Buffer tank - need advice

152NP

I am hoping someone with experience can give me guidance on buffer tank selection.

I am installing a Lochinvar Noble 199 combi, which has 185MBH heating capacity, and a 10:1 turndown ratio. The distribution system is home run pex with TRVs on each radiator. Most radiators are cast iron, but, as things evolve, I might have a mix of emitters. Lowest individual load is about 1,000 btu. Using Lochinvar’s buffer tank sizer suggests that their 30-gal tank would probably be more than adequate.

Lochinvar has two models of tanks for my application: stainless steel with 1.5” connections and 5-year warranty, and glass lined with 2” connections and 3-year warranty.

I am concerned that the 1.5” connections might cause too much velocity and compromise stratification when a boiler this size is running full out with a 4-pipe tank configuration.

The supplier I am purchasing from only offers the glass lined in 50-gal capacity. Even at that size, it is still significantly cheaper than the 30-gal stainless. I am not sure if you can oversize a buffer tank.

The 2” connections would give me the option of doing a 2-pipe configuration whereas the 1.5”, I think, would not give me the hydronic separation at the common piping T.

What are your thoughts? I don’t want to regret buying the wrong tank.

Thank you for any advice!

hot_rod

I've played around with buffer tank piping methods for years and have some thoughts.

The hydraulic separation function happens outside the tank at that large pipe size where all piping combines.
At max flow it would be nice to see velocity below 2 fps in that "large" pipe. So pick the ∆ you plan on running the boiler at and look at the chart for gpm at that rate. Is this a combi? The chart shows 10 gpm at 34∆ on the 199. Smaller models go down to 19 and 26∆ for some reason.

So call it 11 gpm.
1-1/2" copper you could flow 11.4 gpm at 2 fps velocity, just fine. The "fat" header would want to be 1-1/2"

What happens is when the tank ports are small or reduced you get a small velocity increase in that very short section. That is what can mess with stratification, adding some turbulance at that connection. Something like reducing from 1-1/2" to 3/4 might be pushing it. Regardless you are fine with 1-1/2" ports.

As far as sizing the tank, it comes down to your desired on/ off cycle rates. I've used 20 minutes for my calcs. So here is the formula with the time factor included. Poke your numbers into it.

The last 3 buffers I put in were 2 pipes. All things considered for what you are attempting to accomplish I feel it is the best use of the tank. Idronics 10 goes into the pros and cons, 2 to 4 port option.

3 pipe adds yet another function, we discovered it after writing Idronics 10 .
Great for heat pumps and mod cons as it brings the coldest possible return direct to the boiler, no interaction or blending in the tank, a small efficiency advantage, critical for maximizing HP efficiencies.

If the goal is to maximize a buffer, I'd go with a 50 gallon. Sounds like a cost savings also?

Does the heat load require the 199? Or is it the DHW capacity dictating the choice?

152NP

@hot_rod Thank you for the prompt reply. I have been watching your videos about buffer tanks, which is what gave me enough knowledge to know that there were some subtleties I needed expert advice on.

Yes, this is a combi. For what it is worth, I am doing my boiler piping in Schedule 40 black steel. I already have the threaders, so it is the cheaper option for me.

I am totally fine with doing the 3-pipe configuration if that is what will work best. Everything I saw about 3-pipe mentioned it in context of heat pumps, which is why I ignored it.

The 199 is for the heat load. I did a heat loss calculation on the home, a 1890’s Victorian with 3,000 SF of living space. I don’t need it all now, but, when I put some radiation in my 1,500 SF basement workshop, I will.

I have been running a 130k BTU coal fired boiler up until now with a boiler loop, system loop with a pressure differential pump on a home run manifold, and an injection mixing pump between the two with outdoor reset controller. I will be reusing the pressure differential pump on this new system. Bought a new Caleffi manifold to use because the existing cheap stainless unit leaks badly. I also have an indirect DHW tank that I had to stop using because it took too long for the heating system to recover after showers. I have been using a dedicated gas hot water heater since. I will probably use the combi’s built in DHW to a dedicated 2nd bathroom and leave the gas unit alone. I thought about hooking the indirect up to the new boiler but would have to figure out if it is worth doing, and a sensible piping scheme, because the combi has a built in circulator that makes it impossible to pipe an indirect circulator in parallel. Maybe close Ts on the boiler loop would work, but, again, not sure what I would gain. The combi unit was actually cheaper than just a plain boiler unit, and my local utility is giving $500 more rebate on a combi, which is why I went that route.

Again, thank you for your valuable advice! Any further comments are welcomed.

hot_rod

Go with a two pipe. 3 pipe makes it a bit trickier to keep the tank loaded after a heat call. A tank with a large top port would be best also. The hottest water is up there and it maximizes the tank capacity to pull from there. Those top port tanks are not so common.

A sensor well mid height on the tank helps wire a control signal.

You could circulate DHW from the combi to the old indirect, just use it as a storage tank, if you want that dump capacity. A small stainless circ between the DHW side of the combi plate HX and the tank itself, skip the coil.

The main downside to a combi is DHW start up. If you keep the boiler cold, you need the warm up time. Could be a minute or more. Winter months it is not an issue.
Keeping the boiler warm would cost a bit more energy, but speed up DHW to 30 seconds or less on the 3 combis that I have owned.

mattmia2

The glass lined tank will be as durable as the stainless tank as long as your system is leak free because there will be little oxygen to corrode it. Since it will be kept hot all the time it being well insulated is important.

An electric water heater can also be used as a buffer tank although you have to plan for smaller ports and there is no well for an aquastat, it would have to be on the surface like the thermostat.

hot_rod

mattmia2 said:

The glass lined tank will be as durable as the stainless tank as long as your system is leak free because there will be little oxygen to corrode it. Since it will be kept hot all the time it being well insulated is important.

An electric water heater can also be used as a buffer tank although you have to plan for smaller ports and there is no well for an aquastat, it would have to be on the surface like the thermostat.

In theory a plain steel tank would work. The reverse indirect tanks are plain un-lined steel I believe.
I used a couple Harbor Freight portable air tanks for buffer projects at home

Here is how a side port tank could be configured for a 3 pipe tank. Super simple piping. Thanks Siggy for the drawing. Depending on where the side ports are on the top of the tank, you have a bit of the tank not in the flow. A diffuser tube could be used to force fluid into the dead spots. You can find dielectric nipples with a plastic gooseneck to get those spaces.

152NP

AO Smith makes an 80-gallon commercial storage (TJ-80S) tank with 2" tappings on the sides and top, and it seems to be within the price range I am willing to spend, so it sounds like it would work well. My question is, can you oversize a buffer tank? I started with a target of 30-gal, was advised to go to 50-gal, and now I am thinking about an 80-gal to get the best tapping configuration. Would I be making a mistake?

152NP

@hot_rod a follow-up question. You suggested that I go with a 2-pipe buffer because a "3 pipe makes it a bit trickier to keep the tank loaded after a heat call." Then you followed up with another post showing how a side port tank could be set up as a 3-pipe buffer. Just so I am clear, which to you think is best for my application, a 2-pipe or 3-pipe? I am in no position to make that judgment myself. I would even do the 4-pipe if that is the path of lease resistance and doesn't sacrifice much performance over the other configurations. I just want to get this functioning without too much tinkering after the install if possible. Something idiot proof given who you are dealing with here.

I appreciate and value everyone's advice on this!

Hot_water_fan

There’s not much of a downside to oversizing. Maybe you’ll have higher standby losses, but to wildly guess a dollar amount, maybe $5/year? In exchange you get long, efficient cycles. 

152NP

@Hot_water_fan that is what I was thinking, as long as I keep it within the realm of reason. Thank you for your reassurance.

hot_rod

So, to make a long story long...:)

I have not personally done a 3 pipe. I like to sit and watch a system work to confirm I get the expected result. The logic and use by the HP industry indicates it is the best method for their needs.

Correct that 3 pipe is more prone to show up on HP systems. Most are single speed, non modulating so they must have some buffer, and the manufacturers usually suggest the preferred gallons, based on HP tons. They also keep efficiency higher with that return path.

Here are some cleaner pics. Actually from a You Tube training Siggy did for Eden Energy. Maybe watch that for a bit more info than I can type.

The main advantage of 3 pipe over two is the tank is always engaged or part of the picture. With a two pipe, depending on flows it may not be. BUT if a mod con boiler is modulating exactly to the load, any sized load, really no need for the buffer to be engaged. It only job then is to extend on/off cycling. So the fixed input of a HP is what dictates the 3 pipe as the better. For the mod con, it's a toss up.
Most critical on 2 or 3 is the control. The tank MUST have a sensor running the boiler. You would not want a room stat to turn off the boiler, as the tank may not be loaded when the heat call drops. The tank just supersizes the boiler water content at that point.

With a two pipe we did learn of some ghost flow potential, so a pressure bypass or a zone valve should be added. I like the PAB as it does not require electricity. If the boiler reaches setpoint, controlled by the tank sensor and the heat call is still active, in some cases ghost flow through an unfired boiler is possible. Which sheds away a small amount of heat. I've not seen it, but others have. A simple piping component fix.

With a 3 pipe, no need for that extra component, you do want a check however to prevent reverse flow on a 3 pipe.

Just to keep all options available, the two concerns with a 4 port is, in some instances the tank temperature could drop, maybe a day or two of no heat call. So getting heat to the load will be delayed until the tank is warmed and back on line. 2nd, if the boiler has a high flow requirement, it can break up your stratification as flow crosses the top and bottom of the tank. It also requires a speciality tank with 4 large ports.

So considering all this and with proper control of the tank temperature 2 or 3 both work, with the 2 port wanting an extra valve to assure no ghostbusting required

I like differential controls for the boiler operation, you get two sensors, two delta T and two differentials to really dial in temperature control. It is mentioned briefly in the video. We show and explain differential controlled buffers in Idronics 10.

Some mod cons come with a system sensor in the parts bag, wire it to the boiler board and stick it in the tank. The boiler control has some basis parameters to adjust. Nice about that is it shows up on the boiler control screen for setup and troubleshooting, one less component to buy and learn.

Heatloss, physical size and cost are the only reasons not to go big. I suppose you could add an extra heat blanket from HD to get another R-5 around the tank. As always, if it is within the heated envelop, it it really heat loss??

Here is the link to the Eden Energy HP training, somewhere around 50 minutes is where Siggy explains the pros and cons of the three tank piping options.

https://www.youtube.com/watch?v=1DMC0w5GscA

152NP

@hot_rod this is all very helpful stuff. The 3-pipe just makes the most sense to me. It is intuitively easy to understand how it should work. The only thing that gives me concern about it is your earlier comment “Go with a two pipe. 3 pipe makes it a bit trickier to keep the tank loaded after a heat call.” Why would that be? I couldn’t find where Siegenthaler had anything to say about that.
If the only reservation is that there are not a lot of 3-pipe buffers out there yet, I think I am okay with that and willing to give it a try. Worst that can happen is I would have to re-pipe a little to convert to 2-pipe if it didn’t work out. That is the nice thing about working with steel pipe; fittings are reusable.

The Noble combi does come with a system supply temperature sensor that I can use to monitor the tank and control things.

I am still obsessing about the tank itself. I thought I was settled on the AO Smith TJ-80S but realized that it does not have a temperature sensor well mid-way up. The TJ-80A (ASME version) does, but it is silly expensive. The Lochinvar 50-gal tank does have the well where it needs to be, but only has side ports, not the top port that you said would be best for a 2 or 3-pipe. It might be a compromise I have to make. Maybe I can use one of the higher ¾ tappings on the TJ-80S and just set the temperature target a little higher to compensate for its position.



Something you said earlier in this thread still has me confused:
“The hydraulic separation function happens outside the tank at that large pipe size where all piping combines. At max flow it would be nice to see velocity below 2 fps in that "large" pipe. So pick the ∆ you plan on running the boiler at and look at the chart for gpm at that rate. Is this a combi? The chart shows 10 gpm at 34∆ on the 199. Smaller models go down to 19 and 26∆ for some reason.”
What do you mean by pick the delta I “plan” on using? This boiler comes with a built in pump, so the only control I would have over the ∆T would be based on flow rate, which would be dictated by the piping resistance, correct?

Your suggestion about running the output of the boiler’s DHW plate exchanger into my indirect tank’s storage is simple genius! Maybe that should have been obvious, but, sometimes the obvious eludes me.

Thank you for being so generous with your wisdom.

EBEBRATT-Ed

I have used plain steel tanks (ASME) made for air compressors when buffer tanks were not readily available.

The tapping's were not in the best locations but with some double tapped bushings and dip tubes it worked

Dave Carpentier

Im opting to not use a combi Lochinvar in my design, because (iirc) any time there's a heat call to the T-stat connection the internal circulator runs. The non-combi ones (or at least the Lochinvar Epic non-combi) use an external "boiler" circ which can be configured to only run when the System Sensor (mounted to the buffer tank) indicates that heat is needed in the buffer.
With 8 varying loads, I didnt want the boiler circ (in a combi setup) to wash out the "exergy" in my buffer during low flow on the zone heating side of the buffer.

But.. I gotta admit, a $500 discount by going with a combi is appealing too.

152NP

@EBEBRATT-Ed now you have me thinking. I just looked and steel flanged weld on tank fittings are cheap. I have a decent MIG welder and could convert an air tank easily. The only issue would be insulating the thing properly without making an ugly mess.

hot_rod

152NP said:

@hot_rod this is all very helpful stuff. The 3-pipe just makes the most sense to me. It is intuitively easy to understand how it should work. The only thing that gives me concern about it is your earlier comment “Go with a two pipe. 3 pipe makes it a bit trickier to keep the tank loaded after a heat call.” Why would that be? I couldn’t find where Siegenthaler had anything to say about that.
If the only reservation is that there are not a lot of 3-pipe buffers out there yet, I think I am okay with that and willing to give it a try. Worst that can happen is I would have to re-pipe a little to convert to 2-pipe if it didn’t work out. That is the nice thing about working with steel pipe; fittings are reusable.

The Noble combi does come with a system supply temperature sensor that I can use to monitor the tank and control things.

I am still obsessing about the tank itself. I thought I was settled on the AO Smith TJ-80S but realized that it does not have a temperature sensor well mid-way up. The TJ-80A (ASME version) does, but it is silly expensive. The Lochinvar 50-gal tank does have the well where it needs to be, but only has side ports, not the top port that you said would be best for a 2 or 3-pipe. It might be a compromise I have to make. Maybe I can use one of the higher ¾ tappings on the TJ-80S and just set the temperature target a little higher to compensate for its position.



Something you said earlier in this thread still has me confused:
“The hydraulic separation function happens outside the tank at that large pipe size where all piping combines. At max flow it would be nice to see velocity below 2 fps in that "large" pipe. So pick the ∆ you plan on running the boiler at and look at the chart for gpm at that rate. Is this a combi? The chart shows 10 gpm at 34∆ on the 199. Smaller models go down to 19 and 26∆ for some reason.”
What do you mean by pick the delta I “plan” on using? This boiler comes with a built in pump, so the only control I would have over the ∆T would be based on flow rate, which would be dictated by the piping resistance, correct?

Your suggestion about running the output of the boiler’s DHW plate exchanger into my indirect tank’s storage is simple genius! Maybe that should have been obvious, but, sometimes the obvious eludes me.

Thank you for being so generous with your wisdom.

I probably wasn't clear on a few points. Yes the tank is the separation device in all 3 cases. The point with two pipe is that large pipe where the load and heat input connections come together. That is a separator point of sorts as that "bulge" needs to be handle flows from all points without causing unwanted flow. But as we discovered, sometimes that check or PAB is still needed. Notice point "A" on the 3 pipe using the side port, needs to be a large pipe there.

The absolute perfect 3 pipe tank IMO would use that top port. I actually modified a Discal to have a 2" port on the bottom for tank top
The dome of the tank helps air sep, especially when no flow. You better use the entire tank also, compared to side port. It cleans up the piping "look" a bit.

As @EBEBRATT-Ed mentioned any pressurized tank can work, I have even used old expansion tanks with the diaphragm pulled out, welded ports in. 4 different LP tanks from 160- 500 gallons. Some of the drainback tanks I built and the 500 gallon with diffuser tube.
Wessel, pic below, and other still sell plain steel and galvanized tanks. But the insulation and jacketing is not so easy to get done and looking nice. If you have a commercial insulation supplier nearby, they sell those vinyl covered fiberglass sheets or rolls. I always scrounged it at metal building sites, they use it for ceiling and wall insulation often.

Pro and con with sensor wells. They do add "lag" to response on both ends, the thicker the well the more the lag. Sensors must be tight inside and have heat transfer grease. Sensors do get baked into wells occasionally. Wells do allow the sensor to be deeper in the tank.

To be honest I feel strap on sensors actually work fine, better in some cases. Electric water heaters and many hydronic appliances use SnapDisc strap on thermostats. tekmar controls use mainly strap on sensors. Get good contact and insulate over it.
Here is a trick for getting better surface area with round sensors. I've used this trick for years to get solar sensors strapped to the collector piping. Older solar tanks had a stud welded on them to put this type of sensor on them. The box stores sell these electrical "bugs". Many sensors now are 6mm, about 1/4".

Lastly the control logic is important with all tanks. It comes down to how complex to get the job done. A single sensor on the tank firing the boiler is fine. More advanced controls have a top and bottom sensors. The sensor at the bottom quickly responds to cold return and starts firing almost immediately. With a mid point only, you need to work through the stratification until the sensor sees cold, so boiler firing is delayed a bit. And the top sensor always knows the hottest condition in the tank. With wood boiler buffers, boiler fires from the bottom sensors, loads watch the top sensor. I think Siggy talks about dual sensored tanks in that video?


If you want that more complex control, instead of just a boiler sensor, find a Caleffi I-Solar Plus control it is a great option, two 120 outputs, 4 sensors, data logging, digital read out, BTU metering and more. It's out there as a Resol BS Plus also. Solarnetix up in Canada was an importer of Resol, Joe at Alpha Thermal back in MA also. Maybe e-bay. Complex controls tend to overwhelm some folks. Knowing low, high voltage, sensors, NO NC, differential and all the other terminology is frustrating for DIYers without wiring and control experience. Start with a simple boiler or system sensor to fire it up, maybe.

EBEBRATT-Ed

You can insulate the tank with "duct wrap". It comes vinyl coated in white or gray or foil coated. But it's sold in I think 200sq ft rolls 4' x 50' which is more than you need. Maybe you can find an HVAC or duct shop that will sell you a little

152NP

@hot_rod I like the weld on stud with the bug for sensor attachment. That would still have response lag time, would it not?

I am leaning toward using a compressed air receiver tank unless I can find something better. Well pressure tanks caught my attention because they are cheap, but after I saw how they are put together, it seems like a big hassle to tear out all the innards. I haven't found where I can buy a plain tank from Wessel yet.

I found McMaster-Carr has butt weld fittings that look like they would be easy to work with:
Is there a type of fitting you found is best to work with?

As far as controls, I am not intimidated by them. My electrical engineer friend taught me the basics and I wound up creating a monster for my coal fired boiler system (see attachment). What is the logic that you are suggesting needs to be created (or I can watch Siggy's video to see if I can figure it out)? I can probably do it with the parts I already have. The question is, will it result in a noticeable payoff? I did find the Resol solar controller on Ebay for cheap. If I can do the logic with what I have, the only thing more the Resol would give me is logging.

@EBEBRATT-Ed concerning insulating the tank, I was thinking maybe a good way to go would be to build a small wooden enclosure around it that could easily be lined with rigid insulation board. @hot_rod If I did this, would an external tank thermometer get confused by the elevated temperature within the enclosure?

It is surprising it is so difficult to do a system right!!

Larry Weingarten

Hi another thought on insulation is to use relatively thin rigid sheets of foam like 1/2" R-Tech found at Home Depot. They can bend pretty well depending on the radius. I've used multiple layers of 1" foam, held in place with package strapping to insulate a large tank successfully.

Yours, Larry

hot_rod

152NP said:

@hot_rod I like the weld on stud with the bug for sensor attachment. That would still have response lag time, would it not?

I am leaning toward using a compressed air receiver tank unless I can find something better. Well pressure tanks caught my attention because they are cheap, but after I saw how they are put together, it seems like a big hassle to tear out all the innards. I haven't found where I can buy a plain tank from Wessel yet.

I found McMaster-Carr has butt weld fittings that look like they would be easy to work with:
Is there a type of fitting you found is best to work with?

As far as controls, I am not intimidated by them. My electrical engineer friend taught me the basics and I wound up creating a monster for my coal fired boiler system (see attachment). What is the logic that you are suggesting needs to be created (or I can watch Siggy's video to see if I can figure it out)? I can probably do it with the parts I already have. The question is, will it result in a noticeable payoff? I did find the Resol solar controller on Ebay for cheap. If I can do the logic with what I have, the only thing more the Resol would give me is logging.

@EBEBRATT-Ed concerning insulating the tank, I was thinking maybe a good way to go would be to build a small wooden enclosure around it that could easily be lined with rigid insulation board. @hot_rod If I did this, would an external tank thermometer get confused by the elevated temperature within the enclosure?

It is surprising it is so difficult to do a system right!!

My life long quest, finding inexpensive pressurized tanks. When yo find those compression tanks, Wessel etc they are $$ and you still need insulation, maybe a jacket. Over 160 gallon, by far used LP tanks, can be bought for scrap metal prices. They ran about a buck a gallon in Missouri where they were plentiful. LP yards full of them in Texas also. Dealing with the odorant isn't fun

Compressor tanks are inexpensive until you get above 20 gallons for example Harbor Freight has them for under 200 bucks.

What I have found is an electric water heater checks most of the boxed. Insulated, jacketed 300 psi test, tappings for gauges air vents etc. Check all the box stores around you, scratch and dents are usually sold 1/2 off the marked price. Same at the wholesalers, they get "forked" tanks from time to time.

The element holes are 1", so get a 1 X 2" or 1 X 1-1/2 swage nipple like this and increase right at the tank, as close a possible. Graingers has them but $$. Google swage nipples to find a good price. If you want to really maximize, weld a curved tube to get right to the top and bottom. Mark the nipple so you know where to stop when you thread it in. A bit overkill maybe, but..
I think up to 10 gpm or so, that 1" reduction at the tank doesn't upset flow too much.

For sensors, I have also taken a hole saw, cut a 4" hole in the jacket and welded a 1/2 coupling like you show onto DHW tanks. Just buy a steel coupling and hacksaw it in 1/2.

It compromises the glass lining of the tank no doubt, but?? If you are good with a wire feed it is a 1/2 hour job. Protect the insulation from burning, I use Hot Dam a clay product, and pack it around the insulation.

I've also built all sorts of wells for sensors, from 1- 4 sensors in one well. K&S Engineering sells this thin brass tube online in all sorts of sizes, a hobby product. A thin brass or copper well will be you best and quickest transfer, or gold if you have the budget.

Thing is, with an external strap on, it is easy to relocate. Really jamming the sensor between the spray foam lining and the tank works fine. A sensor well with a thin tube is probably about the same response time. We are talking seconds, not hours. Sensors have lag also depending on the spec. These wells are for m6 mm sensors, like Resol and any Euro based control commonly use.

The Ranco controls in that box are setpoint, they can work, but the only measure at one point.
The Resols are also "wave choppers" so you can run any PSC circulator on a variable speed function.
I have a cheat sheet that we built for the Resol that explains what the various arrangements do., if you go that route.

This 850 page book may be available in the bookstore here at HH, if not at hydronicpros.com. It is the bible for building systems like yours. A good education on hydronics in general. A handful of my ideas show up in the pages! The sequence of operation holds the key. A clever person can find and assemble the controls once you have a handle on what happens when and why. Write it down so you refer to the goal as you build and test.

I was surprised and honored to have a mention in this book

152NP

@Larry Weingarten thank you for the suggestion on insulation. That is a good idea! I also have some 2-part expansion foam left from when I insulated some walls in my house that I could fill an enclosure around a tank with. Your idea sounds a lot less intense.

@hot_rod thank you for your further comments. All of this information is very helpful!! I have Siegenthaler's Modern Hydronic Heating, 2nd and 3rd additions, but they don't talk about buffer tanks that much. I just ordered Heating With Renewable Energy and will see if that answers many of my questions. I will try not to bug you until I do my homework. When I understand why I might need a variable speed pump somewhere in the system, I will hit you up for the Ranco cheat sheet. FYI, I do have a Grundfos Alpha2 from my current system for my manifold.

I will have to make a choice on a tank after exploring my options. A trip to the scrap yard is in my future just to make sure I don't miss a gem. You did address my concerns about an electric water heater. I wondered if cutting and welding a glass lined tank would cause it to perpetually delaminate and shed debris into the system. I guess not. I even thought of looking for scrap 20" diameter steel pipe and having two thick steel discs cut that I can weld on the ends. Bottom line is, I don't want to do all this work on a tank unless it is going to last a good long time. It would suck to have a highly customized, labor intensive component be the weak link that fails in the dead of winter.

I am going to do this, one way or another.

As far as you being mentioned in Siegenthaler's book, I am not surprised!

MikeAmann

I like your project. I did a little searching for you:

https://www.douglastank.com/6.2-cu-ft-compressed-gas-cylinder
https://www.ebay.com/itm/324808985332?hash=item4ba01fcaf4:g:1r8AAOSw9URhji58
https://www.ebay.com/itm/165115223049?hash=item2671a25c09:g:UQAAAOSwSDJhX0Bd


IBC also makes a 55 gal buffer tank with 1-1/2" inlet & outlet:
https://www.ibcboiler.com/support/TIPortal/Content/A_Homepage/Series/Buffer_tanks/Models/IBT-55.htm

hot_rod

Buffer on wheels, what a concept. Doubles as a convex mirror to look at yourself in Takes me back to the county fair days.

152NP

@MikeAmann thank you for doing that research! The universe was kind to me. I monitored Craigslist and was very lucky to get a 60-gallon tank for $50!! It was an air compressor that threw a connecting rod. I scrapped the pump and cut off the pump mount.


@hot_rod if you will remember, this project is a Lochinvar 199 combi with home run distribution and TRVs. Mostly cast iron radiators running at 190-deg with a 20-deg drop. I have been reading Heating With Renewable Energy as you suggested and am slowly understanding things.

I am guessing the way this thing should operate is:

• buffer tank temp is managed with a sensor and thermostat with a differential and its own outdoor reset.
  
• buffer tank temp controls the heating call to boiler
  
• boiler's supply and return sensors, and boiler's built in outdoor reset, control boiler's supply temp and firing rate.
  
• distribution system is supplied by a constant pressure circulator always on
  

First order of business is I have to get this tank built. We were talking about doing a 3-pipe configuration. I was going to weld on 2" connections, one to the center of the top, and 2 on opposite sides of the bottom. Sensor wells should be where? One at the middle of the tank and one at the bottom? I am not sure we need one at the bottom. The page from Heating With Renewable Energy that you included with your previous post depicts a system for a solar application. Siggy mentions that that control setup is only justifiable for tanks 500-gal and up. Would one middle tank sensor running on a differential work well on my 60-gallon tank?

I noticed that tank welding spuds are readily available in stainless steel, and I found where there is a special MIG wire you can use to weld them to regular tank steel. Do you have any experience with this? Is it a straight forward procedure?

mattmia2

If the tank is steel there is no point in using stainless spuds.

152NP

@mattmia2 that is kind of what I thought. Just wanted educated opinions to prevent me from doing something wrong.

hot_rod

You might just get a couple 2" street 45s and use the port near the top that is already there. Just rotate them until you get a level or plumb connection.
Steel 1/2 couplings are fine, keep everything 2", that gives you options to add into the system in the future. Use steel couplings not cast iron if you buy couplings. Or buy those weld O lets

You have a few choices on how to run the buffer tank. With a mod con, if you want to keep it condensing run the tank up to 130F, that way returns should be low enough to keep condensing.

Or if you run the tank to 150- 160 or hotter. This gets you more useable time out of the tank, but loose boiler efficiency, you have an 87% boiler essentially. High temperature tanks make sense with solar or wood/ pellet fired as we want long run times on the boiler, hot run times. With solar we pack as much temperature as we can get, into the tank.

I think you have cast radiators? Did you calculate what the lowest SWT you can run to them. That would help decide how best to run tank temperature. You could run the tank at a constant temperature and pull the loads off with an outdoor reset control. This helps maximize the tank, supplies close to ideal temperature and may even get you to constant circulation on the distribution. I think you mentioned you have some ODR parts from the previous system?

Some Lochinvar boilers come with an additional supply sensor. It will be in that plastic bag of accessories. Use that in a well about mid point on the tank.

You are correct in that the tank sensor is the only call to the boiler. I think you have radiators with TRVs?
So you could just wire the delta P circ to an outlet, it will start and modulate as TRVs open and modulate. That was the original intent of delta P circs since TRVs are the most common control throughout Europe and the rest of the world
When they first came to the US the Alphas all had cords with a plug. Not understanding the logic, installers would cut the plug off and wire them to a relay box, turn them on and off.


You might install on master thermostat to start the entire system up also. No sense in maintaining a hot buffer all summer.

If you run the boiler on ODR your tank will not be as warm as a constant temperature tank, so the drawdown capacity is less because the tank is maintained based on outdoor temperature. But it would be super simple to have the buffer controlled by the boilers ODR, I'm not sure how much the buffer would "buffer" that way. Easy enough to try it with or without. It may take an entire heating season with all the load swings to see if you are getting the results you want. I think you main goal is to lessen or eliminate boiler cycling?

You are headed to what I consider an ideal system. Modulating SWT temperature, modulating pump output and TRVs modulating flow based on individual room needs. Dialed in properly it may start up on the first heating day and modulate around load all season long. The cast radiators for a nice feeling radiant emitter. They will not be hot to the touch often, maybe ever, but they will cover the loads if you have everything sized correctly. I'll think some more on nit, see if we have a system like that in a past Idronics showing wiring options.

152NP

@hot_rod to be honest, I would be more comfortable keeping the buffer tank at a constant temperature and modulating the temperature of the water delivered to the distribution system through an outdoor reset. That is essentially how I have been operating my current coal fired system. The boiler loop is always running and I control the temperature of the constant pressure system loop with an injection pump between the two controlled by a Tekmar 361 mixing controller. I plan on building an enclosure around the tank lined with 4" of rigid insulation board achieving R-26, so standby loss should not be much.

hot_rod

Boon indicated that he had to put the boilers sensor in the tank, not the system sensor. Check and see what you have me, some of the sensors are now  now thread in type.

152NP

@hot_rod here is what is in the system sensor bag:

I am not sure what the aluminum tube is. My best guess is that it's for strapping the sensor to a pipe if you don't use the well. The installation manual says nothing about it.

How the boiler decides how to modulate is a mystery to me, and, I don't understand what effect having the system sensor in the tank vs in the tank supply piping would have.

If we were to run the buffer tank at a constant temp, then we would have to circumvent the boiler's built in outdoor reset functionality. The best I figure on how to do that from reading the manual is to NOT install the outdoor temperature sensor.

If I am wrong about wanting to run the tank at a constant temp, please let me know. I am just trying to find the midground where the system works correctly with as few variables as possible. I can imagine where running multiple outdoor resets could put the system into a situation where components are fighting each other.

Thank you for continuing to help me get this thing set up correctly.

hot_rod

I'm pretty sure if the outdoor sensor is not connected on the terminal board it will not use that parameter.
The aluminum tube is a a spacer for putting that sensor into a larger well. Older style aquastats had 1/2 diameter wells.

I believe that sensor can also be used to run an indirect from the boiler in stead of its own aquastat. It's a common resistance sensor and whatever connection you connect it to tells the control what it is reading.

I don't know either how the supply sensor functions differ from what the boiler sensor tells the boiler to do. I've used supply sensors downstream of hydraulic separators and they seem to work fine.

I vote for constant temperature also, mix to the loads. Let me talk to some Loch control pros next week, see if I am missing something.

152NP

@hot_rod thank you again. I look forward to hearing what your Lochinvar guys have to say about it. In the meantime, I will get to work on that tank.

Dave Carpentier

If the Noble is like the Epic , the "System sensor" takes over from the normally used outlet sensor.
So if your ODR value causes the boiler to 'want' 150f, then when it runs it will seek to get to 150f at that sensor (buffer tank). Different outdoor temperatures will vary the setpoint, and thus your buffer tank temperature.
Lochinvar has the installation/operation manuals online.

152NP

@Dave Carpentier that is my understanding. If you install the optional system sensor, then the boiler reads that one instead of its built in outlet sensor. The installation manual is clear on that. What the manual suggests, but does not explicitly state is, if you do NOT install the outdoor temperature sensor, then the boiler will not use its built in ODR to adjust the target temperature, and just try to hit the design water temp. That is what we want in this case, to disable the boiler's ODR and run the buffer tank at a constant temperature (within a differential band), regardless of the temperature outdoors. We will regulate the temperature of the water the radiators get using an injection mixing pump on its own ODR. At least that is the current plan.

hot_rod

Here are the comments from Boon
If the control has enough portions, not all do, you should be able to tame down the firing. Step firing would be one way to prevent aggressive overshoot

Bottom line is determine how the control you have operates and how much trimming you can do. It will come down to how much tinkering you are willing to do to get exactly the result you want. Over the years the control designers have added more and more adjustable options, many requested by installers that want the fine tuning ability. It sometimes looks like a competition between manufacturers to provide the control setting options to one up the others

we are certainly light years ahead of the basic aquastat with fixed differential controls that were mainstay for 100 years😙

152NP

@hot_rod here is what I have so far on the buffer tank. I have space constraints that necessitate that the return inlet and outlet be on the same side, so this is my solution.


Maybe not best, but the best I could come up with while keeping this project moving along. Seems to me like it should work okay.

While I continue to explore what the optimal way to control the system is, I can still make progress on the tank. Other than adding a tapping mid way up the tank for a temperature sensor and a drain at the bottom, are there any other things I should be adding to it, even if it is for "just in case?"

I think I am going to call Lochinvar's tech support to see if they will reveal any details on how the boiler decides to modulate. Understanding that black box would be helpful in making other design decisions.

hot_rod

tank looks great, I’ll try and get to a Lochinvar control guru also

152NP

@hot_rod I had a good conversation with Lochinvar tech support. Very nice people. The tech I spoke to said the Noble boiler modulates by starting out full throttle, and turns down as it approaches the target temp based on the supply sensor. He and I agreed that the only way we will know for sure is to test the behavior of a tank sensor vs a supply sensor to see which gives us the result we are after. He had never seen one installed in a system with an injection mixing pump after the tank. It makes you wonder if regulating the tank temp on outdoor reset and doing away with the injection mixing pump would be the easier way to go, but, for some reason, running the tank at a constant temp makes more sense to me. It seems silly to try to adjust the temp of 60 gallons of water up and down as the outdoor temperature fluctuates.

There is also the idea of running the tank and distribution system at a fixed temp and just let the individual room TRVs do their thing. Many things to consider, and wonder which is going to translate to comfort and meaningful $$ savings.

How you would monitor the performance of the two different sensor locations without some kind of system history logging is beyond me.

hot_rod

Since you have a modulating boiler that can turn down 10-1 running the tank on ODR does simplify the secondary side cost and piping, tough call on which works best.

As for the sensor, which ever sensor does the job best. I think you want to make sure and cover the load, then recover the tank after all loads are off. I suppose if it ramps down as the tank is being heated back up, that would not be an issue. Probably only a 15- 20 minute difference in time between modulating, or boiler at full output to recover.

As long as the load is always getting sufficient heat, then the topping off of the buffer is secondary, maybe not a critical time frame for that to happen. Maybe the boiler runs more efficiently in the slow recovery mode, more condensation, or big HX exposed to a small flame relationship??

Arguments coulad be made both ways for all these questions

152NP

@hot_rod thank you for the generous amount of advice you have offered on this topic. I sincerely appreciate all your help.

I think I am fortunate in that that I can test a few different control scenarios by just changing some wiring connections, or locating a sensor from one place to another. It will be fun to see what happens. Understanding a bit better now how the 3-pipe buffer tank configuration works, I think it affords a lot of flexibility. Worst case scenario is I run the tank and distribution at fixed temp and just let the TRVs take care of things...no big deal. All I have to do to test outdoor reset on the tank is to plug in the boiler's outdoor temperature sensor.

I can ask an infinite amount of questions, and debate hypotheticals, but I think it's time I do some piping already. I am going to leave the injection mixing out of the picture for now to keep it simple. First priority is get some heat working before it gets cold around here!

hot_rod

yes, as long as the piping is correct, you can play with controls, sensor locations, etc.

I saw a German buffer tank at a show in Munich years ago. It had a rail down the side that you could position sensors, multiple sensors anywhere along the tank.
On the pellet boiler buffers with oil fired backup boilers, a sensor a foot down from the top would trigger the oil boiler. So only a very small portion of the tank would come into play when running on the fossil fuel boilers.