Mod Con piping with Buffer Tank

Paul Rohrs_4

with a Knight boiler.

I have 5 zones of high-mass radiant floor heat (single reset temp), DHW on priority, and a buffer tank. My smallest zone is 17,000 btu's at design conditions. My KBN210 low fires at 42,000 btu's (input) It will give some very nice run times in mild conditions.

I am mildly concerned about this piping set up. The Knight boiler will control the P1 boiler pump, P2 DHW pump, and P3 System pump. I like this set up in that the zone pumps will pull the reset temp water from the buffer tank even when DHW calls. As you all probably know, the P1 pump will drop off and the P2 pump enagages and ramps up the boiler on priority. I have an integral flow check on the circ (Return) AND a B&G flow-check on the supply to isolate flow.

The Knight boiler does not disengage the P3 pump on a DHW call. Soooooo, I am mildly concerned that even with the flow checks on the boiler loop and closely spaced tee's, the P3 pump still has the POTENTIAL to draw from the boiler loop when on DHW. (?) I thought about making a thermal trap on the boiler loop or moving the DHW piping a little more closely to the boiler inlets. The "Buffer 2" drawing is probably my preference only because we are pre-building this panel in-house before we hit the job site.

Your thoughts?

Regards,

PR

Brad White_9

Paul

I see little difference between Buffer 1 and Buffer 2, just the closer connection to the Knight- so why not?

As for unwanted flow, why is it written in stone that P-3 continues to run? Why not disable it, if it is of concern?

Still, if no zones are calling I think the flow checks would hold, especially if P-3 has a low head.

If P-1 is off-line, I do not see much if any migration to the buffer tank via P-3 (P-2 being so close-coupled and all).

If the DHW calls for the times you can predict, will that be a problem? A short spike that may not show for a while at the slab level and can be mitigated by PID if you are measuring the slab temperature.

What about... another buffer tank or the same one, off-line and in parallel, that will "store" any DHW overage? Use it as you need it. May complicate it, but thinking a place to put any surge.

Random thoughts.

Brad

Paul Rohrs_4

Thanks

Brad, I'm still waiting to here from Lochinvar to see if I can disable P3 on a DHW call.

Hey, whatever happened to the magnets pulling out sediment at your Susan's place? Any thing notable?

Regards,

PR

ALH_4

P3

Maybe I'm missing something in your control logic, but why not eliminate P3 and put the buffer tank in series on the P1 loop? I can't think of a time when use of the buffer tank would be detrimental in heating mode. There isnt a lot of heat stored in the buffer tank anyway, particularly at radiant floor reset temperatures.

Paul Rohrs_4

Andrew

I have done that in the past and typically that is the way I normally pipe buffer tanks. I just thought in this application I would not have to suspend space heating when DHW calls. It would allow a fully charged buffer tank to feed the zone pumps to draw from during the DHW call.

Regards,

PR

Brad White_9

Magnets

Hey Paul-

Having too much fun elsewhere to pull the filter but I should soon. I will let you know. You and I are alike that way, I think...

I have drawn off some water from a low point after the filter to test pH (a nice 8.0 thank you very much!) and it was clear enough to be drinking water. No visible turbidity. So things seem to be working.

Cheers!

Brad

ALH_4

ideas

I missed your point in the first post. I seem to do that a lot lately. Actually piping it that way could be a good idea, the more I think about it. You could possibly even use it as a connection point for solar with the use of an indirect water heater as the buffer tank if the customer was inclined to do so in the future.

A 40 gallon buffer tank dropped 20F releases only 6664 btu, so that tank would cool off pretty fast if it was just at reset temperature. If it was a 120g solar storage tank run up to 180F you might have around 80,000 btu stored. The cool part would be that it would be a buffer tank when you needed one and a solar storage tank when you needed one. Just by virtue of the heat content of a buffer tank at radiant floor reset temperature means the mass of the floor probably does more to keep the heating system from falling behind than the buffer tank does.

I think I'm getting off the point. lol

One thing you could do is make the P1 loop the primary loop and the P3 loop a secondary loop as I believe Hot Rod has done in the past. That gives you a little more flexibility in how you use the tank if you did decide to use the primary loop as a gateway for another heat source.

Joe Brix

Paul

I like the piping concept - it's one I've never considered, but I think it would work quite well.

How are you sizing the P1 pump? My assumption would be that it is sized to handle the total system flow with all zones calling, and taking the boiler head loss into account? Same with P3, only a lower head would be needed due to the small head loss of the buffer, right?

Starch

Chris_82

computer prob...

Try reposting

Paul Rohrs_4

Buffer Tanks

Here is why I endorse buffer tanks. Just some supplemental information.

Buffer Tank 1 shows a 30 gallon buffer tank (Siggy's software)

Buffer Tank 2 shows no buffer tank in that the software will only default down to a 5-gallon capacity.

Cycle times speak for themselves.

Regards,

PR

Chris_82

Corect me, but p3 is acting as the system circuit and the buffer tank acts as a storage capacitor in a sense so the boiler can do its thing when and where it is most efficent. You can heat a larger mass, run p3 untill the boiler needs to run to reheat the system loop plus mass (storage tank), in other words has/is the manufacturer saying there is an optimum energy savings ratio by not running untill needed at higher delta t's and we will let "flyback" provide heat (again the storage buffer tank) when demand isn't that great. ??

On another note I have experemented with running storage tanks, indirects in series as the system loop, in other words no seperate DHW loop, and using 140 as the set point for the radient loops, but I have been getting some really good numbers in energy savings. (despite the temp swings in the on/off which does cause some noise, but nothing compared with older baseboard.)

Sorry back to your question, the buffer tanks as I am suggesting above, are generally not piped in the p1 cir., because that would mean h20 is flowing thru the boiler and this tends to keep it on a continuious low fire, or its lowest modulation setting, this may not be the most efficent and it can be problamatic to the control paradigm when to fire and when not to fire? In other words, and this raises a really good point is that the control dosen't know if a storage tank is connected in series or not, compared with the load, give tekmar a call and see if this feature is being considered? I will say from personal experience running an indirect in series with the DHW set at 140 and a mix down radient with the system set to 160, the heater runs for a long time at its lowest setting, but then again it is off for a long time as well, very low delta t's, the DWH has the expensive sparco anti scald temp valve, and short cycling isn't a problem, but on really cold days, when the radient zones are maxed out you do run out of DHW, but this is rare and was solved by bumping up the tank size.

I have also done the reverse so to speak, piped a 120 gallon indirect in series at 140, system at 180 and the secondary mix circulator set at 110 for the radient, this requires a seperate control independant from the boiler DHW control and h20 does circulate thru the boiler and the indirect operates as the buffer tank, I never thought the coil would transfer enough but it does, and have had great results so far, with the DHW priority off. Control wise...one independent relay, one 5 zone controller, with the DHW tank as one of the zones.

Dave_4

Full Flow vs Partial Flow

Paul,

Have you thought about directing only part of the flow through the buffer tank? If you're using a 40 gallon tank and your pumping 8 gpm, that tank will be at near system temp in only 5 minutes. If you pipe it is a branch off of the same loop and then throw in some globe valves to control how many GPM are going through it verses bypassing it you may find you can better tune the buffer tank.

For instance, maybe you are only doing 5 minute cycles. Simple enough, close the bypass and have all flow go through the tank as it is now. If the system is doing longer cycles, you may only want 3gpm going through the buffer tank. This way you can keep your return temps lower for a longer period of time.

I think that a buffer tank will work best with a condensing boiler if it takes the full average length of a cycle to replace the cooler buffer tank's contents. That way it keeps return temps cooler for the longest possible period of time.

It will also mean that the BTUs in the buffer tank are released back to the system during an off cycle at a slower rate for a longer period of time.


*edited*

I'd also move the buffer connection point to somewhere after the heating branches (right before the boiler return loop ideally), otherwise the area calling for heat will first get 40 gallons of cooler water on a heat call.

jp_2

mid winter

doesn't the need for a buffer tank diminish during colder periods? seems i would put in parallel, use it during shoulder seasons. if you don't need it don;t use it, remember constant losses from that tank.

ALH_4

loads

The buffer tank could help with diurnal load fluctuations during the colder months too.

Paul Rohrs_4

Andrew

"A 40 gallon buffer tank dropped 20F releases only 6664 btu, so that tank would cool off pretty fast if it was just at reset temperature."

Well, let's say we have 40°F outdoor conditions, and my smallest zone (no longer at design conditions) is 6664 btus. Knowing that I will probably not be running 20°F Delta T's, that buffer tank is extending my on/off run times significantly, which is the point of the buffer tank anyway.

In other words, my smallest zone with a narrow delta t in mild conditions will be optimized by the buffer tank. I think of it more as a run time optimizer and off time extender.

Regards,

PR

Unknown

What's the minimum cycle time you're shooting for?

jp_2

over sized?

andrew, wouldn't that then suggest that the boiler is over sized? certainly there is the point of dimishing returns too

Paul Rohrs_4

Rob

I was shooting for 20 minute off times at the minimum.
I could go with a larger buffer tank, but I am comfortable with this in that milder temps and reduced load will give me narrower delta T and maximize the potential of the buffer tank.

I would have liked to have upsold a TN4 package with zone syncrhonization that it offers. I think it is the unsung hero of the TN4 system. Homeowners opted not to accept that or an additional zone of infloor heat in a bonus room above the garage. (Big mistake- but not my money)

Regards,

PR

PHM

How would partial buffer tank flow be better ?

You wrote: "Have you thought about directing only part of the flow through the buffer tank? If you're using a 40 gallon tank and your pumping 8 gpm, that tank will be at near system temp in only 5 minutes. If you pipe it is a branch off of the same loop and then throw in some globe valves to control how many GPM are going through it verses bypassing it you may find you can better tune the buffer tank."
------------

So what? Partially bypassing the thermal-mass tank will not add any water volume or reduce the boiler's ability to heat it. With the bypass valve cracked open - the tank may heat slower but the loop will heat correspondingly faster. And that's what you wanted to avoid in the first place.

Seems to me that the only reasonable place for the buffer tank is full-flow in series with the return to the boiler.

Is that not true? If so; can you explain to me why not?

Thanks!

PHM
--------

Brad White_9

If I may

The concept of off-line/on-line variable buffer tanks is not new.

Like buffer tanks in general, they are used to correct what was once avoided: Mass deleted to favor faster response. As we know, sometimes low mass boilers and low-volume systems sometimes short-cycle and do not respond to low loads very well.

We install a buffer tank (add mass) to correct a problem caused by a solution. Trouble is, that high mass solution hurts us sometimes. Issue is, how do we bring on part-time help as needed? Enter valves or charging circulators and controls.

The notion of having a variable buffer tank allows full capacity brought to bear when the entire system is calling and short cycling will not be an issue. When bypassing the system (all or in part) you charge the buffer tank for later use. Think if it as an off-line piggy bank. A short-term Certificate of Deposit if you will. Truly liquid cash flow.

Uni R_2

My point was to consider potential advantages of partial flow.

Is your issue with location or partial flow?

As for location that was where he had it. I too think that it should be on the boiler return (before the indirect ties back in) unless it is offline storage (and I really can't see the economics of doing offline storage but that would really depend on the application). So on location, we both see this exactly the same. I just used his drawing to show how easy it would be to make it partial flow (it also makes it easy to isolate if it ever leaked or they wanted to manually bring it off line for any reason). A buffer upstream of heating circuits is obviously wrong.

As for partial flow (the actual point of my post) - why have the coldest water all being returned at the beginning of a cycle if you are dealing with a condenising boiler? Wouldn't it make sense to adjust the flow so that the return is mixed down for the full duration of the cycle, especially if this is a marginal condensing application and there is a chance that the boiler may not be condensing at the end of a cycle? And even if that isn't the case, wouldn't the efficiency of the boiler be optimized by mixing down the return temps for the full duration of the cycle, rather than pumping through the cooler 40 gallons of water in the first few minutes?

Why would this whole concept seem a lot more convincing if Brad posted it? ;-)

PHM

First let me say that I am mean this in the gentlest way -


I say that because I don't want you to think I am arguing - I'm not. I just always want to know more. Maybe: "I'm stupidly curious" would be a better way to describe myself.

And I'll come back to My Actual Tale, and how it relates to all this, a little later.

Your comment is making me think harder -

Now then; I agree with your idea to maximize the cold water returning to a condensing boiler. But I also think that the buffer tank temp would always be colder, longer, than would the boiler discharge water merely cooled 15-20 degrees though one pass of the system. Although the loop temperature is likely to be Initially lower, the small loop volume would soon have it up at boiler discharge minus loop delta T. Right? Whereas the large volume tank temperature would take a longer time to increase and -
--------------
Let me stop to point out: counter-flow through the buffer tank? Yes? That is: system return in the top and then off the bottom to the boiler inlet? Yes? Isn't that how a buffer tank should be piped?
--------------
- so be supplying colder water to the boiler for a longer period of time.

Or; am I missing something?

BTW: the buffer tank's secondary function as "off-line storage" would serve to prolong the OFF cycles. But is this really such a good thing?

Which reminds me that I want to also discuss 'cost effectiveness' too. But I'll do that separately in a minute.

PHM
---------

PHM

This is my project and how it came to be -


I have a Laars Endurance boiler. Plate-frame HX for domestic hot water. Half the space heating load no longer exists. So now the burner will cycle off even during a continuous call from the zone stat. That bothered me. I always like to see long relaxed on cycles.

Clearly added thermal mass, in the form of a buffer tank in the loop, would increase both ON and OFF cycles dramatically

And yet, somehow or other, the idea of having an extremely low water volume boiler and then adding a buffer tank to add a lot of water volume to it bothers me. It just seems so contrary to the original concept! For that matter; why not just install a regular 83% cast iron boiler with a domestic coil and be done with the whole mess for 50 years?

Nonetheless; my boiler is low-volume and short-cycles on low load. If fact; on any load. Now then; I have a presently un-used 80 gallon electric tank here so as of this morning I was going to pipe it in series with the return line from the space heating coil. This would obviously add a larger thermal 'flywheel effect' and make the boiler ON times much longer (at least until the zone stops calling for heat) and certainly make the OFF cycles longer too - as the stored heat would be used on the next call from the zone.

No brainer, eh?

So about lunch time today I happened to be standing next to the boiler (located right under the 1st floor bathroom). Shireen must have washed her hands because I heard the water in the sink drain over my head start running. But what really drew my attention was that I instantly heard the boiler's domestic water flow switch click closed, and then the digital readout on the boiler went from showing ON (which means 'stand-by') to showing the boiler temperature - 156. It immediately dropped to 155 and the boiler came on. Shireen then shut the faucet. The flow switch opened and the boiler stopped soon after. Boiler temp 180 degrees. This entire sequence took less than one minute!

My point is this: the people who made the boiler must have been aware of this potential for burner short-cycling on short calls for hot water - as their HW flow switch starts and stops the boiler.

So if it routinely short-cycles on HW demand, why should I care if it also short-cycles on heating demand? Right? Here I am - going to pipe in a buffer tank to stop heating cycle short cycling but domestic water calls can short cycle it any time all year round? What will I be saving?

And . . . there is no way around it: I have on-demand hot water - so adding a buffer tank to That loop is clearly self-defeating.

Take a shower for example: two gallons of water per minute weighs 16 lbs. and contains 1040 boiler-added BTU's if the CW came in at 60 and the HW is 125 at the shower head. Times 60 minutes in an hour means the boiler has to produce 62,500 BTU's per hour to maintain that water temperature and HW flow rate. The minimum firing rate of this boiler is 81K. So even one shower will have the boiler cycling on and off. And small HW water uses can have the burner short-cycling routinely.

So I called the maker's tech line. On hold for an hour of course and then I asked the questions and described the circumstance. He said: So it ran for a minute? That's not short-cycling. Short-cycling to us is running for 30 seconds or less. A minute's fine.

Say what? Are you Sure about that? Let me talk to engineering. Then the engineering boys tell me exactly the same thing: A minute of run time? Sure; no worries there.

Ok then; what about increased fuel use from all that on and off all the time?

Negligable - almost unmeasurable we find.

So at that point I think: First of all; I cannot prevent (what I consider to be) short-cycling on HW demand. So why bother about what happens on heating demand?

The only harm I can think of is that the HSI may not last as long, and that the boiler may use slightly more fuel. And I'm not even sure about the fuel part. specially after the maker's guys dismissed it. So I may as well do nothing and see what happens - hell maybe the HSI will last five or ten years no matter what I do.

So? What do you all think?

Thanks!

PHM
---------

Brad White

I think

you should start your own thread!

Paul Rohrs_5

What if....

Look back at the original post and the "Buffer 2" picture.

What if the P1 boiler pump had (edit) smart contacts with PID logic to control variable-speed injection mixing circ or a diff pressure pump.

When the "Conditional Priority" of the DHW system gets close to satisfying and the btu's and firing rates become available, why not slowly ramp up the P1 pump as an injection pump to simultaneously satisfy DHW demand and bring the system temp back up based on a target temperature set by the S1 system sensor loop temp.?

Best of all worlds. DHW on conditional priority, system P3 pump continues feeding space heating zones until the P1 can start re-injecting heat as needed based on target temp.

Seems attainable to me. Loooooonnnng run times integrating DHW and space heating. Long off times for high or low mass systems.

Paul

Alex Giacomuzzi

Nice Design --- May Be Able to Simplify..

Paul:

I like your piping diagram and your control questions. May I offer some possible thoughts at maybe being able to simplify the piping scheme and provide you an answer to your P3 question. I am making the assumption that continuous circulation is not in anybodys plan first of all and that we can modify the piping slightly.

From what I surmise, with the piping schematic you have shown, P3 is required to run for two reasons:
1. To reheat the buffer tank
2. To send heated water around your secondary loop or the buffer tank loop when you receive a call from one of your zone thermostats.

I do not know which buffer tank you are using or how many taps it may have available, but "what if" we did the following:
1. Eliminated P3
2. Connected P1 directly to the buffer tank preferably with its own taps or possibly with shared taps.
3. Eliminated the end loop of the buffer tank loop (left side).

The buffer tank still receives the heat that the boiler intends to deliver to it. Each of the zone circulators are more than capable of supplying their individual zone and drawing their supply from the buffer tank. You would want to review the total flow and make sure you have flow checks appropriately located to eliminate any short circuited flows. Provided you have sufficient openings in the tank, you could eliminate a pump along with your control circuit concerns.

I personally do not think you would have any problems with your original concept which ever way you chose to pipe it.

Regards Alex

hr

I'm leaning more towards non storage

for the DHW. After some good track record with several mod cons driving either plate type or small"flash" tanks.

Here is one idea I have been noodling. A Munchkin Contender arrived as the missing link for my idea.

The ErgoMax acts as a buffer and the DHW preheat. The DHW Hx merely has to boost the DHW from the boiler operating point, generally around that 100-120F for my radiant designs.

A plastic boiler! Who'd have thunk it

hot rod

To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"