw's shown here.

http://www.viessmann.com/com/en/products/Gas-fired_condensing_boilers.html



Is seems the US gets the leftovers when these companies develop new product for the European market. What we get and consider "new" is truly "old"....

Mark Eatherton

How LOW can YOU go....

Looks like an 8:1 turn down ratio to me. If the agencies that approve mod con boilers on this side of the pond would recognize the fact that these beauties CAN be turned down to lees than a 5:1 ratio, it would go a LONG way towards eliminating short cycling of boilers.



Seems there needs to be some pressure applied to those agencies from the people who are building these devices. The new Crest has a 25:1 turn down ratio, but the base BTU's are WAY to big for most residential applications, unless you want one appliance to do space heat, swimming pool, hot tub, DHW and snowmelt at the same time.



Priority? We don't need no stinkin' priority!



ME

pipe4zen

Burner output limits

I do like the ability to limit the burner output feature as found on the Buderus GB142. A properly sized unit, then reduce output by as much as 50% helps alot with short cycling.

Jean-David Beyer

ability to limit the burner output

"A properly sized unit, then reduce output by as much as 50% helps alot with short cycling."



I do this with one zone from my W-M Ultra 3. Luckily, that is easy with their controller.

But with a modulating boiler, it should not be necessary to do this, at least with my boiler. Having designed feedback control systems in a past life, I could tell what the short cycling was caused by. At least one of the causes. The trouble is that the controlelr starts the boiler at 50%  firing rate. It then notices that the hot water output is too cool and increases the firing rate too rapidly. By the time the controller notices that the supply temperature is almost at the upper limit, it too slowly reduces the firing rate, so the boiler shuts off at the upper limit. Then the water starts coolinig off as the thermostat is still calling for heat. By the time the boiler approaches the lower limit temperature, the controller cannot refire the boiler; it must wait until it hits the lower limit: classic bang-bang operation. Basically, the lag in the control system is a part of the problem, but changing the proportions of the proportional, integral, and differential components are three parameters not accessable to the homeowner (or the technician either, for that matter). The proportions are correct for the loads for which the boiler is designed, but not for the very light load I have on it for when the small zone is the only one asking for heat.



If I were W-M, I would not make those three constants available because if the person tinkering with them does not understand feedback control systems, they will get the system unstable and get extremely unsatisfactory results. But if they made a smaller boiler, or put in 10:1 or 20:1 turndown, I believe that would solve a lot of problemss. I do not know enough about gas burners to know why they cannot fire at lower rates than 20%. Is it flame instability or what?

CMadatMe

Boiler Industry

Are the powers to be, ECR, Burnham, Weil, Peerless? What we could call     "The Big 3" amercian companies. May the problem be that they cannot continually afford to compete with the Europeans due to the mirco dot market share they have when looking at it as global competition?



Is the regulation held back to allow them to compete in the marketplace? Wouldn't be the first time we put jobs in front of better products. The automotive industry learned that the hard way.

Gordan

I suspect the graph is misleading

Devil is in the details. Boilers in the same family that have different outputs often share the same heat exchanger, up to a point. I fail to see how the same heat exchanger would treat 20kBTU/hr produced by a 60kBTU/hr burner at 30% output, any differently than it would treat 20kBTU/hr produced by a 100kBTU/hr burner at 20% output.



I also agree with others who point their finger at what happens at lower heat loads as a bigger problem than what happens at higher heat loads.

jp_2

good point nrt

my main idea was looking at the period of greatest usage and sizing for highest efficiency.   around here, that would be dec,jan,feb. 



did you read my post on short cycling?  that would clear up inefficiencies in that department.  ODR falls apart under minimum fire.



by the way, I think your math is a little rusty, .04% of the year is a little under 4 hours?

NRT_Rob

sorry

0.4%, moved the decimal point.



ODR doesn't have to fall apart at minimum fire. that's a separate issue from short cycling. What you're concerned with there is your available mass and your firing differential.



I haven't done the BTU analysis yet but I suspect very strongly that design day type usage is not the primary BTU consumption of the year.

scott markle_2

Firing rate Vs. Supply target

Cycling is bad, my point (in agreement with JP) is that it's as much a failure of designing intelligent controls than of over sizing boilers. The graphs provided by the O.P show the thermal efficiency advantages of a large HX and low firing rate,while it may be wasteful from a financial standpoint to oversize a condensing boiler, It could (if well managed) actually produce higher efficiencies. I would argue from my observations that even appropriately sized mod-cons using ODR are frequently operating at firing rates that are determined by supply targets and system inertia as opposed to actual BTU loads.



In a perfect system why should the boiler firing rate exceed the actual BTU demands on the structure? How much of each firing cycle (excluding a flame stabilization period) for a typical mod-con is spent at a firing rate that exceeds the actual loads on the system? My guess is quite a lot. I'v noticed that the new Viessman ODR seems a lot smarter (by comparison to the original vito) in regards to boiler cycling, it seems to get into a low modulation rate quickly and permit a fairly wide overshoot and long burn even in low load situations.



I can imagine a multi zone controller that allows the installer to input the actual design loads required by the individual zones (as well as the boiler modulation parameters). The control could look at the various zones calling, their deviations from room target and based on the outside temperature calculate a firing rate as opposed to or in conjunction with supply target. Actually this seems a bit overly complicated, just thought I take a stab at a solution as opposed to just complaining about limitations. I think Chris's solution (Setting a max. modulation on boilers that permit this) is the place to start.

scott markle_2

Firing rate Vs. Supply target

Cycling is bad, my point (in agreement with JP) is that it's as much a failure of designing intelligent controls than of over sizing boilers. The graphs provided by the O.P show the thermal efficiency advantages of a large HX and low firing rate,while it may be wasteful from a financial standpoint to oversize a condensing boiler, It could (if well managed) actually produce higher efficiencies. I would argue from my observations that even appropriately sized mod-cons using ODR are frequently operating at firing rates that are determined by supply targets and system inertia as opposed to actual BTU loads.



In a perfect system why should the boiler firing rate exceed the actual BTU demands on the structure? How much of each firing cycle (excluding a flame stabilization period) for a typical mod-con is spent at a firing rate that exceeds the actual loads on the system? My guess is quite a lot. I'v noticed that the new Viessman ODR seems a lot smarter (by comparison to the original vito) in regards to boiler cycling, it seems to get into a low modulation rate quickly and permit a fairly wide overshoot and long burn even in low load situations.



I can imagine a multi zone controller that allows the installer to input the actual design loads required by the individual zones (as well as the boiler modulation parameters). The control could look at the various zones calling, their deviations from room target and based on the outside temperature calculate a firing rate as opposed to or in conjunction with supply target. Actually this seems a bit overly complicated, just thought I take a stab at a solution as opposed to just complaining about limitations. I think Chris's solution (Setting a max. modulation on boilers that permit this) is the place to start.

jp_2

yes & no NRT

you are right, the boiler can modulate down to ambient temp.   but the ODR is forcing it to modulate down without giving it room for longer burn cycles with time in between.



this means allowing a larger dT.  remember I'm talking about the shoulder season,  my guess is wider dT at this time of year would go unnoticed.



 the time you want to worry most about efficiency is when you are burning the most fuel per day, around here that would be dec,jan,feb.   in the shoulder season, according to viesmann, you'll be at the top of efficiency away.  



but this is all talk, you only have a couple different size boiler out there to choose  from.



STILL, the ODR should be modified to go off of constant circ  firing mode when under say, 30-40% min modulation.  you could do this in the ODR or the boiler control.  so then it starts acting more like FHA or older style boiler w/tstat.





the BIG saving comes from lower boiler supply temps, the saving don't really come from  running constant circ. though its part of it.  so you widen the dT and you save more fuel.



ADDED:  this now ties the 2 posts together, short cycling and oversizing.

NRT_Rob

again

that's not really related to ODR, as you kind of note, it's about the boiler firing differential.



the action you describe is present in Tekmar controls with "automatic" boiler differential settings. it widens during low load conditions and narrows during higher load conditions. it's a great feature but in modern microzoned hydronic systems it's probably not adequate to avoid short cycling by itself, though it helps.

jp_2

Hmmm

I thought the idea behind ODR was constant circulation and constant firing? so if the boiler can't modulate down any further in mild weather it starts short cycling?



as I said in another post, micro zoning is a bad practice, unless you pull heat from other zones instead of firing a boiler when the zone is only 10% of minimum fire; or you use a buffer tank.



can tekmar also monitor indoor temps a long with outdoor?  can they let the homeowner make small adjustments to room temp,  I understand most  ODR can't do this?

NRT_Rob

outdoor reset

will lengthen firing demands and extend circulation time, but is not by itself a constant circulation strategy. even if it is, that is not related to boiler firing differentials which is what determines, along with mass, what your firing time is for a given output.



Micro zoning is the norm if you want comfort in most homes as many of them have at least one area that really wants to be its own zone. but it does require paying attention to your cycling control.



TekmarNet4 is an indoor feedback system. ODR is the base, refined by indoor temperature feedback. very powerful.

scott markle_2

indoor outdoor

Yes, the tn4 and tn2 stuff is uses communicating thermostats and combines indoor and outdoor feedback to determine supply temperature.

It also attempts to sync. the zones and helps with the micro zoning problem you mentioned.



One of the best part is that the stats have authority over the reset curve, so when the user turns down the stats they are in effect making a parallel shift of the curve. Same thing if they want to crank the heat. I think the new and less expensive two wire tn2 house controls work even better than the older tn4 stuff. It's really a great product. I'v never done a honywell aq2000 but it's very similar.

Nick Ciasullo

Oversizing Bad, mkay?

If you oversize the mod/con, it will have short cycle issues as previously mentioned. Pre-purge and post purge will eliminate most if not all savings by oversizing, imo.

jp_2

I am confused NRT?

I thought  if you hooked up an outdoor sensor to a modcon ,  the circ would run anytime you were below a certain outdoor temp?



  I'm not sure what you mean by "'firing differentials"?



what do you call a short cycle heating wise?



if your ODR calls the boiler to run and the outdoor start up is 60F and its 59F outside how will the boiler act?



I don;t agree boilers need to modulate down to 5 btu's to solve short cycling heating problems.  that becomes inefficient in itself.

NRT_Rob

no

you apparently aren't clear on ODR operation. ODR calculates a water temperature requirement based on the outdoor temperature.



when a demand is placed on the boiler, typically with a normal thermostat, it will run the pump and sample the water to see what the water temperature is. If you're below the temp requirement, it will fire the boiler on a firing differential... i.e. if you need 90 degree water, it might fire at 85 and turn off at 95, for a 10 degree firing differential.



wider differential = longer firing cycle. also a higher minimum temp as you can't cross room temp on the bottom end.

Jean-David Beyer

the circ would run anytime you were below a certain outdoor temp?

Mine does not do that.



My boiler circulator and one of the zone circulators (sometimes both of them) run only if a zone thermostat calls for heat. The ODR sets the supply water temperature; the temperature depends on the outdoor temperature and the size of the range around the reset curve. Right now it is 79F outside, so the Warm Weather Shutdown will prevent anything from happening. But if it were not for that, my reset curve would ask for 75F water, and the range would allow anything between 70F and 80F. If it were 4F outside, the ODR would ask for 120F +|- 5F hot water. This is not the only way to run ODR, but it is how my boiler does it.



The only way I get constant circulation is if I can get the reset curve to exactly match the heat load of the building, opr be less than the heat load.. The best I can get is for it to run about 18 hours a day. I see no reason to run my circulator(s) if no heat is called for.



Actually, the boiler control will run sometimes without the thermostats asking for heat. The controller has a freeze-protection feature, so if the water in the boiler drops below 40F, it fires the boiler at low heat and runs the circulators until it gets up to 45F. This is less foolproof than it sounds. It works if someone turns the thermostat off, or if the batteries go dead. But if the power or the gas to the house shut off, this cannot, obviouly, do anything.

jp_2

standard practice

NRT i didn't know it was standard practice to use an indoor tstat with ODR?



if you increase the Tstat from 68F to 74F how does the boiler know to increase the supply temp?



as your post suggests, short cycling should not exist then.

NRT_Rob

it doesn't

ODR only knows that there is a demand, and what the outdoor temp is. that's a limitation of the technology.



Indoor temperature feedback is what you're talking about. and if you're using THAT, then you have automatic boiler firing differentials and the like as well.

Jean-David Beyer

I feel sorry for whoever gets my house after I die.

I have outdoor reset on my mod-con and it is set up to run for long intervals (up to about 18 hours) to maintain the 69F I like to keep the house at. It supplies so close to the loss that the thermostat does not get satisfied enough to stop calling for heat even though the thermometer on the thermostat says I am at the set point (but I am not quite). If it is very windy, the house leaks enough that I have the water set very slightly too high in the coldest days.



But if whoever gets the house after I die and sets the thermostat(s) to 72F, or something, it may be that they will not get it. And they will think the boiler is broken or undersized, when in fact it is somewhat oversized.



I started writing a user's manual for that, but I got bogged down at explaining how to set the reset curves for the system. I could suggest they read the approprate chapters in John Siegenthaler's book, but I doubt that they (I know who gets the house in my will) would ever read it. And that is not enough to actually do it unless you already understand the idea pretty well already.



It is easy to say, get a qualified heating contractor to do it, but that is problematic. I doubt they would want to pay the contractor for half a dozen visits to get it just right. I know I would not, but I have the luxury of understanding how it all works and  the time to watch the whole thing. Changing a set point takes only a minute or so, but I found it necessary to set one end of the curve on a warm day, and then wait for a cold day to set the other end, and go through quite a few iterations. It would probably take only 2 or 3 days if I could get the extreme temperatures when I wanted them. And after each change, it takes about 24 hours for it to stabilize and see if it is set where it needs to be. So these thing cannot be rushed. I wish my heat loss calculation was exact, but that is too much to expect.