Pressurized storage manifold questions

[cpeltier]

I have owned a Tarm Solo 60 Plus for 14 years and have been running it with an STSS 1200 gallon storage solution. About a year ago the tank started overflowing because my heat exchangers have been leaking. I removed and replaced it with 3 - 250 gallon LP vertical tanks I purchased from Smokeless Heat. I'm in the process of piping the boiler now and would like to know from experience the best way to plumb the tanks. I've studied serial and parallel arrangements and think that maybe the easiest way to pipe it is parallel. I have very tight space constraints and can only place the tanks as shown in the attached vertical view drawing. Does anybody see any problems ganging the tanks this way? I plan to use ball valves on each tank connection as a balancing valve and adjusting where necessary. I've noticed in Europe they are quick to use a Bivalent shunt to select the lower half off the tank for withdrawal first, then switching or mixing to the upper tank port. Is this necessary? I would think the hot water would just rise in the tanks as it is pulled off via convection and the tank stratification would remain intact assuming I was careful with the return temps.

I have both radiant heat loads (via a variable speed injection pump) and 3 fan coils. The return water should be about 110 degrees at design temp from the radiant heat system but could be much hotter from the fan coils. I may run an injection pump from the storage tanks into a secondary loop which feeds the fancoils to lower their supply and return temps (25 delta T) . I'm using an EasyIO FG Beast 32 for a control platform (previously ran Tekmar) so controls are not an issue.

 

[ewdudley]

I'm convinced it's hard to go wrong and easy to over-think it, nevertheless parallel is the only way to go. The manifold you describe should work A-OK since the paths to tanks are more or less the same length, give or take an elbow. I wouldn't fool with the balancing valves, the worst that can happen is the tanks don't run out of hot water at exactly the same time.

It doesn't make sense to me to implement the bivalent shunt unless you are planning on having some solar input that the radiant loads can take advantage of separately from the high temp loads.

Sounds like you have a good handle on return temperature management, but as always I suggest using a hydraulic separator/buffer tank when there are high temperature loads and lower temperature radian loads in the system, if you're looking for additional ideas (post #9):

https://www.hearth.com/talk/threads/connecting-into-existing-boiler.133774/

 

[Tennman]

Agree with EW's over thinking comment. Based on the success of our system I recommend another boiler return manifold on the very bottom of the tank dome. My bottom manifold doubles as the boiler return and storage drain. A very bottom manifold returns the coldest storage water to the boiler. I have a middle mixing manifold but I think it could could be closed and have little impact. I.e. I have five manifolds with the two demand manifolds on the dome to cylinder weld joints and two boiler supply/return manifolds at the extreme ends. They're vertical like yours. It appears multiple vertical tanks aggressively strive for thermal stratification equilibrium. I've seen very powerful thermosyphoning in thse tall tanks.

 

[maple1]

Would it be advantageous to return the radiant flow to the bottom but return the coil flow to the middle?

 

[cpeltier]

I was lucky enough to get the 250 gallon propane tanks with the short dome. Unfortunately these tanks are in a room with an 8 foot ceiling so it is difficult to get anything larger then a 1/2" drain on the bottom port without jacking the top dome and port into the floor joist bays. However since they are short domes I may not be losing much water storage.

The pressurized storage will be pumped via a variable speed injection pump into a secondary loop with closely spaced Ts. A mod-con sits downstream piped the same way but with a fixed primary pump. The fan coils pull from the secondary loop with their own pump (Alpha in pressure mode, 3 zone valves), and the radiant heat variable speed injection pump feeds the radiant loop (Tekmar style) via closely spaced Ts. The radiant loop is primarily driven by a geothermal water-to-water heat pump. The heat pump can't track the upper part of the outdoor reset curve in the coldest weather (130 degrees) so the boiler loop kicks in. In warmer weather the control strategy disables the heat pump when it detects usable water in the storage tanks. When it's very cold out I don't bother since the Tarm 60 can barely keep up with the load and I can't feed it enough firewood without it becoming a full time job.

I'm not sure how a hydraulic separator would help me? They are expensive especially for a secondary loop of 11/2". I think my biggest problem will be keeping the mod-con from short cycling and returning water that is too hot when the fan coils are running.

I've thought about a mid tank return (when fan coil running) but is it optimal under all conditions? It's easy to over think how the tank return is handled. I could use temperature sensors and zone valves attached to my controller or maybe by using a Termovar AF bypass valve set at a carefully chosen opening temperature?

My Tarm has been down for a couple of weeks and I already miss it!

 

[ewdudley]

I'm not sure how a hydraulic separator would help me? They are expensive especially for a secondary loop of 11/2".

A hydraulic separator would be used instead of and not in addition to a secondary loop, and would therefore probably not make sense considering your system's state or development, despite its advantages.

 

[cpeltier]

One more question:

I have a side arm heat exchanger on a 110 gallon DHW preheat tank. It never caused a problem with my old unpressurized system but now with it's low delta T of 4 or 5 degrees pumped via a Taco 007 I'm afraid it will mix my pressurized storage. I could try slowing the flow down or downsizing the pump to get a better delta T across the heat exchanger. Another idea might be to return it into the boiler supply manifold at the top of the tank. Any thoughts on that?

 

[jebatty]

Depending on flow rates and times of use, your high temp fan coil returns to the bottom of the tanks may mix the tank quite quickly to a temp near that return temp. That will pose no problem for the radiant, as it likely can use that high temp fan coil return water just fine. I also agree not to over think or over design to squeeze maximum use of high temp water and prevent mixing. The only real world example I have is at DP, a 56000 sqft facility with fan coils designed for 180F water. The whole facility is heated very well with 130-160F water, the fan coils just are in operation more than otherwise as they operate on lower temp water. And a separate DHW storage tank which may get as low as 120F rarely results in insufficient hot water, although the kitchen uses a booster heater to get the very hot water it needs for the dishwashers.

 

[maple1]

One more question:

I have a side arm heat exchanger on a 110 gallon DHW preheat tank. It never caused a problem with my old unpressurized system but now with it's low delta T of 4 or 5 degrees pumped via a Taco 007 I'm afraid it will mix my pressurized storage. I could try slowing the flow down or downsizing the pump to get a better delta T across the heat exchanger. Another idea might be to return it into the boiler supply manifold at the top of the tank. Any thoughts on that?

I experienced that here. It wasn't much of an issue during heating season when things were kept relatively hot all the time, but when trying to heat just DHW from just storage, it really showed. I added a FPHX in and it works great (I just plumbed it in series with the sidearm). I also added in a small recirc pump (B&G E3Ecocirc) on the DHW side - a sidearm will work by convection if supply is hot enough, a FPHX not so much. I heat our 80 gallon electric hot water tank with it. The circuit breaker to that has been off for over a year now - but having said that, I might turn it on again for next summer depending on the situation when that time comes. I control all that DHW heating with a Johnson A419 - it can raise setpoints when I am burning, so in actuality in the 7 day period between burns in the summer, almost the first half of that is gotten through just from the charged up DHW tank. If I time things right with burns & wash days, that is. I also slowed the supply flow down through the HXs considerably - it is pumped by an Alpha, I have that flow throttled down to somewhere in the 1-2 gpm range.

EDIT: And still speaking of DHW heating, I also preheat my incoming feed for it via 2 rolls of 3/4" pex stuffed inside my storage enclosure, most of it just spread out on top of my horizontal tanks, under the insulation. There's a bit of a pressure drop at the taps comparing hot flow to cold flow, but it's still quite adequate.

 

[ewdudley]

I have a side arm heat exchanger on a 110 gallon DHW preheat tank. It never caused a problem with my old unpressurized system but now with it's low delta T of 4 or 5 degrees pumped via a Taco 007 I'm afraid it will mix my pressurized storage. I could try slowing the flow down or downsizing the pump to get a better delta T across the heat exchanger. Another idea might be to return it into the boiler supply manifold at the top of the tank. Any thoughts on that?

Study the stratified buffer/hydraulic separator cited above. With a DHW zone in the top of the buffer you can maintain a minimum DHW temperature with a sensor near the bottom of the DHW zone. The drawing shows heat exchangers in the tank; to supply heat to an indirect you would pull from the top of the buffer and return to the middle instead.

And then there is a separate zone in the bottom of the buffer that supplies the high temperature zones. The maximum return temperature of the bottom zone is controlled with a sensor near the bottom of the lower buffer tank zone.

Then lower temperature radiant loads pull 'spent' water from the bottom of the buffer as shown.

 

[Tennman]

I had the same height problem with 500 gal tanks in our barn's 10' ceiling. The bottom of the tank is ~2-3" off the floor (enough for some insulation). Our bottom manifold pipe comes in parallel with the floor and nearly tangent to the bottom of the dome. It requires cutting an ellipse in the side of the dome rather than a simple round hole, but gets you the bottom of the tank with minimum height. That gives the greatest deltaT to the boiler and uses all the tank's water for storing energy. I'll stay out of the buffer/separator discussion. Pretty sure biggest deltaT into/out of the boiler yields the greatest boiler efficiency which defines the optimal place to charge and pull from storage.

John Siegenthaler's Hydronics training document he's kindly offered to folks here is helpful.

http://www.nyserda.ny.gov/-/media/Files/EERP/Renewables/Biomass/biomass-hydronics-training.pdf

Cheers

 

[cpeltier]

I gutted my boiler room and installed the new Primary/Secondary loop with 2 parallel Mod Con boilers staged for a 10-1 turn down. I ran the house off of a geo water to water unit for almost 3 weeks as this work was being done. Luckily it hasn't been very cold in eastern Pennsylvania and we got by with minor annoyances.I winched the 3 - 250 gallon propane tanks into position, standing up. I have been studying the plumbing and insulation strategy's. Luckily I got the house running off of the Mod Cons just before Christmas and was able to take a break.Now it's time to address the wood boiler subsystem. The 15+ cord woodstack is not getting any smaller and it's now January!

I put 2" of foam board against my interior wall and am thinking that the tanks should be spray foamed (2 or 3"?) with the close fittings on but not fully piped together. After the foam hardens then I will move the tanks into their final position and pipe it up. I read John Siegenthaler's Hydronics training document (thanks Tennman) and it has a wealth of information. Of notable interest it appears that I should abandon the top and bottom of dome ports because of stratification destroying currents. There is no way to modify the tanks again without a huge amount of work so I'm guessing that it's better to abandon these ports and use the horizontal ports just below and above the dome for supply and return. I will lose the small amount of water in the domes unfortunately (I have the short dome tanks) but it appears to be an important trade-off.

Some of the information here seems very interesting and could be applied if I had more room and added a multi-function buffer tank in place of the secondary loop but I also think the load consideration is important. I have noticed that the radiant heat load is continuous and represents most of my load. The fan coils do occasionally come on and this increases with frequency as it gets colder outside, Some of the strategy's will work well if both demands are equally present but I think my strategy (outdoor reset PID injection pumps into secondary loop) will work based on the continuous demands of the outdoor reset, PWM modulated, IFRH. Besides some of this does not look easy to implement and requires custom fabrication. Another consideration is what happens if I want to sell my house. I could remove the wood boiler system and storage tanks without affecting the heating system.

 

[ewdudley]

Of notable interest it appears that I should abandon the top and bottom of dome ports because of stratification destroying currents. There is no way to modify the tanks again without a huge amount of work so I'm guessing that it's better to abandon these ports and use the horizontal ports just below and above the dome for supply and return. I will lose the small amount of water in the domes unfortunately (I have the short dome tanks) but it appears to be an important trade-off.

FWIW, I've come around to thinking that mixing that results from vertical ports is nothing to worry about at all, at least for vertical tanks. As a practical matter with vertical ports you'll still have well defined stratification, and whatever small mixed layer there may be will not affect the performance of the system significantly.

When the boiler is charging the tanks, the flow from the boiler will be on the order of 5 gpm, which is then split two or more ways, and so the feet per second is low. As the charging cycle starts there will be a small mixed layer at the top that then decends. As the charging cycle completes whatever mixed layer there is will end up on the bottom as return water and will therefore disappear.

When discharging, the flow to the loads is even less gpm than when charging, so there will only be a small amount of turbulent mixing at the bottom of the tank when the discharge cycle starts. Should the small mixed layer reach the top of the tanks before the next fire is lit it's no big deal because you're about to run out of hot water altogether and the fading temperature is the least of your worries.

For chilled water there the density gradients are tiny, a small amount of vertical jetting can be a big problem. But for hot water storage where the hot water is so much lighter than the spent water, and the feet per second entering the tank is quite low, it is important to realize that resulting currents will not affect stratification enough to worry about.

 

[cpeltier]

I have a partial roll of 1 1/4" oxygen barrier PEX and I am tempted to use it between the boiler and storage tanks and back to my Secondary Loop. It will save me about a 100' of 11/4" copper or black.
My storage tank ports are also not level with each other. The PEX would solve that without putting angled elbow/nipple combinations on every port. The PEX rating for heating tops out at 200F which should be safe. The inside diameter of PEX is much smaller then copper or black. Other then that are there any thoughts on doing this?

 

[Deering]

My understanding with multiple tanks is that if you plumb them in series you effectively replicate a single tall tank with superior stratification...almost. Boiler supply and distribution supply both connect to of the top of tank #3, with returns coming out of the bottom of tank #1. And the connections should definitely come in the sides rather than the top to help maintain stratification.

Siegenthaler addresses that issue as I recall from his class that I attended (highly, highly recommend!). He also has an interesting approach to DHW that I think you'll find from his training material. He has recently published a book covering this subject (which I have not acquired yet), and it might be worthwhile to read before you jump into this. Here's a link: http://www.amazon.com/Heating-Renew...8&qid=1420222151&sr=1-3&keywords=siegenthaler

 

[cpeltier]

The system has been running almost 4 weeks now.I plumbed the tanks in parallel following John Siegenthaler's Hydronics training document recommendations. Specifically I plumbed the wood boiler output to a larger manifold near the top (seam of the dome) of the 3-way connected 250 Gallon LP tanks. The system supply also connects to the same manifold. The returns are plumbed near the bottom in a similar fashion. The dome tops have vents and the bottoms only have drains. I run this primary loop from the tank to a variable speed injection pump to my Secondary loop. I have two staged LP boilers also feeding the secondary loop downstream of the storage injection. I slapped together a basic web user interface for the control system (EasyIO fg-32) which runs on the controller and there were few surprises!
The storage injection pump is a bit touchy using the Smartfan Nimbus controller in a PID loop configuration and I may have a small ghost/convection flow into the secondary loop from storage into the Secondary loop when idle and hot.
I'm running three fan coils with a combined flow rate of 8GPM at 125F and it heats the house to 70F when -5F outside without staging from the Propane boilers. I running the boiler twice a day when the outdoor air is below 20F. So far it's working well and I'm really impressed with the simplicity of the EasyIO DDC/SCDA controllers.