Finding A Contractor To Install A Buffer Tank

JustinS

Hi

I've been trying to find a contractor to help fix my TT Prestige Solo 110 setup that has some serious short-cycling for about 6 months now

The system is a Triangle Tube Prestige Solo 110 that heats 3 zones of low-mass baseboard (Haydon Heat-Base 750, I think) and a TT SMART80 DHW tank off its dedicated DHW port.

My 3 zones are as follows:

1st floor - ~39 ft plus kickspace heater => ~26600 BTU/hr
Master BD - ~25 ft for ~45% of upstairs => ~14500 BTU/hr
2nd floor balance - ~25ft for ~55% of upstairs => ~14500 BTU/hr

I have discussed this a fair amount on this forum and the solution appears to be installing a buffer tank

I tried using the 'Find A Contractor' search engine on this site but there are no contractors nearby; I contacted a few but I am either out of their service range completely OR sufficiently far enough that they haven't been able to visit.

Are there any options beyond this search engine? I live in Bellingham, MA...

Thank you

Rich_49

Could you possibly give us those heat load numbers at 40* ODT ?

JustinS

Rich said:

Could you possibly give us those heat load numbers at 40* ODT ?

By 40* ODT, I assume that you mean 40* Outside Design Temp?

If so, those numbers aren't the heat loads on the zones but rather what each zone can dissipate at 180F.

The only heat load/loss I have on the house is approximately 40K to 45K BTU/hr at 9 DegF ODT

Does that help at all?

Rich_49

May I ask why the high supply water temps / The reason I ask is that besides your short cycling issue you are not condensing and quite probably have an efficiency south of 90% .

Could you use whatever program or spreadsheet you used prior and change 9* to 40* and share that number with us ?

Your 55000 heat loss for the combined zones would suggest that you can get the job done with cooler water and take advantage of the condensing equipment you have . This may also make the needed buffer tank a bit smaller .

I think JA Phinney travels as far as Bellingham . They are out of Boston

JustinS

Rich said:

May I ask why the high supply water temps / The reason I ask is that besides your short cycling issue you are not condensing and quite probably have an efficiency south of 90% .

Could you use whatever program or spreadsheet you used prior and change 9* to 40* and share that number with us ?

Your 55000 heat loss for the combined zones would suggest that you can get the job done with cooler water and take advantage of the condensing equipment you have . This may also make the needed buffer tank a bit smaller .

I think JA Phinney travels as far as Bellingham . They are out of Boston

Hi Rich

I'm not actually using those high temperatures - I just included those numbers as a baseline to show how oversized the boiler is - even when using 180F, I have problems

My ODR curve tops out around 165F or so - I haven't tried to run as low as possible (which I suspect is quite low since last winter, it was -10F and I was fine at 165F or so) because it seems to me that running at cooler temps would only exacerbate my short-cycling as it would increase the delta between input and output

My only resource is the heat loss report an HVAC company provided - they indicated 46100 BTU/hr with area 2137 sq ft and inside/outside temp of 72/9 DegF => 731 BTU/hr per DegF

Scaling for a inside/outside temp of 72/40 DegF, I think that the heat loss would be 23416 BTU/hr

I've run some numbers and I believe that a 40-gallon BT would be sufficient to address my problems (see attached image for calcs)

I've spoken to JA Phinney but unfortunately, he's been really busy and hasn't been able to come take a look... since it's getting cold, I figured I'd see if there were other options...

Thanks

Rich_49

What type of circs does this system have installed on it ? Why are you attempting to push 4 GPM through the system when 1 would most likely be good enough . When you return colder water to the boiler that means more was delivered to the reservoir . If the water returning is colder your short cycle problem becomes much less of a problem . Let's use that as a starting point .

When was the last time you attempted Phinney ? This is a funny business and you can be busy as hell one minute and sitting home the next

JustinS

Rich said:

What type of circs does this system have installed on it ? Why are you attempting to push 4 GPM through the system when 1 would most likely be good enough . When you return colder water to the boiler that means more was delivered to the reservoir . If the water returning is colder your short cycle problem becomes much less of a problem . Let's use that as a starting point .

When was the last time you attempted Phinney ? This is a funny business and you can be busy as hell one minute and sitting home the next

There are 3 Taco 0015 Multi-Speed, set to LO

I'm not sure exactly what my flows are; I attempted to estimate them, as shown in this other thread... I am giving some thought to installing some VR1816s so I can turn down the flow

I emailed James on Monday and yesterday, haven't heard back; last reply I had from him was last week, asking me to wait until this week...

Rich_49

You don't need the VR1816s , get the VT2218 instead if you want to pump this system right .

JustinS

Rich said:

You don't need the VR1816s , get the VT2218 instead if you want to pump this system right .

Could you elaborate? Are you recommending replacing each 0015 with a VT2218? I liked the VR1816 because I can adjust to whatever flow is appropriate, as opposed to being constrained to a set of fixed curves...

Rich_49

The flow that is appropriate is the one that is required at any given time . a Delta T circ does not have a set of fixed curves unless , of course you choose to use it that way . You have 3 zones and intend to use 3 pumps . View the webinar , "Its in the way that you use it "

http://www.taco-hvac.com/products/water_circulation_pumps__circulators/variable_speed_products/viridian_family/index.html

hot_rod

If the load is below 50K as you listed, one circ pump and zone valves would work fine. Probably be able to circulate that with a delta P circ at less than 40W power consumption.

JustinS

hot rod said:

If the load is below 50K as you listed, one circ pump and zone valves would work fine. Probably be able to circulate that with a delta P circ at less than 40W power consumption.

OK, one thing that I'm confused about - it may function better, in terms of cooler temps, to replace my 3 circulators with either different ones or valves... but it still seems to me that my zones aren't really capable of dissipating the boiler's minimum firing rate of 30K BTU/hr, especially if I start running at cooler temps.

Still seems like I would need the buffer tank or am I misunderstanding something?

Thanks

hot_rod

Well all boilers will cycle if the output cannot exactly match the smallest load on a mild condition. That is tough to obtain unless the boiler ramps down to 500 BTU/hr.

So the question becomes how much cycling is too much. Some say a 10 minute run cycle is minimum every time a boiler fires.

Does the control in that boiler allow any adjusting to firing rate. Lochinvar, for one, allows a ramp delay function that all but eliminates short cycling, if the boiler is sized close to the load.

JustinS

hot rod said:

Well all boilers will cycle if the output cannot exactly match the smallest load on a mild condition. That is tough to obtain unless the boiler ramps down to 500 BTU/hr.

So the question becomes how much cycling is too much. Some say a 10 minute run cycle is minimum every time a boiler fires.

Does the control in that boiler allow any adjusting to firing rate. Lochinvar, for one, allows a ramp delay function that all but eliminates short cycling, if the boiler is sized close to the load.

Definitely aware that there will always be some measure of cycling under mild conditions.

My issue is that the minimum firing rate is so much more than my individual zones even on the coldest day and using 180F water... it's pretty much the case right now that my boiler will cycle every 90 to 150 seconds unless all three zones are calling OR if it's really cold out and then I might be able to get away with the 1st floor and just one of the 2nd floor zones.

Unfortunately, the TT doesn't have any adjustment parameter for the firing rate

hot_rod

Well a buffer tank is one excellent way to solve exactly that type of problem, which is somewhat common in zoned or micro-zoned systems.

This journal will take you step by step through the process of sizing and piping a buffer tank.

I hope to try the two pipe method on my own system this fall, it has some nice advantages for various flow rates and heat to the load without first "loading" the buffer tank.

http://www.caleffi.com/sites/default/files/coll_attach_file/idronics_17_na.pdf

JustinS

hot rod said:

Well a buffer tank is one excellent way to solve exactly that type of problem, which is somewhat common in zoned or micro-zoned systems.

This journal will take you step by step through the process of sizing and piping a buffer tank.

I hope to try the two pipe method on my own system this fall, it has some nice advantages for various flow rates and heat to the load without first "loading" the buffer tank.

http://www.caleffi.com/sites/default/files/coll_attach_file/idronics_17_na.pdf

Hi HotRod

Quite the fascinating read, thanks for providing it...

Question about the two-pipe vs four-pipe - is the only real difference between the two the amount of water and the speed at which it enters the tank?

In the example on pg 8, the same 2gpm flows downwards in both configurations but I guess there's a lot more mixing in the 4-pipe due to the higher velocity of the water entering the tank? If I understand correctly, this would reduce the temperature stratification and make the tank temperature more uniform?

Also, could you elaborate more on the advantages with varying flow rates in the 2-pipe?

Thanks

hot_rod

Good questions. To answer them you would need to know the flow rates.
2 piping option: use figure 3-4 and assume you have that 10gpm from the boiler and exactly 10 gpm going to the load. No flow goes into the tank, all the flow from the boiler goes to directly, and quickly to the load.

As load drops off, that % goes through the tank fig 3-4 again, and starts the buffering function.

With a 4 pipe option all the flow must cross or blend through the tank. Even with 10 gpm from the boiler and 10 gpm to the load, it still interacts with the tank as it crosses, and does some temperature mixing.


With zoned systems the flow rates through the tank are always changing, possibly never being exactly the same in as out so blending is occurring.

The second point is about stacking or stratifing the tank. Ideally we want to stir the tank a little as possible to enhance that stacking.

Stacking maximizes and optimizes the amount of energy you can store, try not to create too much turbulence as you enter and leave the tank. Sparge tubes, side port connections help.

Lastly the goal of a buffer is to store thermal energy and limit boiler cycles. Once all the loads drop off 100% of the boiler input recovers the tank. So use the math in Idronics 17 to optimize your design and tank sizing.

So we are asking the tank and piping to provide 3 functions, separation, storage, and mass to eliminate cycling of the boiler.

I have a tall skinny tank in my shop connected to solar and a wood boiler, about 8 feet tall, a former LP tank. I can see a 50 or higher ∆T from top to bottom on a sunny day like today.

All things considered, if the tank has large enough ports to allow you to build the hydro-separator into the piping connections, I think 2 pipe offers better usability.

With a 4 port tank, the extra ports are ideal to add a plate type HX for DHW production fig 7-7.

This morning my tank top temperature was 120F, and I still had plenty of heat energy to supply a good hot shower. I no longer store any DHW in a tank.

At some point it would be fun to to run these examples through SolidWorks and actually "show" what happens. Similar to the race car drafting example at the SolidWorks site, wild stuff.
The actual flow may not always follow the red and blue lines exactly as drawn.

I think we have that simulation capability in Italy, if only I spoke the language

ced48

Rich said:

You don't need the VR1816s , get the VT2218 instead if you want to pump this system right .

The last thing you want to do is install a delta t circulator. All that will happen is that your circ and boiler computer will be working against each other, adjusting and readjusting, until they both get so confused, the boiler will shut down on some kind of error. Go with the 1816, it is a fantastic circulator.

Also, I think you are correct, a buffer tank is your only solution. Your boiler is just to big for your house.

JustinS

ced48 said:

Go with the 1816, it is a fantastic circulator.

How would you say the VR1816 compares to the Grundfos Alpha?

SWEI

A bit less head and flow at maximum output. Lower minimum curve -- important on most residential systems. Fully variable fixed speed (versus three fixed speeds on the Alpha.)

Rich_49

Ced48 ,
Maybe you could explain to me why the boilers computer and the circ would fight each other ? I am interested in why you believe that your statement is true . Let me start by saying that the circ does not care what temp the SW is and does not recognize that this parameter changes , it only recognizes and adjusts for Delta across the system on it's side of the boiler

SWEI

The problem is that the control loops in the boiler and the circulator have different coefficients.

ced48

What SWEI said, both to the Taco-Grundfos, and to why a modcon won't work together with a delta t circ-I tried a Taco Bee with a Loch, and within minutes, the boiler had shut down.

The two units will be constantly adjusting to the changes each is making, eventually confusing one another. Maybe I am over simplifying, but that seems to be what happens.

SWEI

Same experience here. Three on direct-pumped mod/cons and two that were used as primary pumps on a multi-boiler setup. Had to set them back to fixed speed to stop the lockouts. Still better efficiency than a fixed speed pump, but nowhere near optimal.

If the pump makers would give us 0-10V control inputs on the smaller circs, we could sell them all day long as primary pumps on commercial multi-boiler setups. Even if the boiler does not have an output (like Lochinvar nicely provides) we can manage them using the outboard controls we provide. The first system we did this way used 0-10V analog inputs to the 399k boilers and we simply paralleled the 0-10V inputs on the pump controllers to the same controller outputs. By setting the minimum and maximum speed on the pumps to match the modulation rate of the boiler, they actually track within a couple degrees after the boiler has warmed up. Simple, but effective.

hot_rod

Hmmm, just as many here predicted, a bad marriage.

Thermal Equilibrium, all systems will seek and find the point of thermal equilibrium, it's basic physics. No need to try and trick them into some off the wall operating condition.

While selecting a delta T number is important for design, in our hydronic world that can be anything from 10- 40°∆T or any number between depending on the heat emitter.
The ∆T is important for sizing and predicting system performance, there really is no valid reason why a system should, or needs to work at exactly that design number, all or most of the time.

As the conditions that the heat emitters "see" changes, so does the delta T within the system.

As as many have observed, those that take the time to sit and watch a system go through a heating cycle, the system delta T moves until equilibrium is found. Unless you try to over control it with limitations.

Plenty of great applications for delta t circulators, concentrate on those.

I agree Kurt, why is it so hard to get those small 0-10V ECM circulators. They are all over the European market. Maybe we should get into the 120- 240 inverter business. That is how the original Alphas came to this market.

Rich_49

Pardon me Gentlemen but I mistakenly thought we were discussing installing a buffer tank in this thread and that the Delta T circs would be on the opposite side of the buffer from the boiler pump . Those Delta T pumps would go a long way toward keeping the buffer tank stratified , maybe I read # 17 wrong which verified my thoughts about keeping the buffer layered .
3 1816s would certainly work but as you all know , I think moving more water than you need to in order to deliver the required heat is counterproductive (sorry that was not there earlier) . Maybe it's just me , I design my systems with Delta T , proper flow for as long as possible , and varying water temps . Remember , I am speaking of the system side of the buffer . I could also almost swear that someone in this discussion once stated that he uses Delta T circs as zone pumps for series type stuff . Again , maybe I was mistaken .

hot_rod

Rich said:

Pardon me Gentlemen but I mistakenly thought we were discussing installing a buffer tank in this thread and that the Delta T circs would be on the opposite side of the buffer from the boiler pump . Those Delta T pumps would go a long way toward keeping the buffer tank stratified , maybe I read # 17 wrong which verified my thoughts about keeping the buffer layered .
3 1816s would certainly work but as you all know , I think moving more water than you need to in order to deliver the required heat . Maybe it's just me , I design my systems with Delta T , proper flow for as long as possible , and varying water temps . Remember , I am speaking of the system side of the buffer . I could also almost swear that someone in this discussion once stated that he uses Delta T circs as zone pumps for series type stuff . Again , maybe I was mistaken .

I think you are the one that brought pumps into the discussion on your Oct 1 post?

Certainly we want to stratify a buffer tank for max. performance, but a delta circ is not needed to assure that. If it is on the boiler side with that TT boiler, sounds like they conflict? Are you seeing this as Kurt and others are? mod cons and delta T pump conflicts?

On the system side of the buffer, if it is zoned as this system is, delta P has a proven track record, and is exactly what all the ∆P circs are being marketed for.

I don't think a circ knows or cares which side of the tank it is on. A buffer looks a lot like a high mass, large water content boiler to a pump. And in either case the system is striving for equilibrium.


I like the avatar pic, by the way.

JustinS

How

hot rod said:

Good questions. To answer them you would need to know the flow rates.
2 piping option: use figure 3-4 and assume you have that 10gpm from the boiler and exactly 10 gpm going to the load. No flow goes into the tank, all the flow from the boiler goes to directly, and quickly to the load.

As load drops off, that % goes through the tank fig 3-4 again, and starts the buffering function.

Trying to understand something... In this example with 10 gpm from boiler and 10 gpm going to the load, the buffer tank is essentially bypassed... Presumably, this represents "full load" and as zones stop calling, the flow to the load drops off with flow now going into the buffer tank...

I see how the buffer tank absorbs excess heat from the boiler as zones drop off but what I'm having trouble with is how to handle changes in the water temp, as the weather changes, for a given number of zones calling...

As the temp drops, it seems that I would need to reduce flow to the zones so that more of the boiler flow starts going into the buffer tank; otherwise, I imagine I'd be right where I am now, with no buffer tank in effect...

Do I need delta-T pumps on each zone?

I hope what I'm asking makes some sense...

Rich_49

Yes Bob . I inquired about circs . In order to demonstrate that one type of pump would outperform the other in this particular instance .

There are as we all know , two ways to vary the output of an emitter . One is to vary SWT , the other to vary flow . We can do both and exceed what has been our prior capability .

There is no denying that as heat loads and SWTs lower so does flow and head .

Nobody here or anywhere for that matter ever heard me advocate using a Delta T circ as a boiler side circ or directly connected to any mod con with the exception of The HTP Pioneer or Versa Flame . Maybe we can all start recognizing that we are chasing our proverbial tails searching for a lower turndown boiler so we can provide the realtor with 25 more sellable sq ft . Boiler , 2 circs and a buffer tank .

With the new homes that are being built and the Deep energy retrofits becoming commonplace we all better wake up or learn to watch a bunch of Foreign , electric sources take over while you and I d%#k around debating what circ to use . Want a solution ? Use a boiler with mass that can be fine tuned with programming on a system by system basis , use whatever circ you want , IDGAS which you prefer and give the customer a good job and value . The products exist right now if anyone really gave a damn about doing it right .

Justin ,
A Delta T circ on each zone would insure that you are moving the least amount of water necessary to cover the load whilst allowing the Buffer not to mix for the longest period of time until which time that the Delta T circ has reached the low end of it's duty cycle .

hot_rod

JustinS said:

How

hot rod said:

Good questions. To answer them you would need to know the flow rates.
2 piping option: use figure 3-4 and assume you have that 10gpm from the boiler and exactly 10 gpm going to the load. No flow goes into the tank, all the flow from the boiler goes to directly, and quickly to the load.

As load drops off, that % goes through the tank fig 3-4 again, and starts the buffering function.

Trying to understand something... In this example with 10 gpm from boiler and 10 gpm going to the load, the buffer tank is essentially bypassed... Presumably, this represents "full load" and as zones stop calling, the flow to the load drops off with flow now going into the buffer tank...

I see how the buffer tank absorbs excess heat from the boiler as zones drop off but what I'm having trouble with is how to handle changes in the water temp, as the weather changes, for a given number of zones calling...

As the temp drops, it seems that I would need to reduce flow to the zones so that more of the boiler flow starts going into the buffer tank; otherwise, I imagine I'd be right where I am now, with no buffer tank in effect...

Do I need delta-T pumps on each zone?

I hope what I'm asking makes some sense...

At the top post you mentioned low mass baseboard, so I assume all zone run about the same temperature, no mixing required for the three zones?

You also mentioned 165F at design -10F, yes? Do you have output tables for the board that shows output at lower temperatures? how low, do they show? maybe 120F?

When typical no-con boilers, wood, pellet, solar, are buffered we run the tank to 180 or so, then pull the load via ODR. That give us the stacking and most amount of BTU storage.

With your condensing boiler that may not be the best way to run for optimum condensing efficiency. Probably put the boiler sensor in the buffer and reset the tank via the boiler. You give up some buffering leverage this way, as the tank temperature is set by the ODR demand, and you want to keep the boiler condensing as much as possible. You can't have it both ways,

If you are sure of the required flow rates , you mentioned 40K at design? No sense in 3 pumps for that load.
I'd suggest a single ∆P circ, zone valves, and circuit setters to dial in the exact flow.

The ODR controls the required temperature, the smart pump supplies the required flow based on valves opening and closing.

Here is some good reading on pumps and smart pump settings. This issue was developed to clear up confusion about smart pump operation and application.

http://www.caleffi.com/sites/default/files/coll_attach_file/idronics_16_na_0.pdf

JustinS

Hi Hot Rod

hot rod said:

At the top post you mentioned low mass baseboard, so I assume all zone run about the same temperature, no mixing required for the three zones?

You also mentioned 165F at design -10F, yes? Do you have output tables for the board that shows output at lower temperatures? how low, do they show? maybe 120F?

That is correct, all 3 zones are low-mass baseboard and there are no mixing valves or the like to vary the temperature by zone... each of the three zones just draws from a common supply header and its return goes to a common return header...

Supposedly, the house was designed to a 0F day but I never saw a heat loss at that temp; as I've struggled to fix my issues, another HVAC company came in and did a heat loss at 9F for "average construction" of ~46K BTU/hr... this works out to ~730 BTU/hr/F, based on a 72F indoor temp...

Last year in Massachusetts, it got to be -10F for a few days and I didn't have any trouble keeping the house warm (~70F) during that time... based on my above heat loss of ~730 BTU/hr/F, this would have been a total heat loss of ~58K BTU/hr.

It's not so much that my design temp was -10F but rather that the ODR curve topped out at 165F for 10F outside; since the house was staying warm when it was colder outside, I saw no reason to change it and just left it at that...

My baseboard is Haydon Heat-Base 750 and its output table actually goes as low as 90F (see here)

Based on my 90ft of baseboard (rated at 485 BTU/hr/ft at 165F) plus my kickspace heater (rated around 7400 BTU/hr at 165F), my total heat dissipation would have been ~51K BTU/hr...

This leads me to question the accuracy of the heat loss since it would have been higher than my system was putting out at the time... My intent (once the buffer tank is installed) is to play with the ODR curve over time and see just how low I can go...

Would you agree with my reasoning?

hot rod said:

With your condensing boiler that may not be the best way to run for optimum condensing efficiency. Probably put the boiler sensor in the buffer and reset the tank via the boiler. You give up some buffering leverage this way, as the tank temperature is set by the ODR demand, and you want to keep the boiler condensing as much as possible. You can't have it both ways,

I think what you're saying here is that the buffer tank temp would essentially move to whatever the boiler setpoint is over time, right?

Thanks a lot!

hot_rod

At this point it really doesn't matter if the load is 40 or 50K. You have data showing that you can maintains the home at below design conditions, that's the good news. It would be better yet if you could do that with 140F or lower supply temperature instead of the 180 you mentioned. Dialing in the ODR may lower that max. number.

Remember at low supply temperatures the convection through a baseboard really slows, and you may not "feel" the heat. Some homeowners complain about not feeling warm heater emitters, be it BB or radiant slabs.

Seems like you still have a boiler that is twice the size you need, and I think that is what you are looking to deal with.

The boiler has limited capacity to be down-rated like some of the newer controls with ramp delay, ramp limiting, etc.

So does it make sense to add a buffer to help minimize short cycles.

More good news is the fact that a smaller buffer can be used with modulating heat sources, page 28 of Idronics 17 has the math for that..

Until we have VS pumps that can ramp with the boiler, although Viessmann has, modulations as Kurt mentioned, you have to deal with the fixed speed circ on the boiler. TT should list min flow rate accepted with that boiler. No need to pump 10 gpm thru the buffer if the load is never over 50K?

Essentially you are "site building" the boiler that Rich mentioned above, and you can pump the distribution side with a VS circulator and not "upset" the boiler, if you add a buffer.

Now decide if the upgrade is worth the time and money for what it will add to your system? Or just replace the TT110 with a 50K mod-con, there are some available with 7 or 10 to 1 turn down, that would fit your system to a T.

SWEI

Had to go back and re-read this one from the beginning. On the system side, a single pump with zone valves will be more than adequate for that heat load. ΔT or ΔP we could argue for days about, but either will work far better than a conventional circ.

JustinS

Hi again - started talking seriously with Phinney this past June, discussing piping plans and such.

Unfortunately, I've not heard back from him in several months - last notice was that he was handling some personal things.

Any other folks able to help in my area?

Thanks

Steve_Q1969

Are you still looking for a contractor?

JustinS

Steve_Q1969 said:

Are you still looking for a contractor?

Hi, yes I am

Steve_Q1969

I am in the South shore area of Massachusetts and could come out and take a look and see what is needed.