Open Expansion Tank Questions

  • Active since 1995, Hearth.com is THE place on the internet for free information and advice about wood stoves, pellet stoves and other energy saving equipment.

    We strive to provide opinions, articles, discussions and history related to Hearth Products and in a more general sense, energy issues.

    We promote the EFFICIENT, RESPONSIBLE, CLEAN and SAFE use of all fuels, whether renewable or fossil.
Status
Not open for further replies.

ED in CO

New Member
Oct 23, 2013
28
Colorado
I have been connecting pipes for the wood boiler fireplace and I'm confused about the open expansion tank set up. I may have located the pump on the wrong side. I have it on the return, just after the bypass and before the drain valve.

1. Should I move the pump to the supply side?
2. Should I eliminate the open tank and just use a closed system pressure relief valve?

I'm trying to do something like this. Except I left out the cold water return because it is much easier to fill the boiler from the storage tank side where the water is located.

Open System Header Tank.gif
I have the pump on the return side--I read that it is better for the pump to be on the cold side, but pushing the water into the stove instead of pulling it away from the stove makes me think the water will go right up into the expansion tank and keep on filling it until it overflows. Would it be different if the pump were on the hot side? Would a cold water return from the overflow tank help avoid the pump sending water up the vent pipe?
Spectra plumbing with plate exchanger.png
I am using a storage tank and a plate exchanger from that tank to the heating system.
Spectra with Accumulator Tank.JPG
Could I forget the open expansion tank and just try installing a pressure relief valve? The directions specifically state that the boiler side should be open though. This is how the overflow pipe will connect to the stove--into the upper pipe. The lower one was removed and plugged after the system held air.The pipe will have a curve where it enters the overflow tank.
Wood Boiler Overflow.jpg

Am I correct in thinking that the pump in the return will pump water right out of the top of the system to the overflow tank instead of or in addition to the supply pipe?

By the way, I used a new gas tank for the expansion tank because it was very inexpensive--only $22 on Amazon. It has plenty of room for this system's expansion and open holes for air and an overflow drain. But it would not be missed if I switched to a closed system on this side. Just hesitant to go against the directions. A non-pressurized boiler is a safer boiler.
 
Please stop work. I think you'll need to get some local talent in to look the job over and make recommendations based on the boiler install guide and its specifications.
 
Last edited:
A few things/questions:

1. If you are going to do an open system (non-pressurized), you need to ensure that everything in your system is free of any corrosive metals (cast iron, steel). If you don't know that to be the case, then it would be foolish to go with an open system.

2. Regardless of system design and type, I would stick with equipment and materials that are designed for that purpose. Even if it worked OK, I would be very concerned about the loose connection at your gas/expansion tank. Just be aware that materials (especially PEX) will move, bend, flex, etc. a great deal when they come up to temperature. I wouldn't be surprised if that line just slid right out of fill port, once it warmed up.
 
Thanks for your comments.
Yes, I know steel should not be part of an open system, all of the pumps, valves etc. are stainless or brass.
But properly designed the idea is that fresh water does not enter the system. The header tank is a fail safe option in case the stove overheats.
This is not the type of open system that uses fresh water constantly--I've seen those for radiant heat where the household hot water is also heated by the system. It is generally not recommended precisely because the constant introduction of fresh oxygen is so corrosive.
I will take your advice about pex seriously. For now that pipe was temporary--just to understand the workings of the system. I am out of 3/4" copper for now, but all the pipe near the boiler is copper. Plus I need to build a structure that puts the tank directly over the stove. I don't want it on top of the cabinet. The directions I found said that the tank should be above all the radiators and about a meter above the stove.
I was looking for advice about the setup of an open system. I think it is recommended because the stove should operate under low pressure. I was planning to change the in and out pipes (eliminate the return side of the header tank) and wondered if anyone had experience with this type of system--to let me know if there is a thermosyphon reason for the return pipe. I'm pretty sure I should move the pump to the supply side, but thought I could get some advice on that here.
I'm extremely doubtful there exists an expert in the Denver suburbs with this type of expertise. I talked to a couple of installers for the gas boiler and they did not seem to exhibit the expertise I expected. They were not interested in Manual J calculations and tried to talk me into a larger boiler etc. Of course it is different in upstate NY where systems like these are more common. That is why I'm asking the questions here where the experts truly exist. My self installed gas and radiant floor system is working great. I actually wonder if a wood system is even necessary since we use so little natural gas. But we have heated with wood most of our adult lives and if we have a fire we like the idea that we are contributing to our heat too. Plus I already bought the boiler stove.
If no one volunteers an explanation of the requirements of the open system tank here then I will experiment to get the right answer, same as most of the folks who post here.
 
Hydronic systems are hydronic systems - whether solid, gas, oil, or electric fueled; so installers out there should be able to help you out. If you're looking to do it on your own to save some $$, I completely respect and encourage that. When we discuss open or closed systems though, the biggest difference is whether the system is open to atmosphere or not - not that the system is a "once through" system which continually dumps water down the drain.

I think everyone will agree that open systems present more challenges than closed systems do. You need to ensure that the reservoir/expansion (this is what's open to the atmosphere) is at a higher elevation than everything else in your system and ensure that it always maintains a water level to ensure the entire system is filled. In a closed system, this is not an issue as the expansion takes place in a pressurized tank. Open systems will allow oxygen to enter the system continuously, so whether you're refilling it with fresh water or not, your system will promote corrosion.

As far as the pump location, as you mentioned before, I would suggest that the pump be installed on the return side of the boiler; such that it pushes the water through the boiler. The suction side of the pump (which is system's lowest pressure location) is where you should tie in your expansion. I would think this would be ideal for either a pressurized or open system.
 
Last edited:
You don't have to keep adding fresh water to an open system to get oxygen into it. It will get in the water through the exposed water surface.

My boiler specs open expansion tank to be 8' above highest system point to help reduce ogygen saturation. If I was set on doing open expansion, that is what I would do. I would consider an electric hot water heater tank for the expansion vessel. Size at minimum 5% system volume. I would run one line, from a point close to the inlet of my circulator, up to the bottom of the tank. I don't understand why the two lines are needed in these 'header tank' situations - seems to me an unecessary complication. But I think after going to all that trouble, I would go further and size the tank a bit bigger, and close it off to make it a closed tank (can valve a top fitting to atmosphere & keep the valve closed). And install a pressure relief valve on the system. But that is without knowing anything about your boiler or what it is specd for pressure wise. If it insists that the boiler be open - I suspect it isn't specd for any pressure, so you will have to do open expansion.

Also, the way you speak implies a pressure relief valve is a substitute for expansion. That is not true - expansion is needed of some kind in any case, open or closed. And if the boiler specs open, the boiler might calve before the relief even does its thing.

Which would also make me leery about the overall construction & ruggedness of the boiler to start with.

I would also consider ewdudleys advice - given what has transpired so far and the assortment of portions of different piping diagrams shown, you might end up doing something that might lead to a bad result. Hopefully not towards the badder side of bad.
 
Thank you for these helpful replies. I think the system is specified to be open because it should operate at a low pressure and I understand that open tanks do not let the pressure in the boiler rise over atmospheric pressure. Sort of like a pressure release valve for 15 psi instead of 150. I'm not too worried about the strength of the stove because it was sold by Buderus in Europe although manufactured by Cichewicz in Poland, the stove seems like a quality unit in fit and finish.

I prefer to leave the pump on the return side, so glad there is that recommendation.

An open system can introduce air just by being open to the atmosphere. I thought the air was limited if the aerated water did not circulate through the system. So good to know. Closing in the water with an enclosed tank would help reduce corrosion.

I looked for a used small hot water heater for months on craigslist and could not find one that was not too expensive for used. I don't need one that works, but didn't see anything so thought I would try an inexpensive tank. The gas tank was the least expensive vessel that holds enough water that I could find. I'm intrigued by the idea of using a closed tank and only having the overflow open.

I have an expansion tank and a pressure relief valve on the storage side of this system. Didn't want a tank of hot water sitting there without both safety devices. If I were to use a pressure valve, I would include an expansion tank. But thanks for the reminder.

It seems that the return line from the overhead tank doesn't serve any purpose if the boiler has an alternate way to be filled. That means I should be able to stick with my design.

I still don't really see why the pump doesn't just fill the overflow tank--even with the check valve there, my understanding of pump pressure and water is that the pressure from the pump will push the water right up the vent pipe! But maybe not.

I did use a lot of examples but that is because each one shows a different aspect of the problem. I posted my diagram here a couple of weeks ago but didn't get any comments.
Fireplace Boiler Piping-14_sm.jpg
Even then I used an open system tank that was plumbed differently. It was drawn before I found the tank I thought I would use and is plumbed more like a hot water heater would be. The storage tank in this drawing is in the utility room where the gas boiler is located.

For the expansion tank height, I don't have 8 ft. to work with and the specs in my installation manual say 3' so hopefully the smaller amount of water makes a difference in the required height of the tank. I'll put it as close to the ceiling as I can though. More distance is better.

As far as getting expert help, I have no idea who to call or what to say to this HVAC person. They will surely think I am nuts and not appreciate that I've done much of the work myself. I just don't have patience for a naysayer.

I will have to see how the system runs once I get water into it and I'll find out about the overflow tank's operation then.
 
An open system can introduce air just by being open to the atmosphere. I thought the air was limited if the aerated water did not circulate through the system. So good to know. Closing in the water with an enclosed tank would help reduce corrosion.
During normal operation an elevated expansion cistern will cycle a percent or two of system volume in and out of the tank depending on how much and how often the temperature of the system water cycles. Elevated expansion cisterns have a reputation for doing a very good job of minimizing the amount of oxygen that gets into the system due to the isolation of the water in the tank and the limited surface area. It is often recommended that a layer of "mineral oil" or "paraffinic oil" be maintained on top of the water in the tank to eliminate oxygen absorption.
The gas tank was the least expensive vessel that holds enough water that I could find.
Whether or not your expansion tank functions correctly in the end is one thing. Nevertheless it is important to have your project look like it is even possible you know what you're doing, which will not happen with that tin tank and the PEX hanging out of it.

I'm not too worried about the strength of the stove
The pressure limit of the boiler is not affected by its apparent strength. Whatever pressure limit the boiler manufacturer specifies is the limit your expansion device must maintain.

It seems that the return line from the overhead tank doesn't serve any purpose if the boiler has an alternate way to be filled. That means I should be able to stick with my design.
I still don't really see why the pump doesn't just fill the overflow tank--even with the check valve there, my understanding of pump pressure and water is that the pressure from the pump will push the water right up the vent pipe! But maybe not.
I cannot figure out what you're trying to say here.

At any rate an elevated expansion cistern needs (at least) two lines running to it. One fat unobstructed pipe, generally at least 1" ID, running directly from the top of the boiler to a point above the surface of the water in the tank so that steam and slugs of water can blow off into the tank should the need arise. The other much smaller line enters at or near the bottom of the tank and is connected to the system somewhere near the inlet of the pump. The height of the fat pipe where it "goes over the top" must be high enough relative to the surface of the water in the tank such that the pressure differential between the little pipe and the fat pipe does not cause a waterfall in the tank.

Placing the circulator pump on the supply side of the boiler as shown in your first diagram above insures there will not be a problem. If you place your circulator on the return side then you need to do some homework that proves you know how it really works and that you know what you're doing.

I did use a lot of examples but that is because each one shows a different aspect of the problem. I posted my diagram here a couple of weeks ago but didn't get any comments.
No. See the discussion above about the two pipes and what they are for.

Even then I used an open system tank that was plumbed differently. It was drawn before I found the tank I thought I would use and is plumbed more like a hot water heater would be. The storage tank in this drawing is in the utility room where the gas boiler is located.
I have no idea what you are trying to say here.


I will have to see how the system runs once I get water into it and I'll find out about the overflow tank's operation then.
No. You need a fat unobstructed blow-off pipe made out of steel or copper.
 
Last edited:
Thank you for your comments and the time it took you to try to understand my issues.
It is not unusual to get contradictory answers to questions on any forum. Your answers are a bit contradictory to the summary I just posted of the answers in previous posts. I'm sorry you couldn't tell what I was saying, but building on former posts does require that one read them.
Didn't realize I had to know what I was doing before I could post a question. Most folks on here don't really know but are learning. I'm not going to blow myself up--I have some common sense and I hope you are not thinking that I won't be able to solve these problems on my own.
I think you read too much into the tank and pex photo--it is not the final piping and was to illustrate the ideas I was presenting.
If you read the other serious answers to my questions you would know that I was responding to their opinions on the need for two pipes from the header tank and some doubt about the quality of the stove, etc. I understand the stove is only designed to work at atmospheric pressure, thus the recommendation for an open header tank.
I have read about the mineral oil on top of the tank water. I also looked at several English header tanks and they are made from steel and even plastic. Galvanized metal is not really tinned.
The question whether two pipes are necessary coming out of the header tank is a serious one. Someone said that they didn't think two were necessary, although my intuition told me that there was some type of balance between the two pipes. Another person recommended that I keep the pump on the return side, although again my thought was with this open tank, the pump should be on the supply side as in SOME of the diagrams. But I don't really want to move it.
There are also diagrams of the cold pipe from the header tank entering the hot larger pipe instead of the supply pipe further down the line. That did lead me to believe it is not always necessary to have two pipes.
You say both pipes are required but don't really explain why. I have been researching this question too and have not found an answer.
You don't understand why I might think the pump would fill the overflow tank? Think of a pump pushing water down two pipes, why wouldn't it fill both? Seems like a simple thing to interpret. If the purpose of the header tank cold water return is to keep the boiler filled then it would not be necessary if there was also a boiler fill valve on the system. Another reasonable statement, or I meant it to be.
On my gas boiler, the secondary loop pump is on the supply side and the primary is on the return. Both seem to be working fine so I'm not sure what other research/knowledge I would need to install the pump on the return. Except of course if it matters because of the open system tank.
Also if you look carefully at the diagram of my system, that does NOT have a cold water return pipe, it shows the overflow being pumped into the bottom of the tank, unlike the diagram where the water enters above the tank. I have also found diagrams of the water entering at the side as my tank would be set up to do. That is why the diagram of my system is not plumbed the same as my current plan. All of these systems seem to work under the right circumstances.
If you don't have enough elevation to the tank you can actually install a fatter pipe and it will work as sort of a pre-tank--I got this info from English boiler forums--where open systems are more frequently installed. In fact, horizontal installations can also work--I could post that diagram as well. I have read English experts who say they prefer the open systems as safer with less to go wrong. It may just be that is what they know.
Sometimes a solution just requires some experimentation, but before I continued, I thought this was a good place to ask my questions.
I am also considering a low pressure relief valve and a boiler expansion tank instead. It seems like that is a much simpler system and I don't have to worry about waterfall effects or pumping too much water into the overflow tank or mineral oil or keeping the tank filled etc. etc. It seems if the relief valve is close to atmospheric pressure then the system would operate under spec.
 
We haven't heard yet, I don't think, the pressure specs on your boiler. All we have heard is that it specs to be left open - which implies it is not rated for any pressure.

And does it have a manual with installation diagram - which should show how to do this?

I am no pro, and have no experience with header tanks. So I'm out of this one now I think.

Good luck.
 
Yep, that diagram is one way to install what I was trying to accomplish, as posted in my original question.
Operate at No pressure? I mentioned that a header tank would provide atmospheric pressure. Air is pushing on the water which is pushing against the circulating water in the stove. Plus there is pressure from the pump which seemed to me would be great enough at that point to push water right up into the overflow tank. The specs say the maximum working pressure is 2 bar. Sorry I didn't include that figure. That is almost 30 psi so anything under than should be OK.
It seems that using a header tank is going to cause too many problems for this setup. I'm going to proceed using a 15 psi pressure relief valve and an expansion tank for the boiler. It is a cheap solution and your answers to my questions have helped me make that decision. Thanks.
I have learned many things using this forum so thanks to all for posting.
 
If you are doing an open system (atmospheric tank), the pressure in the system can never exceed the maximum elevation head that your water level can achieve. If it did, it would overflow that atmospheric tank.
 
Yes, that makes sense. What concerned me was that the pump on the return at the bottom on the stove would overcome the head pressure and fill the tank, something I had not thought about before I set up that tank and pex line. None of the setups I looked at had this issue, youtube shows header tank systems that fill because there is a leak in the coil that is allowing reverse flow of the water up the vent pipe, but nothing about pumps causing the problems.
It seemed to me if the pipes to the tank were just before the pump instead, that would create a suction that would be less likely to fill the overflow. So the solution might have been to move the pump. The answer probably is that the pump on the return side is unlikely to create enough pressure to push too much water up into the overflow tank due to the head pressure we have been talking about with the open tank.
I don't really have a complete understanding regarding the two pipes from the header tank. Most of these systems fill from the header tank and I don't want to do that. I have a fill valve that will take care of keeping the system full of water.
I have two options, continue to experiment with the header tank, or move to the system I'm more familiar with, the pressure relief valve and an expansion tank. With all we have discussed, there are too many requirements for the header tank in my estimation and I think the clarification about the working pressure of the stove leads me to believe it will operate just fine with the low pressure relief valve instead.
 
I understand what the rationale is for the concern, but don't think you have anything to worry about. I may have a difficult time trying to explain it though. Although the pump is intended to build pressure, it is also pulling suction at the same time. Since your system will always maintain a positive head pressure, the entire system will remain flooded at all times. This means that instead of the pump "building pressure" it will just serve to move water.
 
Your explanations have been very helpful. It is good to know that the pump is unlikely to build enough pressure to blow the low pressure relief valve. I figured that if it does, I can buy a higher psi valve--as long as the boiler pressure stays below 2 bar, it should be OK. I'm relieved that I don't have to figure out the open system tank and this information has helped to clarify that I can install a closed system.
 
Welcome to the hearth forum Ed.

Just so you know, ewdudley is top shelf in hydronics. Free advice from someone like him is a God send. Most of the posters here have years of experience with wood fired boiler systems and do not advise inferior solutions.

We all hope the best for everyone in their system builds and your ability to say good bye to the fuel suppliers. Rock on with your setup.
 
Thank you! I am thankful for all the posts from those who were willing to give advice. Folks like ewdudley and others here who have years of experience with hydronics and actually know the physics behind the workings of the systems are rare and appreciated. Where else could I get such good advice and recommendations? Being able to get feedback helps further the thought process. I may have bristled a bit at being misunderstood--sorry if I gave offense.
Thanks for the good wishes for my setup. I think it will be a great system. Low BTU's for a superinsulated passive solar house and the comfort of radiant vs. circulated air heat. Plus the beauty of a wood fire in the evenings. Not to mention a choice in fuel supply for energy independence.
I have been lurking for a couple of years now and research using the search functions. Lots of great info here.
 
Brant2000-I quoted your reply about the workings of the header pressure vs. the boiler pressure on my building blog post about this issue.
http://blog.twinsprings.com/2014/12/04/wood-boiler-clarification/
and further explanation of the questions I posted here
http://blog.twinsprings.com/2014/12/02/wood-boiler-confusion/
Your explanation helped clear up my confusion about the open system tank and helped me to conclude that the low pressure relief valve would serve a similar and easier to install purpose.
I also wrote about ewdudley's concerns about the use of pex and the overflow tank as illustrated in the photo-so that nobody will look at that setup and think that is what they should do!
Thanks again.
 
Last edited:
Still seems to be a bit of misunderstanding in your blogs. A circulator pump won't build pressure in the system. It will just move water around. In order for it to build pressure, it would have to introduce more volume into the system as it pumps.

What is the total volume of your system? A 5 gallon expansion tank is not very big - I would hazard a guess that it is going to be too small and you will overpressurize the system on heating it up. Also wondering why use a heat exchanger if the whole system will be closed.
 
YES! That was my confusion. I thought the pump on the system, located on the return side would push water through the stove and into the open overhead tank. I was hoping the blog clarified that was an error in thinking. :confused::p
I didn't see what was keeping the water from moving up that open pipe in addition to moving past it to the storage system. So I thought if I moved the pump to the supply side, it would create the suction at the overflow pipe and pull the water into the pump instead of up the pipe. That was my original question, should I move the pump?
It took several tries here but I finally understood due to Brant2000 that the pump will NOT overcome the pressure from the tank if there is sufficient head and the same thing is true as long as the stove is operating below the level of the pressure relief valve.The pump itself will not "build" pressure in the system.
The reasons I decided to switch to a closed system are that the solution was inexpensive, the complications of the install and maintenance of the overhead tank, and the understanding that as long as I had a low psi pressure relief valve on the stove, it would operate below its maximum pressure rating unless something went wrong. Those were all concerns that I had and these folks helped me figure it out.
I decided to use the heat exchanger because the stove is steel and the rest of my system is copper, stainless and pex. I spent a lot of money keeping any and all cast iron out of the main heating system as well as treating our water to minimize iron build up. I didn't want any contamination from the steel wood stove in the main system. At the time, I also thought I had to have an open tank on the main system and I wanted to isolate that too. But I have changed my mind and will have both systems closed.
Thanks for the additional clarification. Even though I was wrong about the system pressure and the pump, I am glad I asked the question because it resulted in a new design for the system.
The stove itself is about 14-15 gallons and the storage tank is 40. There are two runs of 1" pipe to the utility room where the storage tank is located of about 20 ft.
I have an expansion tank on both sides of this run of pipe, a smaller one is on the 40 gallon tank. Both are on the wood boiler side of the heating system. The gas side of the heat exchanger has its own expansion tank--I'm thinking it is also about 5 gallons. That is for all the water in the radiant and the radiators. I don't remember the totals right now but it is a sizable amount.
 
Last edited:
Hi, What is your reasoning for thinking the expansion is undersized?
Is that because of the heat generated by the boiler or the system capacity? Or something else? Am I too close to the estimated requirement?
I bought a Watts non-potable tank and used their calculator. I just rechecked my figures and measured all the pipes around the boiler and the water heater storage tank--total is 68 gallons. I have a 2 gallon expansion tank on the water heater--based on the Amtrol calculator and here are the numbers I used to buy the 4.7 gallon tank. I figured since there are only about 15 gallons of water on the boiler side, that the 4.7 would be sufficient. What am I missing?
Watts Calculator.png
If I increase the incoming pressure 1 PSI and do the calculations for the total of 68 gallons, I get 6.27 and I will have 6.7 on the system. But if the incoming pressure were 10 PSI then I would need 16 gallons of expansion. So maybe I should have planned for the possibility of higher incoming water pressure. The fill valve is adjustable from 3 to 60 PSI and the system is supposed to operate at low pressure. Did I set it too low?
Thanks for your input.
Watts Calculator @ 68 Gal.png
 
OK - if your total system volume is only 68 gallons then you might be OK.

That is very small for a combined boiler & storage volume - and you didn't answer before (unless I missed it) when asked how much system volume you have. So if that's really what it is, then that's what it is. But I am not sure of the adjustability of your pressure fill valve or the accuracy of that range on the low end - so you want to be very careful with that since you will be very close to the max on your expansion volume calcs. If the fill valve lets in just a little more water than you want when the system is cool, it won't take much to push you to blow off pressure when hot. Personally I would like a wider cushion of safety.

Also I think your air pre-charge on the expansion tank should be set to close to or at your cold system pressure.

So if you get any of that stuff off a couple PSIs when setting up, you will likely blow some out of your relief valve.
 
You are right, I didn't mention total system volume, but just the volume of the stove and the storage and off the top of my head said the water heater is 40 when it is actually 50.
I measured all the pipe and added that volume of water too (2.6 Gal.) and calculated a total of 68 gallons rounded up.
It would be better to have a larger margin of error in the expansion tank as you suggest. Reducing the charged air to 6 PSI increases the size of the tank needed to about 8 gallons. Reducing the pressure to the system pressure at the installation point is recommended in the install instructions but I didn't reduce it in the calculation. Thanks for pointing that out. Reducing the output temperature of the system reduces the size of the expansion tank too.
I know it is an unusually small system. Our house is super insulated and passive solar heated too. It is 2213 sq. ft. We will use the stove mostly on cold evenings. It is likely that the stove will have to re-charge the temperature of the storage system when it is operating and I figure the heated storage tank will only substitute for the gas boiler for a few hours at a time. The gas boiler is a low temp system on an outdoor reset. (Modulating/Condensing) Last night it was about 32 outside and the boiler when it was firing had 123 degree water being sent to the valves and the return gauge said 100 degree was returning. That delta T is perfect. The floors are designed to heat the house at zero at a temperature of 86 degrees. A digital temp gun measures exactly 86. At just below zero outside the system was sending out 139 degree water, more often. So I'm happy that the system works as designed--since I designed and installed it.
I will consider replacing the expansion tank with a larger one especially if the pressure relief valve pops.
 
Status
Not open for further replies.