Feedback needed on new system piping diagram

[700renegade]

Sometime this summer I need to pipe up a new install in a 60x60 workshop I built. Using a EKO 40 with two 1000 gal vertical LP tanks for storage. The diagram attached assumes I'd use something like a Taco 0010-VS primary pump at the boiler (set on reverse acting, sensor set at 185* on boiler output). I have a 30,000 btu electric boiler and a Central Boiler Maxim corn boiler sitting around so may pipe them into the loop as backup sources. Any automated control interfaces between the three boilers are yet to be determined ( both the electric and Maxim can self start themselves based on temp setboints)

A number of the diagrams I've studied have limited means to prevent mixing at the tanks and can return quite hot water to the bottom when drawing off storage. The 3-way valve scheme on this is an attempt to take water returning from my house (water - air HX) thru my shop hydronic system to get the return temps down to a minimum.

Unless I'm missing something obvious, this seems like a relatively straightforward way to drive the exact correct BTU out of the boiler into storage at 185* without mixing. Running off storage requires no pumps (other than the heat circuits themselves) as compared to secondary and injection loop systems

Please provide comments:

 

[woodsmaster]

6 of 1 half dozen of the other.

 

[ewdudley]

A number of the diagrams I've studied have limited means to prevent mixing at the tanks and can return quite hot water to the bottom when drawing off storage. The 3-way valve scheme on this is an attempt to take water returning from my house (water - air HX) thru my shop hydronic system to get the return temps down to a minimum.

Would definitely mitigate hot return problem but wouldn't solve it unless loads and flows all match up just so. To address the problem head-on you could consider having the house WAHX pulling with its own pump and returning to the shop-to-house pair though a thermostatic diverter valve, or run a small buffer tank in the house with a transfer pump controlled according to return temperature in the bottom of the buffer tank.

In any event, it could work out nicely to eliminate the three-way valve by pulling the low temp hydronic load supply upstream of the house loop flo-check, so the hydronic load pulls all the way through the house loop whether the house loads are running or not. The net flow to the hydronic loops should be quite low so the hydronic pumps wouldn't see much extra resistance even pulling all the way over to the house and back. This would work particularly well in conjunction with a buffer tank strategy on the house side since the buffer tank could maintain a reservoir of return water that the hydronic load could draw from.

Unless I'm missing something obvious, this seems like a relatively straightforward way to drive the exact correct BTU out of the boiler into storage at 185* without mixing. Running off storage requires no pumps (other than the heat circuits themselves) as compared to secondary and injection loop systems

Please provide comments:

There seems to be plenty of guys who are controlling supply temperature with an aquastat and it appears to work A-OK. It implies that the boiler will be stratified, but I don't see how that could be a disadvantage as long as the water on the bottom is above 140 degF, which the Danfoss guarantees.

 

[Bob Rohr]

Are the BVs at the tanks to bypass storage and send boiler output directly to the loads? Without isolation you will need to heat the tanks to the Danfoss setpoint before you could pull much heat to any loads.

It's very hard to predict how much flow goes where under different operating conditions. With the boiler pump only running, perhaps 8-10 gpm, which is plenty to move the output of a 40KW boiler. But as load pumps kick in, which are piped in series, the flow gets un-predictible. Boiler pump and house pump running may prevent any flow thru the radiant as it probrably has a much larger pressure drop. It could be simulated if you can define all the load GPM and pressure drops.

You have some great potential to use one of those buffers as a hydro-separator. Or add a small separator into the system, add all the inputs on one side, the outputs on the other. With this current piping it seems you will need to do a lot of manual balancing and change settings as loads come and go?

I not convinced you ever need a balance valve on a thermostatic mix boiler valve. It defeats the intention of the valve to start regulating the bypass flow with a valve. We have tested many valves like that in our lab, the only ones the "need" a bypass restriction valve are the ones that have "leakage" either built in or do to poor tolerances or design. You want as much bypass flow as possible to get the t-mix temperature up to setpoint as soon as possible.

Here is a building block schematic for using a hydraulic separator with all inputs on one side, loads on the other. There are ways to bypass the large buffers until some of the heat loads are satisfied, it depends if you want to manually control that or use some 3 way motorized valves to make the change overs.

When I look at Euro designed wood, fossil, solar fueled systems it is not un-common to see a handfull of 3 way valves to select different operating conditions. If you want an automatic or somewhat automatic system it takes valves and operaters that can respond to temperatures. but as you know it adds cost and complexity to the system. You really cannot "pipe in" as dependable, predictable control logic without some sensors, operators and hydraulic compatability.

hr

 

[700renegade]

Thanks for the great replies so far.

EWD - I love the idea of pulling the shop hydronic off the house loop return. It eliminates the three way valve, and when the shop hydronic is running alone, it keeps my house W-A HX and my sidearm waterheater hot. I'll get a certain amount of heat into the furnace ducting just from convection off the hot W-A HX without the furnace fan running.
The reason I was thinking of a Taco variable speed setpoint pump on the boiler is to maximize the stratification in the storage tanks. Seems guys have problems matching gpm with the tank temps when they get close to saturated. Initially I'd only need 6.2 gpm ( 140k/500/(185*-140*)) but at the end of loading the tanks may need up to 14 gpm (140k/500/(190*-170*)). I suppose if the tanks are really well stratified it wouldn't matter as the change from 110* water at the bottom to 180* water would happen in a small number of gallons ( sort of like my electric water heater, when it runs out, it gets cold FAST ).

Bob- The only reason for the valves at the tanks are to isolate for whatever maintenance reason, at the end of shoulder season I could drop down to one tank charging, etc.
I'm puzzled why the tanks in their current configuration don't work as a 'hydro seperator'. What functionally is different if I happen to draw the load piping on the right hand side of the 1000 gal tanks? Lets say my boiler is producing 185* water and 6 gpm headed uphill towards tanks and loads - if I'm pulling 8 gpm in my house loop, 2 gpm pulls off the top of my tanks. If they are charged at 185*, great.... if they are cold ( 110* ) my 8gpm water to house is blended, but still 162*.
I don't see the need for any "load balancing and changing settings" with what I have here. The only variable is what return water temp is shoved back into the bottom of my tanks. I'm probably missing something.....
Regarding balance valve on feed to Danfoss valve - I'm tickled you pointed that out. I've always wondered what in the world it would ever be used for, but it seems like every diagram has one, so I figured I was too dense and better put one in just in case. If you've tested them and have a recommendation on what model TV to use, I'm all ears.

 

[maple1]

Regarding balance valve on feed to Danfoss valve - I'm tickled you pointed that out. I've always wondered what in the world it would ever be used for, but it seems like every diagram has one, so I figured I was too dense and better put one in just in case. If you've tested them and have a recommendation on what model TV to use, I'm all ears.

That one's had me puzzeled from the beginning also. The only way I could make it make sense in my head, is that the bypass (hot) side of the Danfoss must be wide open all the time, and the return-from-storage side opens as it sees warmer return water. Not sure that is the way it works, but is the only reason I could come up with for using a valve to throttle the bypass loop - which seems an odd way to have it work, overall. I would have thought it would have been designed that the hot bypass side closes at the same rate the return side opens, and nothing would need throttled.

 

[ewdudley]

Regarding balance valve on feed to Danfoss valve - I'm tickled you pointed that out. I've always wondered what in the world it would ever be used for, but it seems like every diagram has one, so I figured I was too dense and better put one in just in case. If you've tested them and have a recommendation on what model TV to use, I'm all ears.

Well according to the Danfoss data sheet:

To ensure proper flow to the system an adjustable balancing valve is recommended to be installed on the bypass between the supply and return piping. The piping of the balancing valve creates a similar resistance as the system to reduce the constant recirculation of heated fluid through the boiler and not out to the system.

Seems like normally a balancing valve wouldn't be required since the piping to storage would be pretty similar in resistance to the bypass loop.

 

[Bob Rohr]

Well according to the Danfoss data sheet:



Seems like normally a balancing valve wouldn't be required since the piping to storage would be pretty similar in resistance to the bypass loop.

I can't speak for the other brands but when I asked the engineers that designed our valve they told me if the boiler is connected to a storage vessel there was no need for a balance valve. It could be other brands do not shut off tightly and the valve is intended to add some flow resistance. We know that is a fact when we flow test the competitors in the lab.

I know our TRVs use a wax filled actuator while may others use a gas. The wax has better properties that allow us to shut off higher pressures. It could be that is true of the actuators use in some brands of these valves.

How would you determine where to adjust the balance valve if used. In one case the manufacturer shows a valve with p/t ports that a differential meter is required to balance. I doubt many installers own such a meter to even dial them in properly?

 

[Bob Rohr]

Thanks for the great replies so far.

EWD - I love the idea of pulling the shop hydronic off the house loop return. It eliminates the three way valve, and when the shop hydronic is running alone, it keeps my house W-A HX and my sidearm waterheater hot. I'll get a certain amount of heat into the furnace ducting just from convection off the hot W-A HX without the furnace fan running.
The reason I was thinking of a Taco variable speed setpoint pump on the boiler is to maximize the stratification in the storage tanks. Seems guys have problems matching gpm with the tank temps when they get close to saturated. Initially I'd only need 6.2 gpm ( 140k/500/(185*-140*)) but at the end of loading the tanks may need up to 14 gpm (140k/500/(190*-170*)). I suppose if the tanks are really well stratified it wouldn't matter as the change from 110* water at the bottom to 180* water would happen in a small number of gallons ( sort of like my electric water heater, when it runs out, it gets cold FAST ).

Bob- The only reason for the valves at the tanks are to isolate for whatever maintenance reason, at the end of shoulder season I could drop down to one tank charging, etc.
I'm puzzled why the tanks in their current configuration don't work as a 'hydro seperator'. What functionally is different if I happen to draw the load piping on the right hand side of the 1000 gal tanks? Lets say my boiler is producing 185* water and 6 gpm headed uphill towards tanks and loads - if I'm pulling 8 gpm in my house loop, 2 gpm pulls off the top of my tanks. If they are charged at 185*, great.... if they are cold ( 110* ) my 8gpm water to house is blended, but still 162*.
I don't see the need for any "load balancing and changing settings" with what I have here. The only variable is what return water temp is shoved back into the bottom of my tanks. I'm probably missing something.....
Regarding balance valve on feed to Danfoss valve - I'm tickled you pointed that out. I've always wondered what in the world it would ever be used for, but it seems like every diagram has one, so I figured I was too dense and better put one in just in case. If you've tested them and have a recommendation on what model TV to use, I'm all ears.

Here is what I understand about a hydraulic separator. It basically provides a wide spot in the piping. The key is a 3-1 ratio. The center barrel needs to be 3 times the diameter of the pipe connected to it. This allow flows to cross within the device without pushing flow thru the opposite side. So, in fact, different sized circs can pump through a separator without adding flow to each piping system.

With a tank that size the ports would not need to be direct opposite on the tank, as long as there was some separation to allow multiple flows. I've built some of these separators with clear plastic to demonstrate the concept in my training. A large high head pump on one side, a smaller one on the other. red food coloring in one side, run that pump for 20 minutes and no color moves into the other side. Then blue color in the B side, turn off A side pump and the colors stay the same. Only when both pumps run does the color blend. The amount of blending of temperature depends on the two different flow rates. The formula in I-dronics 1 tell how to calculate the mixed temperature, but you need to know flow rates.

If you have a means to calculate the pressure drop in the various zones you can determine with a pump curve and system overlay, exactly what the operating point is on the pump curve, then you will know exactly how many gpm is flowing in all the zones.

Or add a flowmeter that will give you a visual of how much flow is moving in various loops. The only way to assure you have the flow you want, or need, in every zone is to calculate all the resistence, or add a flow setter/ balancer to force it into the desired operating point. The system will never just self balance and provide the flows you write down on the design sheet . That is why engineers always include flow setters on larger or complex piping systems, you need to balance systems with complex piping. There are actually companies that specalize in balancing hydronic systems and on most large jobs the M.E. require a balance contractor sign off.

Then you need to decide what type of balance valve. In your drawing I'd look at a PIBV pressure independant balance valve. You buy this valve with the flow rate you need speced out. This type of valve has a spool that slides around and assures that piping loop ALWAYS gets the correct GPM regardless of how many other zones or pumps turn on and off.

Even with hydraulic separation you may need balance valves on some or all of the piping zones, or at least a flow meter to see where the system is operating.

In many cases the various piping layouts we see here work and move heat energy, and good is good enough, trial and error design.

Systems that short cycle, make velocity noise, wear out pumps, have in-adequate heat output, air lock, etc... these are all flow related problems that can be determined or corrected with flow meter and balance valves.

hr

 

[ewdudley]

I'm puzzled why the tanks in their current configuration don't work as a 'hydro seperator'. What functionally is different if I happen to draw the load piping on the right hand side of the 1000 gal tanks? Lets say my boiler is producing 185* water and 6 gpm headed uphill towards tanks and loads - if I'm pulling 8 gpm in my house loop, 2 gpm pulls off the top of my tanks. If they are charged at 185*, great.... if they are cold ( 110* ) my 8gpm water to house is blended, but still 162*.

You are exactly right, a hydraulic separator wouldn't provide any advantage for what you're trying to do. Any flow imbalance will sort itself out where the load supply tees into the hot side.

However you need to convince yourself that the load pump won't pull undesired parallel flow through the boiler over and above what the boiler is supposed to be supplying, whether or not the boiler is operating. If you trust the Danfoss to pinch off as soon as TRet drops below 140 degF then you're golden. If there's any doubt you can simply move where the load tees into the hot manifold higher and closer to top of storage where hot water rising and the relative resistance will guarantee that load supply will draw from storage exclusively.

 

[700renegade]

EWD - in my case the loads will pull off the piping very near to the tanks. I could easily bump up my near tank piping to 1.5" instead of the 1.25" which will go 25' over to the boiler room.

Instead of the $600 Taco VS pump, has anyone used a cheap Grundfos 15-58 3-speed and used an aquastat to control the speeds? I'm thinking - put a thermowell at the bottom of my tanks and anytime the temp there is above roughly 150* to force it to bump the main boiler pump from low speed to high. Logic says at 150* my delta T is getting tight at the boiler and I'm going to need more GPM to drive BTU to the tanks. I'd have to pencil out a way to wire a DPDT relay to cut power from the low speed lead and apply power to high. This would be the poor-man's 'bang-bang' version of a VS pump. If it worked I can easily handle the $450 lump it will leave in my wallet.

Since these tanks are 41" dia and 16 feet tall, I'm assuming I'll have above average stratification and perhaps won't need the higher pump rates.

 

[ewdudley]

EWD - in my case the loads will pull off the piping very near to the tanks. I could easily bump up my near tank piping to 1.5" instead of the 1.25" which will go 25' over to the boiler room.

Instead of the $600 Taco VS pump, has anyone used a cheap Grundfos 15-58 3-speed and used an aquastat to control the speeds? I'm thinking - put a thermowell at the bottom of my tanks and anytime the temp there is above roughly 150* to force it to bump the main boiler pump from low speed to high. Logic says at 150* my delta T is getting tight at the boiler and I'm going to need more GPM to drive BTU to the tanks. I'd have to pencil out a way to wire a DPDT relay to cut power from the low speed lead and apply power to high. This would be the poor-man's 'bang-bang' version of a VS pump. If it worked I can easily handle the $450 lump it will leave in my wallet.

Since these tanks are 41" dia and 16 feet tall, I'm assuming I'll have above average stratification and perhaps won't need the higher pump rates.

Nofossil is out in front on this one, he even calls it a poor man's VS pump control!:

http://www.nofossil.org/index.php?choice=poormanvsc

 

[700renegade]

Nofossil is out in front on this one, he even calls it a poor man's VS pump control!:

http://www.nofossil.org/index.php?choice=poormanvsc

We'll I'll be danged.
I remember opening up a 3-speed a couple years ago and thinking one could switch it's gears electrically. I hadn't gone to Nofossil's site till now. Good stuff in there and I guess there is no more guessing if this is possible.

 

[jebatty]

Seems guys have problems matching gpm with the tank temps when they get close to saturated. Initially I'd only need 6.2 gpm ( 140k/500/(185*-140*)) but at the end of loading the tanks may need up to 14 gpm (140k/500/(190*-170*)).

In my experience it is more complicated that this. I will assume you achieve a very high level of stratification, especially with vertical tanks, as I know that also can occur with a horizontal tank. The thermocline separation between very hot and cold water will be quite sharp. When the bottom of your tanks reaches 170F, there will be very little water volume between the 170 and the 190 water. 14 gpm will work only for a little while because the delta-T will close rapidly. If your boiler is operating at 140K output at this point, then it will be in high burn state, and unless you have substantial other loads you will quickly close the delta-T and your boiler will idle. Ideally your boiler should be operating close to fire out at this point (output perhaps at about 35K and falling). The boiler fire dying out operation will then coast the storage up to 185-190F.

In a no load, tank charge only scenario, a 15-58 on M or a 007 (1.25" pipe, boiler close to storage) works just fine to bring storage up to 190 with no idling. With some experience unique to your situation you can easily calculate how much wood to load in the boiler to result in the wood load burning out as storage moves towards a full charge and delta-T closes so that boiler output falls, flow remains constant, and the entire storage achieves a 185-190 charge. I do this all the time. And with a little more experience you will get a good feel, based on outside temperature, what your heat load will be during a burn cycle and you similarly will be able to load and burn to achieve the same result with demands on the system.

Another method to prevent mixing from warm return water is to inject the return water at one or more tank mid-points rather than at the bottom. A very good solution of course is to have a heating system that only needs supply at 95F or so, and then mixing doesn't make much difference.

 

[Bob Rohr]

In my experience it is more complicated that this. I will assume you achieve a very high level of stratification, especially with vertical tanks, as I know that also can occur with a horizontal tank. The thermocline separation between very hot and cold water will be quite sharp. When the bottom of your tanks reaches 170F, there will be very little water volume between the 170 and the 190 water. 14 gpm will work only for a little while because the delta-T will close rapidly. If your boiler is operating at 140K output at this point, then it will be in high burn state, and unless you have substantial other loads you will quickly close the delta-T and your boiler will idle. Ideally your boiler should be operating close to fire out at this point (output perhaps at about 35K and falling). The boiler fire dying out operation will then coast the storage up to 185-190F.

In a no load, tank charge only scenario, a 15-58 on M or a 007 (1.25" pipe, boiler close to storage) works just fine to bring storage up to 190 with no idling. With some experience unique to your situation you can easily calculate how much wood to load in the boiler to result in the wood load burning out as storage moves towards a full charge and delta-T closes so that boiler output falls, flow remains constant, and the entire storage achieves a 185-190 charge. I do this all the time. And with a little more experience you will get a good feel, based on outside temperature, what your heat load will be during a burn cycle and you similarly will be able to load and burn to achieve the same result with demands on the system.

Another method to prevent mixing from warm return water is to inject the return water at one or more tank mid-points rather than at the bottom. A very good solution of course is to have a heating system that only needs supply at 95F or so, and then mixing doesn't make much difference.

\The last paragraph really sums it all up. Lowest possible design temperature, and multi port tanks. The Europeans are really keen on stratification tanks with a handful of ports on the side. The highest temperature loads from the top, generally DHW which must now run to 140F once a day for anti bacteria issues in some countries, then panel rad loads, and at the bottom the infloor radiant.

 

[stee6043]

Much discussion has been had on the stratification topic. It certainly can be a rather complex issue. But I can tell you from experience that a single dip tube within a few inches of the bottom of the tank combined with a single port on the top of the tank seems to work very well with the flows, pumps and piping you are looking to work with. I measure my tanks top and bottom (6" from each +/-) and I very consistently maintain 20 degrees delta within the tanks. I didn't do anything special with my dip tubes. So I guess I'm saying maybe we don't need to worry about it so much?

I charge my tanks with a 15-58 3 speed on the medium setting and I supply my load with another 15-58 3 speed on low.

Off topic - that EKO 40 of yours is going to have it's work cut out for it to charge two 1,000 gallon tanks. I've occasionally wished I had a 60 just to charge my tanks a bit faster...and I'm half the storage.

 

[700renegade]

Off topic - that EKO 40 of yours is going to have it's work cut out for it to charge two 1,000 gallon tanks. I've occasionally wished I had a 60 just to charge my tanks a bit faster...and I'm half the storage.

I know the EKO 60 would have been a better choice, but I stumbled into a 2 year old 40 for a good price. I'm not a purist like some however - if my fuel oil furnace has to run a bit on some cold days or my shop dips below 50* I'm not too concerned.

So far I've not heard any comments that the single pump system shown won't work. Unless someone is able to point out or sketch up a better option, I think I'll go with what's drawn, with the minor changes EWD suggested.