Welcome! Here are the website rules, as well as some tips for using this forum.
Need to contact us? Visit https://heatinghelp.com/contact-us/.
Click here to Find a Contractor in your area.

expansion tank

i am removing a expansion tank and want to replace it with a bladder tank, but i do not know how to size it. I can not size it by boiler size because i have two, four million BTU steam boilers that do several hot water heat exchangers. all i have to go by is the old tank size which is 20" diameter by 50" long. WHAT SIZE TANK DO I NEED

Comments

  • Mark Eatherton
    Mark Eatherton Member Posts: 5,853
    Want some pi???

    pi x radius square times the length = cubic inches.



    Cubic inches divided by 1728 = Cu Ft.



    Cubic Feet times 7.5 = gallons of volume.



    Match the gallons of volume to a Watts model number and you're good to go.



    So, 10 X 10 X 3.14 = 314, times 50 " length = 15,700 cubic inches.



    15,700 divided by 1,728 = 9.08 Cubic feet.



    9.08 times 7.5 = 68 gallons of volume.



    Go to the next size larger tank, so use an Amtrol # 90 floor mount expansion tank.



    The correct way of doing it is to determine system volume, minimum and maximum allowable pressures, minimum and maximum water temperatures and size accordingly, but I assume getting the volumes down pat is an issue. You can NEVER oversize and expansion tank.



    Heres a link to the Amtrol sizing program.



    http://www.amtrol.com/sizing.htm





    ME
    It's not so much a case of "You got what you paid for", as it is a matter of "You DIDN'T get what you DIDN'T pay for, and you're NOT going to get what you thought you were in the way of comfort". Borrowed from Heatboy.
  • Brad White
    Brad White Member Posts: 2,398
    Mark is absolutely correct when he says,

    "The correct way of doing it is to determine system volume, minimum and

    maximum allowable pressures, minimum and maximum water temperatures and

    size accordingly, but I assume getting the volumes down pat is an issue.

    You can NEVER oversize and expansion tank."



    Let me tell you a little story about an engineering mis-adventure. Pull up a chair. Yes, it was me. I was asked as an, "oh by the way" additional service on an air handler replacement in a student housing complex, to size and specify replacement expansion thanks, bladder style for the older open tanks which were about 35 years old.



    Upfront, the lessons-learned are "There is No Such Thing as a Small Job", and, "The Fundamentals of Engineering Cannot be Ignored", rapidly followed by, "Success Absent Calculations is Not Engineering, But Sheer Luck."  Oh, and ask questions, such as, "If the tanks work, presumably, why replace them? Say "Amen" for me, for I am saying, "Duh".  Their statement of "air binding" and the even less specific, "air problems", made a sealed bladder tank the solution of choice. If you do it correctly.



    I followed our then standard procedure to replace such tanks. We calculate the volume of the old tanks and have a program/spreadsheet from Amtrol which will give you an equivalent size. The old tanks had a volume of about 450 gallons and a static fill pressure of 50 psi for this complex. What we specified as a replacement actually had 150 percent of the acceptance volume as the originals, so we were safe, right? Uh, no. Not by a long shot.



    Add to this system a make-up water meter they never had, plus a leak detection system. Thirty seconds of unattended flow sounds an alarm. Simple.



    System is placed in operation and warms up via steam heat exchangers to an outdoor reset scale. All is good so far until the first cold snap. The penthouse air handler freeze-stat trips and there is no heat on the top floors. There was a make-up water volume of over 400 gallons in one night. No known leaks. None.



    After cogitating a bit, I deployed some data loggers at the HEX to get coincident temperature and time data. Fishing a bit, but a place to start. I put my hand on the relief valve pipe and the sucker is HOT. That was where the water was going, in bursts. For over 30 years. Lots of water only NOW we can measure it. No one knew.



    I told the owner, "frankly, the only thing which would make this behave this way is that the expansion tanks are undersized. Changing the relief valve will not help. There is not enough expansion tank volume. I think we messed up."



    Well we calculated the system volume from old and incomplete drawings, but decided to do a dye test which is our new office policy thanks to this.  We also back-calculated what volume the original 450 gallon tanks would serve and came up with about 2,500 gallons system volume. What do you think the dye test revealed? The average of three readings one percent apart gave us 13,000 gallons!



    Our tanks were off by a factor of 3.5 and the old tanks by a factor of over five.

    And for over thirty years, they probably dumped,  made up and heated millions of gallons of system water down the drain and suffered inconvenience.



    Had we not installed the make-up water meter, we would not have found out nearly so soon.



    Anyway, a variety of Taco CXA tanks are now being installed. And I want to see their water bills next year.
    "If you do not know the answer, say, "I do not know the answer", and you will be correct!"



    -Ernie White, my Dad
  • Mark Eatherton
    Mark Eatherton Member Posts: 5,853
    FIddle Faddled by the Fickle Finger of Fate....

    Been there, done that, got credit card receipts to prove it :-)



    Tell me more about the Dye Method of volume determination. I've got a rather large district heating system with lots of unknowns.



    Am familiar with sodium titrate tests, but not dye.



    Thanks for coming back pal. We've missed your wisdom and humor!!



    ME
    It's not so much a case of "You got what you paid for", as it is a matter of "You DIDN'T get what you DIDN'T pay for, and you're NOT going to get what you thought you were in the way of comfort". Borrowed from Heatboy.
  • Brad White
    Brad White Member Posts: 2,398
    Thanks, Mark

    It is good to be back. Still busy but things are settling in good ways.



    Upon browsing the site anew, it would seem that this post about the expansion tanks might be in the job hell category.  :) But in the end we stumbled upon the problem and the Owner seems happy except for the tank farm we are building for him.



    The dye test, I do not have all of the technical details, (we have this done for us by a local boiler service company), but they inject a known quantity of a dye which has fluorescent properties and can be read with a spectrofluorophotometer.  (That is what I call it anyway.)



    The stuff I have seen is green and for the system we tested, they used a one liter bottle of the stuff, if I recall correctly.  It circulates for 48 hours and a sample is drawn off and read by the instrument, (hand held no less).  Probably no different a methodology than titrate tests, but I am not familiar with those specifically.



    By calculating the diluted concentration against the known quantity injected, they can get a determination of volume, sort of a back-calculation. Three samples are taken and averaged. Our readings were all within one percent of 13,000 gallons, so "close enough for government work".



    We like this because the results are nearly immediate (from sample drawing, not injection), but also because of the repeatability and accuracy.  The cost was about $200 and the dye itself was about $40 of that. Such a deal!



    Now, my own house (new to us, 1913 with 3" and 2.5" gravity hw piping), I am draining her down into 42 gallon barrels, of which I have two. Old fashioned way, for sure, but I can filter out the stray iron and stuff before reinjecting it. There is no drain point in the basement so I have to take it to barrels anyway, or pump it.
    "If you do not know the answer, say, "I do not know the answer", and you will be correct!"



    -Ernie White, my Dad
  • Mark Eatherton
    Mark Eatherton Member Posts: 5,853
    WIll have to run that one by Dwight Hedgpeth

    with Rhomar and see if he's familiar with it.



    Sodium titrate test is essentially the same. Measure residual sodium in the system to begin with, then inject a known quantity into the system, circulate fully for a couple of days, then read and calculate dilution ratio..



    BTW, there is a "system" out there that is a single tank system, that can eliminate the need for an expansion tank, tank farm. I believe it is put out by Wessells (pronounced vessels, kind of like veelo)



    It uses a compressor, micro processor and bleed valve to maintain ideal system pressures. Not a big fan of bleeding to control pressures, but it works.



    HTH



    Thanks



    ME
    It's not so much a case of "You got what you paid for", as it is a matter of "You DIDN'T get what you DIDN'T pay for, and you're NOT going to get what you thought you were in the way of comfort". Borrowed from Heatboy.
  • Brad White
    Brad White Member Posts: 2,398
    Dwight is Quite

    the guy. Rhomar used to do that test for you with a sample bottle but stopped doing it, last time I asked. Still, Dwight is a fount of knowledge. I never knew what chemical they used, but yes, same principle.



    With the dye, you do not send the sample back. It was rather fun to watch. We all had bets going, highest volume without going over. I guessed 8K. Way off but I won :)
    "If you do not know the answer, say, "I do not know the answer", and you will be correct!"



    -Ernie White, my Dad
  • Mark Eatherton
    Mark Eatherton Member Posts: 5,853
    HMmmmm....

    In visiting the Vessels web site, I see no mention of their "Control System". Seems they have abandoned it...



    I suspect sticking solenoid valves caused a high rise to empty ;-)



    Tank farm indeed.



    ME
    It's not so much a case of "You got what you paid for", as it is a matter of "You DIDN'T get what you DIDN'T pay for, and you're NOT going to get what you thought you were in the way of comfort". Borrowed from Heatboy.
This discussion has been closed.