Tankzilla!

[DaveBP]

In the year or so I've been following this forum I've had the privilege of using the expertise and pioneering experiments of others to lead me to design my own storage system for my Tarm Solo based heating system. I had gone through various unpressurized pool and copper coils plans when I ran into a couple 500 gallon propane tanks and sailed off into various pressurized storage schemes. How to connect the two tanks and where to put them in an existing basement with only 7 1/2' ceiling height?

Well, some of the pioneering being done on this side of the Atlantic is old news a few thousand miles northeast of here. There are so many schemes to switch boiler output from heating loads to storage and back I began to wonder if any of that was necessary. Why not just directly couple the boiler to the storage tank (this would only be practical for a pressurized storage tank) and take all the heat loads from the tank directly. Seems it is common in Northern Europe.

Seems to me the weakest feature of a directly coupled storage tank is when the tank is depleted of hot water and it's cold out there and you just got home from a weekend away... How long will it take to get hot water out of that tank and start heating the house and get some hot water for a good long soak in the tub? Do you have to heat up the whole tank first? Hydraulic seperators do this quickly by being narrow vertical tanks with a short distance between the hot water in and the hot water out ports so the boiler supply water flows the short distance across and is drawn into the heating circuits by the pumps feeding the heating loads. How to do that with horizontal tanks 10 ft. long? And how to put 2 of these tanks together to get 1000 gallons?

Well, here is my latest scheme and I have the materials and started cutting and fitting. At some point you have to stop tweaking designs and start gettin' 'er done. I have use of world class welding facilities (and welder, not me) where I work so weld quality will not be an issue.

Boiler supply comes into the tank at the top left and goes back to the boiler bottom left. The hot water to the heating loads goes out top right and cooler return water comes back into the tank bottom right. My expectation is that once the boiler supply water starts flowing out of the boiler it will only take a few minutes before that hot water will float across and be available to the heating loads at the opposite end of the tank. There is only a few feet between the pipes in the tank and I have enough of that big 3" pipe to put them even closer if I get worried.

I would appreciate any input from you folks out there because there has to be at least a couple things I haven't fretted over yet and I haven't welded anything yet.
So let's see if my first attempt at photos works.

 

[Justburnit]

With both your inlet and outlet on the same end of the tank you may want to have one of them extend to the far end to prevent thermal short circuiting.

 

[sdrobertson]

What I did was take the boiler out line and put this into a manifold. This manifold would connect the line going to the house load and to the top of the storage and the boiler out. I made another manifold for the bottom of the storage which connected the return line from the house load and the storage and the boiler in. If the boiler was running and the house called for heat the water would run directly from the boiler to the house through this manifold. If the boiler was not running then the water would draw from the storage through the same manifold. I installed a motorized 2-way valve that would only allow water that had reached 185 degrees to leave the boiler. This water would either go to the house or go to the storage. If the boiler wasn't on, then the valve would be closed and the water would come from the tanks when the house needed heat. Another good thing about using manifolds like this is that you only need one hole on the top of the tanks for the hot water and one hole on the bottom for the cold as the water dosen't care which way it runs through the manifolds.

 

[Nofossil]

As far as the plumbing to storage goes, I think this is essentially equivalent to the 'simplest pressurized storage' design that I posted a while back. Consider the difference between the two diagrams below:

 

[jebatty]

I am doing essence of what your propose. The difference is that I heat storage only, and the storage is the radiant heating source for my shop. But based on the performance described, also drawing hot water from storage and returning to storage should work very well. The opposite end approach is also very good, as system return water to the tank may well by at a higher temp than the bottom of the tank, thus maintaining highest possible delta T between bottom of the tank return to boiler and boiler output. System supply water from the tank should be very close to boiler input to the tank due to the high stratification. You will want to pay attention to gpm's. At the flows described below, I experience very little mixing. Higher flows may produce a different result.

Short answer: under the operating conditions described, extreme stratification. Example: If entire tank was 120F, fire boiler, nearly the whole tank will rise to 160F, top to bottom, before the top of tank will start to rise much above 160F, and stratification continues as top of tank rises, gradually moving to bottom of tank, etc. You should be able to draw from top of tank and return to bottom of tank, using the hot top water. Why doesn't top of tank rise faster? With a delta T of 40, assume 6 gpm's, btuh is 120,000, pretty much max average output of the Tarm (I find 100,000 btu to be the reasonable average Tarm output; high burn easily reaches 140,000). I easily can heat the entire tank to 190F with very little idling until delta T closes less than 20, then idling begins occasional, and gradually increases. Final heating of storage at less than full burn (later part of the burn) works very well, as nearly full boiler output goes to storage with minimal idling.

1000 gal LP tank (39" x 19' approx). Hot water inlet is about 1/4 from left end, with a 90 degree fitting inside the tank that "shoots" the water horizontally towards the right end of the tank. Cold water outlet is from a diptube to the bottom of the tank and located a little to the left of the inlet. 1" lines to and from the boiler. Tarm Solo 40. I have temp sensors on the exterior of the tank and under insulation that measure temp top L and R and bottom L and R; also a thermometer in a well near the hot water inlet but opposite the direction of water inlet flow. Top L, R and thermometer all will read quite close; bottom L and R will read close.

Water flow into the tank ranges between 2.25 and 8.75 gpm depending on the temperature of return water to the boiler, which controls the opening of the Termovar, and depending on boiler output temp. I have 2 circs; a primary circ directly on boiler return (Grundfos UP-58 set on MED). This is controlled by a differential controller which turns "on" when boiler output is +16F over tank cold water return and turns "off" when this differential falls to +8F. I have a second circ (Taco 007) directly on the tank cold water return to boiler that is controlled by an aquastat that turns "on" when boiler output > 180F. The boost in flow at high boiler output does a lot to eliminate or greatly reduce idling

Good luck. Good plan, IMO.

 

[DaveBP]

Back home now. Thanks for the replies.

This virtual model is of the tank assembly only. I didn't bother to show all the other necessary hardware, isolation valves, expansion tanks, etc. And if there is anything new here it is only in how the two propane tanks are connected together, not in how there are integrated into the rest of the system.

The possibility of thermal 'short circuiting' is probably important in some tanks with big circulators pushing water in and out at high FPMs. I'm using 3" pipe for those long tubes you see in the section view. Even at high speed the Grundfos 1558s will only be pushing water in and out at inches/second not several feet/second like some tanks get. After watching Jebatty's tests this spring and seeing how well the hot water in his long horizontal tank stays horizontal with much higher flows than I will use, I quit fretting over it. I'm also counting on the projecting lips on the end of the pipes to help direct the water across the tank rather than try to go straight down.

Nofossil, your diagram on the right, the one with the 2 edits to the sticky diagram, is closer than you know to what I am going to do. Swap out the baseboards for radiant floors fed by a Tekmar controlled mixing valve and I could just use your diagram as it is. I'm attempting to find a simpler than simplest way to configure the boiler at the cost of much more tricky welding and fittings on the tank. The Froling and other computer controlled boilers are designed to adapt to different loads and the varying water flows that accompany them and maintain an efficient burn. Seems to me that the manual gassifiers can be tuned to burn best under a set of conditions that are best not varied a lot. So tuning for only one task, heating the tank and only the tank, seems simplest to me. But, again, at the cost of a more complicated tank and its plumbing fittings. Most people are going to want to just use whatever fittings are on their propane tank as is. And that can work.

Jim, I may use the tank-as-radiator scheme to heat my basement/workshop, too. I'm thinking along the lines of enclosed and insulated with ducts in and out around the bottom tank with a small computer type fan and thermostat. It's a well insulated, dry space.

My tank design is probably overly elaborate but it has been a "fun project" for me and a good excuse to learn how to do simple assemblies in Solidworks. It would be a daunting, expensive solution for anyone that didn't have top notch welding at hand and racks of all manner of steel stock to choose from.
My boiler will have a very low head circuit through the tank so I'm hoping to do it with little circulator power. I can see I may be heading toward some kind of variable speed, delta T control on it but I'll start out at the brain-dead simple level and work my way up as I experiment and learn. Don't want to deprive myself of all the adventure that you guys that have been running your systems for a season or more have been sharing with us all. And it will have a number of bypasses so I can try different configurations to compare and contrast.

Thanks again for your comments, Dave

 

[jebatty]

Jim, I may use the tank-as-radiator scheme to heat my basement/workshop, too. I’m thinking along the lines of enclosed and insulated with ducts in and out around the bottom tank with a small computer type fan and thermostat. It’s a well insulated, dry space.

My tank is wrapped with 6" of fiberglass insulation. I leave about 4-6' exposed, depending on outside weather, and I end up getting the heat I need without any additional fans, power, etc. Your'e headed in the right direction.