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.

Original ARCO Gravity Hot Water

Kevin Pulver
Kevin Pulver Member Posts: 67
<BR>

Comments

  • Kevin O. Pulver
    Kevin O. Pulver Member Posts: 380
    SAVE THE RADIATORS!

    This is a 1920s 2 story home with ARCO radiators and coal boiler converted to gas. It works beautifully, but potential buyers are terrified of it. The mystery and inefficiency are too much I guess. I've been approached about upgrading it. The owner was terrified I'd want to remove the radiators and add ductwork. Isn't that a nice switch? I will do a heat-loss and am considering a mod-con in pri-sec but have never done a conversion like this. Any ideas from you Pipe-Artists as far as piping layout and circulator sizing would be appreciated. Something tells me I wouldn't need much circulator since it was gravity in the first place, but I know I need something. What do you estimate efficiency is on this?
    Thanks for the voice of experience! Kevin
  • Kevin O. Pulver
    Kevin O. Pulver Member Posts: 380
    SAVE THE RADIATORS!

    This is a 1920s 2 story home with ARCO radiators and coal boiler converted to gas. It works beautifully, but potential buyers are terrified of it. The mystery and inefficiency are too much I guess. I've been approached about upgrading it. The owner was terrified I'd want to remove the radiators and add ductwork. Isn't that a nice switch? I will do a heat-loss and am considering a mod-con in pri-sec but have never done a conversion like this. Any ideas from you Pipe-Artists as far as piping layout and circulator sizing would be appreciated. Something tells me I wouldn't need much circulator since it was gravity in the first place, but I know I need something. Must I figure every foot and diameter of pipe in the circuit or is ther another way? What do you estimate efficiency is on this?
    Thanks for the voice of experience! Kevin
  • Mike T., Swampeast MO
    Mike T., Swampeast MO Member Posts: 6,928


    Utterly no need to measure piping and compute head loss. Head loss to 2nd floor rads of gravity systems had to be less than 2.4 inches and ground floor rads generally less than a single inch.

    Here's an article from "Hot Tech Topics" regarding circulator sizing for gravity conversions. Taco 007 or B&G 100 are very frequently used.

    TRVs (thermostatic radiator valves) are a GREAT addition to gravity conversion systems. They're not free, but not particularly expensive and are a great if the budget allows.

    Mod-cons thrive driving gravity conversion systems. The radiation is typically so generously sized (made even more so by any later insulation/weatherization measures) and required supply temps will be quite low in most weather.

    Some form of primary/secondary is required if TRVs are not used with a mod-con. If TRVs are used, it's possible to use only a single (perhaps built-in) circulator with some mod-cons.

    Presuming the system still has its original open expansion tank (usually in the attic), you'll have to replace it with a closed compression tank. Plain (non-diaphragm) tanks were formerly used--and no problem using them now if you can find. If you do use a plain tank, install a B&G Air-Trol fitting to prevent gravity circulation moving air through the system and the tank waterlogging. If you use modern diaphragm tanks, check mfgr. literature carefully. Depending on the size of the system you may need multiple tanks. Impossible to have too large of a compression tank, but you can certainly have one that's too small...

    When you pipe in the new boiler here's a simple rule-of-thumb for sizing the new common piping. Take the largest pipe size in the mains. Cut in half then subtract one further size. For example, if the largest mains are 3", you would use 1 1/4" common piping. (If the returns currently join together into a single huge pipe don't count the size of the common pipe as the "largest in the system".)

    As to efficiency: You really will get the greatest efficiency using a mod-con. The only thing about the piping that affects efficiency is the "backwards" gravity pipe sizing that finds the highest, furthest rads with the greatest head loss and the lowest, closest with the least. The head losses are tiny on an absolute scale, but on a relative scale 2nd floor rads will typically have at least 2x the head loss of ones on the 1st floor! To keep reasonable balance through the system this means you have to move significantly more water than originally moved under gravity--by increasing the flow rate considerably you add head loss to everything thus lowering the relative difference between upper and lower rads.

    The significantly increased flow rate means significantly reduced delta-t which means increased return temperature. Mod-cons prefer the lowest possible return temp. TRVs on all radiators will keep the flow rates very similar to what they were under gravity without any balance problems. The increased return temp without TRVs isn't a huge issue with regards to efficiency of a mod-con but it's at least modestly significant.

    DO NOT OVERSIZE THE REPLACEMENT BOILER--PARTICULARLY A MOD-CON!!!

    Do not size the new boiler by measuring the radiation! Do not size the new boiler using the rating of the existing boiler! Size the boiler using a room-by-room heat loss calculation. If Manual-J based, my personal opinion is that you can safely use a boiler with a gross output rating up to 25% less than the heat loss calculation. If at all possible do not choose a boiler larger than the loss calculation.
  • Brad White_9
    Brad White_9 Member Posts: 2,440
    What he said. (NM)

  • Kevin Pulver
    Kevin Pulver Member Posts: 67
    Let me see if I got it

    Guys... The outlets are 2" main on the left and 3" main on the right. I think the returns are 1 1/2" at the boiler.
    The system was balanced originally, but since we are adding a low-mass boiler we must move more water for the boiler's sake. Moving all this extra water adds head which throws the system out of balance even though the piping is the same. Correct?
    So the TRVs would allow us to re-balance besides offering a high level of zoning control. Correct? The part I don't understand is why I could POSSIBLY delete the pri-sec if we have TRVs?
    And if I don't use TRVs, would I need to put balancing valves on one or both of the supplies?(the 2" & 3")
    I understand clearly what you said about sizing the boiler and expansion tank and what an Air Trol is. I've read all Dan's books, and have the I.B.R. binder but sometimes I need a tutor besides. Thanks for your patience! Kevin
  • Steamhead (in transit)
    Steamhead (in transit) Member Posts: 6,688
    Kevin, since you say

    "but since we are adding a low-mass boiler we must move more water for the boiler's sake"

    That's as good a reason as any to use P/S. The flow thru the boiler is then independent of that in the system.

    To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"
  • Kevin O. Pulver
    Kevin O. Pulver Member Posts: 380
    Correct Steamhead.

    I understand the WHY of Pri-Sec and the part I don't understand is why TRV's would make any difference in regards to Pri-Sec. I must not be understanding correctly something Mike was trying to say. I just posted a diagram in a new thread called, "save the radiators! Diagram." Thanks for all input. Kevin
  • Mike T., Swampeast MO
    Mike T., Swampeast MO Member Posts: 6,928


    The connections you describe sound unusual. I can see the 2" supply on one side and the 3" on the other, but 1 1/2" returns sound truly goofy.

    How many square feet EDR of radiation? I'm going to guess somewhere around 700 but that's just a wild guess...

    -----------------------------------------------

    Regardless, I'll reply to your message:

    since we are adding a low-mass boiler we must move more water for the boiler's sake

    You have it backwards--VERY easy with gravity conversions--believe me!!!

    You're moving more water for the PIPING'S sake. The low-mass, mod-con boiler is EASILY satisfied with a flow rate following the old rule of thumb of 10,000 btu/hr per gallon per minute with a 20 degree F temperature drop. The typical Taco 007 or B&G 100 connected to a gravity conversion will EASILY move 20 gpm--and likely more. If this system is anywhere near the size I suspect, it likely only moved 5-6 gallons per minute as originally designed with the old solid fuel manually modulating boiler. When you convert to forced flow you need the extra flow to overcome the goofy gravity piping. This is where it's VERY easy to get things backwards!

    The part I don't understand is why I could POSSIBLY delete the pri-sec if we have TRVs?

    Use the TRVs and a SINGLE circulator with a automatically modulating boiler and you'll wind up with flow rates VERY similar to the original. The system is almost certain to have ZERO problem operating at 30 or even 40 degree F delta-t during periods of high load or setback recovery--just like it did under gravity. 6 gpm with an 80,000 btu/hr or so input mod-con boiler will be ABUNDANT flow for a TRVd gravity conversion (provided there are no later goofy modifications like a zone of fin baseboard).

    Compared to the old ARCO boiler, a mod-con has EXTRAORDINARILY high head loss. ALL mod-cons can be directly connected but I've yet to see a U.S. manual with instructions for such. Nearly all however do give head loss for the boiler itself. With a TRVd gravity conversion using a single circulator just consider the head loss of the boiler plus about 6". Contact the mod-con manufacturer. Tell them it's a gravity conversion using TRVs and follow their advice.

    ------------------------------------

    If I've rambled too much or not stated things clearly, please ask further. TRVd gravity conversions driven by mod-cons are my true passion.



  • Brad White
    Brad White Member Posts: 2,393
    I see the TRV part

    as being essentially balancing devices, indirectly balancing the system by trimming back the satisfied rooms. With constant circulation to the radiators, life will be very comfortable IMHO.
    "If you do not know the answer, say, "I do not know the answer", and you will be correct!"



    -Ernie White, my Dad
  • Dave Stroman
    Dave Stroman Member Posts: 761


    Check out this old coal converted beast. I clocked the meter at a whopping 650,000 Btu per hour. I installed a 200,000 boiler for the main house and a 150,000 for the carriage house, and even they were larger that was really needed.

    Dave in Denver
    Dave Stroman
  • Mike T., Swampeast MO
    Mike T., Swampeast MO Member Posts: 6,928


    And in an original gravity system where you have essentially zero head loss combined with the ****-backwards pipe sizing TRVs are the only reasonable way to restore the original flow rate. A mod-con takes the extra step of restoring the original control method--automatically!
  • Steamhead (in transit)
    Steamhead (in transit) Member Posts: 6,688
    Oh yes, thee olde Spencer

    we replaced one of those last fall. Nice boiler in its day, but both had horrible conversion jobs done to them.

    To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"
  • Dave Stroman
    Dave Stroman Member Posts: 761


    Here is a shot of the burner. And look, this boiler may have been converted to oil before it got the gas.

    Dave in Denver
    Dave Stroman
  • Greg_40
    Greg_40 Member Posts: 43


    I've done a few conversions of that exact model boiler using mod/cons each time. The owners experienced some noticeable differences afterward. One- prior, the ARCO was virtually silent, now they hear circulators, either through the floor or transmitting via piping. Two- fuel bills are greatly reduced. Three- home warms up faster and they all say 'feels more comfortable', but I can't substiantiate the last one. P/S piping will be necessary and I always installed a screen Y-filter in the return just in case of some crud breaking loose in the 100 yr. old piping network. The most difficult thing we experienced was breaking the old fittings apart. Finally had to cut and weld in reducers at one project, just could not break loose the old flanges.
  • Kevin Pulver
    Kevin Pulver Member Posts: 67
    EDR RATING

    Mike, your "wild guess" is awful close! You said 700, and Dan's book says the 5 2nd story Rads total 265 and the 5 on the main floor total 356 for a grand total of 621 EDR.
    As far as I can tell, nothing has been added or removed. Do you have a favorite brand of TRV? And where is a good supplier? I have never bought any and don't know if my local guys even have them. Lots of times, they don't even know what I'm asking for-even if I give them a part no.
    And are you saying that if I have TRVs and the boiler manufacturer blesses it; that I would need just one circ sized for the boiler HX plus 6' and no primary-secondary necessary? Thanks, Kevin
    Thanks, Kevin
  • Kevin Pulver
    Kevin Pulver Member Posts: 67
    The Y strainer

    sounds like a really good idea Greg. Definitely worth the money. Thanks, Kevin
  • Steamhead (in transit)
    Steamhead (in transit) Member Posts: 6,688
    It probably was

    here's the Spencer we pulled out. When I first worked in that house, I saw the original oil burner had a gas pilot for ignition. The drop pipe was still there. By the time we replaced it, it had a Beckett SF which was a poor match for the old firebox. No one bothered to try to replace it.

    To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"
  • Mike T., Swampeast MO
    Mike T., Swampeast MO Member Posts: 6,928


    I've only used Danfoss TRVs. Nobody stocks them locally so I order from Blackmore & Glunt in St. Louis. I don't know if you're anywhere near, but their phone number is (314) 878-6028. I won't mention price other than to say that I once had to special order a standard hand valve from a local plumbing house and it cost me more than a TRV body!

    The "Side-Mount Angle" body style is almost always used with standing iron rads. They come in 1/2", 3/4", 1" and 1 1/4". Unlikely that you'll have 1 1/2" or larger in that system unless there's a monster of a rad very close to the boiler.

    The TRV body is a direct replacement for a hand valve or angle union. Of course you have to remove the old tailpiece as they won't mate properly. Personally I just remove the tapping bushing as well as the tailpiece and put the new tailpiece in a new bushing. Sometimes the tailpiece will come out with a spud wrench and some people cut out the tailpieces. Working by myself I can usually install a TRV in about 15-20 minutes. I'm pretty good at manhandling even the largest rads, but a helper certainly makes the job easier--but not much faster.

    TRVs MUST be installed on the supply. Hand valves are usually--but not always--on the supply. Verify the piping! If you install on the return the valve will bang loudly and rapidly once it tries to throttle down and it MUST be relocated to the supply.

    Count on needing about 8" from the edge of the radiator (not the tapping bushing) for a the side-mount angle TRV body and self-contained actuator. If a wall or door is in the way there's a couple ways to avoid moving the rad. A remote-operating actuator will remove a touch over an inch. You can also use an angle valve body but this means the operator is vertical and a remote sensing operator MUST be used as the self-contained operators MUST be installed horizontally. Using the angle valve will keep the overall project very similar to the original.

    A very strict interpretation of Danfoss literature shows that a remote sensing operator should be used if the TRV is on the bottom rad connection, but many use the self-contained in this position and I haven't heard of problems. The remote operator and/or remote sensors are connected via tiny capilliary tubes--not wires--so keep that in mind. Despite their tiny size they seems amazingly robust. While it's rare to find the supply connected to the top of a standing iron rad (except with a downfeed system), such is truly an ideal location for the TRV and self-contained operator--not only for sensing room air temp but for operation without stooping to the floor.

    If the rad is under a window and the operator will be covered by a drapery, you MUST use a remote sensing operator. If the window is at all drafty you're well-advised to use a remote sensing operator. A remote sensor installed on an uninsulated outside wall MUST have a piece of rigid insulation installed behind it. What all this means is that the sensing element MUST be able to sense room temperature in a reasonably representative area. It cannot be obstructed nor can it be exposed to drafts.

    And yes, install TRVs on ALL radiators and you'll be able to use a single circulator with a mod-con. Every manual I've reviewed for mod-cons introduced in the last two years or so alludes to this ability but does not give instructions. Do however consult with the manufacturer for their specific recommendations. If they say it can't be done, I'd say to choose a different manufacturer. (I'd shy away from aluminum heat exchangers in an old gravity system.)

    That wasn't 6 feet of head I said to add to that of the boiler--it was 6 inches. On further review that recommendation could be a bit shy in some circumstances. 2' for the TRVs themselves plus whatever head for the new near boiler piping will be generous. Again, no need to compute head loss in the gravity piping since at these flow rates (both in mains and branches) it will be so close to zero as to be inconsequential. And by the way--if it's more convenient to bring the main pairs together via a "bullhead tee", do so--with the TRVs it won't matter in the least.

    Do not forget to install a differential bypass valve in the new near boiler piping! This MUST be used to protect the circulator in the event that all or nearly all of the TRVs are closed. The easiest way I've found to adjust the differential pressure bypass is by ear. I use Braukman (Honeywell) valves. They have an indicator but head loss in gravity systems (even with TRVs) is so darned low that I've NEVER seen an indication--even during "torture tests". With the system maintaining temperature in the house during at least "cool" (say 30s) weather, you want NO bypass. Then set all but 2 or so TRVs all the way down and adjust until you hear a fair amount of bypass. Then return the TRVs to the normal setting and listen again--if you hear even a hint of bypass, use less.

    You'll want your circulator to run constantly anytime that the house needs heat. If the boiler has thermostat connections (e.g. it's not a Vitodens), there are a number of simple ways to control. The easiest is to merely jump the T-T connections. This will be OK if the boiler itself provides a warm-weather shutdown function. If it doesn't have a warm-weather shutdown function you can simply use a setpoint controler with the sensor OUTSIDE as the "thermostat" with it closing the T-T connection anytime the outside temp is below some point (55F is often a good starting temp--know it sounds cold, but it usually works well).

    You can also install a traditional thermostat in a central location. During "normal" operation the thermostat setting will be HIGHER than needed anywhere in the house. This will produce a constant call for heat. This is a very simple way of instituting whole house setback in a fully TRVd system--merely turn down the thermostat and the house will cool until it reaches setpoint. Again, if the boiler itself does not have a built-in warm-weather shutdown function, I'd highly suggest putting an outdoor temp sensor in series with the wall thermostat--this will automatically shut down the system during moderate weather.

    If the mod-con has an "auto-boost" function (many do), it MUST be disabled in such a system. The last thing you want the boiler to do is keep jacking up the supply temperature when you've intentionally produced a constant call for heat!!! Every manual I've read finds "auto boost" on by default, so be careful--fortunately every manual also states that it can be completely defeated.

    If the boiler is a Vitodens you'll merely use the built-in circulator, and there's utterly no need to use the "room temperature sensing" RS remote control. The WS remote (it merely provides remote adjustment for the sun and moon dial settings) can be used or the entire programming unit can be remote-mounted to operate in the WS (not RS) mode. The 6-24 (smallest) Vitodens will certainly have enough residual circulator head for the system. I SERIOUSLY doubt you'd need the 8-32 unless the climate is utterly brutal and the house more like a barn, but even then the built-in circulator would most likely have sufficient residual head. If the heat loss calls for a Vitodens larger than the 8-32 then something is VERY wrong with the heat loss calculation given that amount of radiation!!!

    When it comes time to set the reset curve, my best advice is to "start low" and work your way up if necessary. A curve with 0.8 slope and 0 shift should prove sufficient--for most gravity conversion systems provided little or no daily setback is used. If the homeowner appears at all capable and enthusiastic, give them instruction on how to adjust the reset curve. (The Vitodens appears unique in giving the user access to and well as reasonable instructions in the user manual for adjusting the curve.) Particularly with TRVs, you should wait AT LEAST two days between curve changes--longer if you're "homing in" on an efficiency-idealized curve that produces barely adequate supply temps.

    Sorry if this is so long, but it could be much longer... Nothing particularly difficult about turning an old gravity system into one whose only rival for efficiency and comfort is highly conductive radiant floors, but it does take a somewhat different way of thinking.
    Murphy38
  • Kevin Pulver
    Kevin Pulver Member Posts: 67
    Excellent information Mike

    I appreciate it. So your recommendation is TRVs with a pressure differential bypass and single circ, no pri-sec.
    (consulting the boiler folks in Tenn. of course)

    If the budget doesn't allow that, go with Pri-Sec, and balancing valves on the 2 returns.
    The former would allow better control and slightly better efficiency due to lower return temps.
    The latter won't balance out like the gravity did. Do I have this right?
    It almost makes me think that the "right" way is with TRVs. Am I overstating that?

    All your trivia about installation and mounting is wonderful. If I get the job, I'll pipe it in black and post photos. (And I'll save what I can for souvenirs)Thanks again, Kevin


  • What I dont understand is why you can pipe direct with a single circulator when using the TRVs, but must use pri-sec piping if no TRVs are used?

    If the room temp is below the setting of the TRVs then they are open and the system would circulate just as if no TRVs were present. If the head loss of the gravity system was so low as to not be a limit to flow, why would you need the pri-sec in either case?

    I understand the need for pri-sec piping with a modcon so that the system piping to the radiation does not limit the flow rate through the heat exchanger. But if the radiation piping effectively produces no head loss, and the modcon heat exchanger is the sole cause of the flow restriction, why do you need pri-sec circulation?

    While a system with TRVs is probably the ultimate, a gravity conversion can work really well without them. Even if the flow characteristics change with forced circulation, you still have possibilities for adjustment. Besides the two balancing valves on the returns, you can always adjust the individual radiator valves to compensate for the different flow.

  • Mike T., Swampeast MO
    Mike T., Swampeast MO Member Posts: 6,928


    For ANY gravity system still operating under gravity my HIGHEST recommendation is for TRVs on ALL radiators and a directly connected mod-con with stainless steel HX and a SINGLE ciculator. (It's bastardized original gravity systems that have been converted to forced flow with wildly inappropriate "zones" like fin baseboard that are a problem...)

    In all honesty I believe you can tell the homeowners to expect at least a 50% reduction in fuel costs (as long as they don't drastically change their habits and keep things significantly warmer because of the fancy, new, efficient system).

    Gravity hot water systems were BY FAR the most extravagent and expensive heating systems in common use at the time. I'd compare them to the finest modern radiant floor systems in this regard. Overhead gravity hot water systems were the Rolls Royce of Cadillac systems. Only indirect systems were more extravagant and only the VERY rich could afford to install and operate them... Gravity hot water was ALWAYS known for being fuel-efficient, clean, comfortable and controllable--provided the solid-fueled modulating fire was well-controlled.

    In my opinion you are NOT overstating that this is the "right way"!!!

    Add the TRVs and a directly connected mod-con and you retain the Cadillac nature of the original system--anything less and you can well end up with a doggy Dodge.

    If TRVs utterly cannot be fit into the budget, I'd STILL use a mod-con. Use a low-loss header--NOT traditional primary-secondary! In this case where you have different sized supply mains (they're usually the same size) and what sounds like strange return sizing, I'd say that the "balancing valves" in the return are a prudent addition--but do find high-quality valves that are actually made for this purpose. Adjusting them might prove difficult--I really have no idea.
    Murphy38
  • Mike T., Swampeast MO
    Mike T., Swampeast MO Member Posts: 6,928


    Everything I've said in this thread is in the context of a original gravity system. The system in this thread can VERY likely be perfectly satisfied with 4-5 gpm of flow under the highest load--VERY similar to the orginal gravity flow.

    Try to move only 4-5 gpm of forced flow through a system designed for gravity flow and I can nearly guarantee severe balance problems in a multi-floor system. The pipe sizing is literally backwards. The forced water will favor the path of least resistance--the lowest, closest rads.

    Modern systems can certainly be designed for a directly connected mod-con, but I'd still be reluctant to do so without the TRVs unless you're heating one or two logical spaces--not the many rooms of typical homes.
  • Kevin Pulver
    Kevin Pulver Member Posts: 67


    Mike T,
    Mike Ks question made sense to me also, but reading Dan's gravity stuff has helped me understand I THINK.

    YOUR QUOTE:Try to move only 4-5 gpm of forced flow through a system designed for gravity flow and I can nearly guarantee severe balance problems in a multi-floor system. The pipe sizing is literally backwards. The forced water will favor the path of least resistance--the lowest, closest rads.

    SO: Gravity flow and Forced flow are nothing alike, and the piping that works FOR you in Gravity will work AGAINST you in Forced flow. Therefore, proper balance will be extremely unlikely even with original rad-valves.
    Hence the TRVs. Correct?
    So, if the budget rules out TRVs, we can still do the job, but advise them accordingly? And TRV's could be added in the future, but a differential bypass valve would be needed if there is no Pri-Sec. And the reason you favor ONE circ over pri-sec is a little lower return temps. Correct? And instead of just Pri-Sec piping you'd opt for a low loss header like the Caleffi, right? Thanks, Kevin
  • Mike T., Swampeast MO
    Mike T., Swampeast MO Member Posts: 6,928


    EXACTLY:

    Gravity flow and Forced flow are nothing alike, and the piping that works FOR you in Gravity will work AGAINST you in Forced flow. PERFECTLY STATED!!!!

    Utterly impractical to use the original hand valves to "balance" the heat when it's converted to forced flow. Believe me--I've tried. Even the little "anti-freeze" hole in the hand valve typically allows enough flow to fully heat a rad when it's converted to forced operation!
    If you do manage to achieve balance by carefully adjusting the hand valves all you have to do is move ONE and the balance through the entire system is shot! The hand valves were designed to balance gravity flow--not forced flow! Look at the guts of one and you'll see what I mean.

    Don't consider a mod-con using a single circulator without TRVs in an original gravity system. The only possible exception is for a gravity system serving a single floor.

    Yes, TRVs could be added later to a system with a low-loss header or traditional primary/secondary. A differential pressure bypass valve will be required but you might as well retain the low-loss header unless you just want to eliminate it and the additional circulator.

    Yes, if you directly connect the boiler to the TRVd system via a single circulator you WILL have the lowest possible return temperature as you'll be able to keep flow as low as possible. (This gets a bit dicey as flow through a constantly circulating TRVd system is inexorably linked to the reset curve as modified by the differential pressure bypass, but as long as you have a reasonable reset curve there won't be a problem.) I have numerous and long white papers devoted solely to differential pressure bypass in constantly circulating, reset systems...

    If you don't use the TRVs, DO use a low-loss header (not traditional primary/secondary) with a mod-con on a gravity conversion. Secondary (emitter) flow will be MUCH higher than primary and the low-loss header will guarantee that return temperature is as low as possible under the conditions at the time.

    If you use a conventional boiler, I'd suggest use of an ESBE type TV thermostatic bypass--still with a single circulator. This gives positive, automatic protection against low return temperatures without the need for primary/secondary and two circulators. It's kind of like the bypass line Dan mentions in his book "How Come" (regarding gravity conversions) but it's thermostatically controlled.

    Don't forget. Many of these comments and recommendations are based on an original gravity system that hasn't been screwed up with later modifications (other than simple conversion to forced flow).



  • Kevin O. Pulver
    Kevin O. Pulver Member Posts: 380
    I think I've got it!

    > EXACTLY:

    >

    > _I_Gravity flow and Forced flow are

    > nothing alike, and the piping that works FOR you

    > in Gravity will work AGAINST you in Forced

    > flow._/I_ PERFECTLY STATED!!!!

    >

    > Utterly

    > impractical to use the original hand valves to

    > "balance" the heat when it's converted to forced

    > flow. Believe me--I've tried. Even the little

    > "anti-freeze" hole in the hand valve typically

    > allows enough flow to fully heat a rad when it's

    > converted to forced operation! If you do manage

    > to achieve balance by carefully adjusting the

    > hand valves all you have to do is move ONE and

    > the balance through the entire system is shot!

    > The hand valves were designed to balance

    > _I_gravity_/I_ flow--not forced flow! Look at

    > the guts of one and you'll see what I

    > mean.

    >

    > Don't consider a mod-con using a single

    > circulator without TRVs in an original gravity

    > system. The only possible exception is for a

    > gravity system serving a _I_single

    > floor_/I_.

    >

    > Yes, TRVs could be added later to a

    > system with a low-loss header or traditional

    > primary/secondary. A differential pressure

    > bypass valve will be required but you might as

    > well retain the low-loss header unless you just

    > want to eliminate it and the additional

    > circulator.

    >

    > Yes, if you directly connect the

    > boiler to the TRVd system via a single circulator

    > you WILL have the lowest possible return

    > temperature as you'll be able to keep flow as low

    > as possible. (This gets a bit dicey as flow

    > through a constantly circulating TRVd system is

    > inexorably linked to the reset curve as modified

    > by the differential pressure bypass, but as long

    > as you have a reasonable reset curve there won't

    > be a problem.) I have numerous and long white

    > papers devoted _I_solely_/I_ to differential

    > pressure bypass in constantly circulating, reset

    > systems...

    >

    > If you don't use the TRVs, DO use a

    > low-loss header (not traditional

    > primary/secondary) with a mod-con on a gravity

    > conversion. Secondary (emitter) flow will be

    > MUCH higher than primary and the low-loss header

    > will guarantee that return temperature is as low

    > as possible under the conditions at the time.

    > If you use a conventional boiler, I'd suggest use

    > of an ESBE type TV thermostatic bypass--still

    > with a single circulator. This gives positive,

    > automatic protection against low return

    > temperatures without the need for

    > primary/secondary and two circulators. It's kind

    > of like the bypass line Dan mentions in his book

    > "How Come" (regarding gravity conversions) but

    > it's thermostatically controlled.

    >

    > Don't

    > forget. Many of these comments and

    > recommendations are based on an _I_original

    > gravity system that hasn't been screwed up with

    > later modifications (other than simple conversion

    > to forced flow)_/I_.



  • Kevin O. Pulver
    Kevin O. Pulver Member Posts: 380
    I think I've got it!

    Your thoughts on the original hand valves and balancing are very clear.
    It seems to me that I either need TRVs OR a low-loss header. I have 10 rad locations. By the time I buy a low loss header and another circ and flanges, that is eating into the "savings" of no TRVs. I think you have me converted.
    I'm sure others could post experiences of how it will "work" another way, but this seems like the "best" way to me.
    I'm open to suggestions, but this makes sense, and it's not theory with you. I really appreciate it Mike! Kevin
  • Mike T., Swampeast MO
    Mike T., Swampeast MO Member Posts: 6,928


    You're quite welcome. I'm almost certain that the final cost using TRVs will be higher than if you make a simple conversion--but almost certainly less if the customer wants conventional "zoning" that would require extensive re-piping.

    I've installed TRVs on three gravity systems--one mine. Only mine uses a mod-con (the other two had boilers similar to the ARCO). ALL were exceptionally pleased with the comfort. All saw some fuel savings--I saw EXTREME fuel savings by adding the Vitodens. My Vitodens was installed after 3 or 4 heating seasons with the TRVs, constant circulation, mechanical reset, T-T connections jumped, and warm-weather shutdown so I have a very good basis for comparison for just the change in boilers.
  • Mike T., Swampeast MO
    Mike T., Swampeast MO Member Posts: 6,928


    Mike:

    Have you actually connected a mod-con to a multi-floor gravity system using only a single circulator?

    If so, which mod-con and what circulator?

    I have not tried. All of the numbers seem to say "don't even consider".


  • No. Actually I have never seen a modcon installation that did not have some form of P/S circulation. With the exception of the Vitodens, all of the modcon installation manuals I have read recommend or require P/S. My earlier question was more addressed to why TRVs made P/S less necessary, rather than suggesting the use of a single circulator was the best solution.

    Typically the modcon manufacturers concern is to insure adequate heat exchanger flow, which might be limited by the head loss of the radiation and piping. As you say, a gravity conversion is a different situation, where the heating circuit is not the limiting factor of flow. Rather, the heat exchanger is the flow limiting factor, with the piping and radiators being capable of and requiring larger flow rates than the exchanger may allow.

    I think your suggestion to use a low loss mixing header makes a lot of sense, whether you use TRVs or not. With the header, both the boiler and the heating circuit may each operate at their optimum flow rate.
  • Mike T., Swampeast MO
    Mike T., Swampeast MO Member Posts: 6,928


    Both the Triangle Prestige and Munchkin Contender manuals state something along the lines of, "Primary-secondary is shown in all of the examples, but is not always necessary. Contact us."

    Again, it's the goofy gravity pipe sizing that can cause problems and require a secondary flow rate MUCH higher than the primary rate required by the mod-con. The significantly higher secondary flow causes extremely low delta-t in the secondary which in turn results in an elevated return temperature.

    As long as secondary flow is greater than primary through a low-loss header, the LLH guarantees the minimum return temperature to the boiler, BUT that minimum return temperature is exactly equal to the return from the secondary. If the greatly increased secondary flow has already resulted in a substantially raised secondary return temp there's no to get a lower return temp to the boiler.

    The situation is even worse if you've used TRVs and a low loss headed and the primary circulator operates at a fixed speed. In this case, the TRVs are regulating the secondary flow and in many cases it will be lower than primary. When this happens flow through the LLH goes in reverse with the extra primary supply shortcutting its way right back to the primary return thus again increasing boiler return temperature!

    With TRVs and without the low-loss header (or primary/secondary) the entire system has proportional flow even when one single-speed circulator is used. As long as the resulting flow is within the acceptable range for the boiler used, there's simply no need for a low-loss header--in fact it becomes counter-productive.

    Modern systems very frequently suffer from a relative deficit of emission potential. In other words the emitters cannot instantly liberate the the full output of the boiler to maintain room temperature in even the most brutal weather. If a conventional boiler, it has to cycle--if a mod-con it has to modulate down.

    Gravity systems are very unlikely to suffer that problem as the radiation is typically VERY generously sized. Any later insulation/weatherization measures in the structure only add to this relative oversizing of the radiation. Size the boiler properly to the heating load and you can most likely run the boiler flat-out for hours if not days without ever reaching the system high limit.

    I once ran my 210,000 btu/hr input conventional boiler for about 10 hours straight to see what would happen. Weather was in the 50s and I had about 850 sq.ft. EDR of radiation (now have 1,049 sq.ft.). Opened all TRVs wide--and eventually many of the windows/doors as well. The supply temp pretty much stopped at about 167F and had only risen a degree or two in the last two hours of firing. Room temp was somewhere in the 80s.

    Now, with a properly sized mod-con (91 mbh max input), there's simply no rational way for the boiler to ever encounter a high limit problem. I'd have to torture it by adding an insane amount of positive shift to the reset curve while turning all of the TRVs way down and set the differential pressure bypass such that it would NEVER bypass.

    I've run the Vitodens in test (full input) mode for at least 30 minutes with only one TRV wide open (the rest satisfied or closed) and still didn't approach high limit.
  • Mike T., Swampeast MO
    Mike T., Swampeast MO Member Posts: 6,928


    [Correction to post beginning with "EXACTLY"]

    If the TRVs are added at a later date and you have a low-loss header and the primary circulator operates at fixed speed you will want to remove the low-loss header and secondary circulator. The existing primary circulator will probably be adequate for the entire system however.

    For the reason why, see a later message. It has to do with the TRVs producing a secondary flow lower than primary--when this happens you're back to the "increasing the return temperature" problem.
This discussion has been closed.