Need some help with an existing Radiant / Baseboard system

Constantin

... so I imagine that's why the local companies don't want to touch it.

If you could be so kind, perhaps now would be a good time to post a diagram of the system and what is connected to what.

I second using the Find-A-Pro service to see if any radiant experts are in your area.

ScottFP

Existing hydronic system is not functioning correctly

We just purchased our home in July, The previous owner had installed a radiant / baseboard system in the home and it had been used for 1 heating season only (last winter). When it was time for us to use the heat this year, the system components all seem to be working properly, but it was using a lot more NG than i would have thought it should. I checked the lines of the system loops when they were running, and the return lines are only about 1/4 the temperature of the supply lines (which are set to run at 160 degrees). I have contacted all the HVAC companies in our area and no one wants anything to do with the system, as they didn't install it. I am comfortable with doing the repairs / changes myself but i don't really know where to begin. I have read about everything that i can find on hydronic heating, and the problem, from what i can guess; is in the system design. Maybe the pumps are undersized? The baseboard part of the system is functioning great, it is just the radiant floors on the other 2 levels of the home that i'm having problems with. The pumps, 3 zones and the primary, are all Taco 007 models. All system copper is 3/4 inch, the radiant PEX is 1/2 inch, and the supply and return to the baseboards is 3/4 inch PEX. The radiant is a underfloor joist system with foil insulation, the loop lengths according to the markings on the PEX are all around 250 feet (give or take), and there are 4 loops per level, with 2 radiant levels total. I have attached a few pics of the system, the domestic hot water tank is not connected to the system, they are just very close in proximity, . I would like to salvage as much of the current system as possible, as we did just buy a new house. Any advice or assistance anyone can provide would be greatly appreciated as the colder months of winter are sneaking up.

Thank You,
Scott Berdine

Al Letellier_11

radiant system upgrade

> We just purchased our home in July, The previous

> owner had installed a radiant / baseboard system

> in the home and it had been used for 1 heating

> season only (last winter). When it was time for

> us to use the heat this year, the system

> components all seem to be working properly, but

> it was using a lot more NG than i would have

> thought it should. I checked the lines of the

> system loops when they were running, and the

> return lines are only about 1/4 the temperature

> of the supply lines (which are set to run at 160

> degrees). I have contacted all the HVAC

> companies in our area and no one wants anything

> to do with the system, as they didn't install it.

> I am comfortable with doing the repairs / changes

> myself but i don't really know where to begin. I

> have read about everything that i can find on

> hydronic heating, and the problem, from what i

> can guess; is in the system design. Maybe the

> pumps are undersized? The baseboard part of the

> system is functioning great, it is just the

> radiant floors on the other 2 levels of the home

> that i'm having problems with. The pumps, 3

> zones and the primary, are all Taco 007 models.

> All system copper is 3/4 inch, the radiant PEX is

> 1/2 inch, and the supply and return to the

> baseboards is 3/4 inch PEX. The radiant is a

> underfloor joist system with foil insulation, the

> loop lengths according to the markings on the PEX

> are all around 250 feet (give or take), and there

> are 4 loops per level, with 2 radiant levels

> total. I have attached a few pics of the system,

> the domestic hot water tank is not connected to

> the system, they are just very close in

> proximity, . I would like to salvage as much of

> the current system as possible, as we did just

> buy a new house. Any advice or assistance anyone

> can provide would be greatly appreciated as the

> colder months of winter are sneaking up.

> Thank You, Scott Berdine

Al Letellier_11

radiant system upgrade

Help is available, Scott, but where are you located. The photo shows some potential problems, but is too limited to say what needs to be done. Use the find a pro area of this site.
al

Mike T., Swampeast MO

What problems are you having besides what seems to be excessive fuel consumption?

Don't forget that radiant systems have bring lots of mass up to temperature. Unless you're in a really cold climate, the system has likely only been used sporadically this fall. Once true cold weather arrives and you need heat continually, things may change.

tom_49

Hey! How did a pic. from the halloween thread get here.

ScottFP

Responses

I'm located in central Pennsylvania, so the climate here is pretty cold in the winter.

I tried the find a pro locator, and the closest installer is located 3-4 hours away in York or Greensburg.

The system does run periodically as the temperatures haven't been really cold yet, but it runs for approximately an hour or so to raise the room temp 1 degree.

They originally had setback thermometers on the radiant loops with a setback at night going back to 60 degrees, it was taking all day to warm the temp back up, but after some reading i discovered that they were incorrect in the setbacks for radiant loops and that these can in fact make the system much more inefficient. So i changed it to setback only 1-2 degrees at night and that seemed to help.

I'm just concerned that when the temps do get really low around here the system either won't be able to keep up or will be running 20 out of the 24 hours a day.

I'll have to put together a diagram, as they didn't leave anything other than the manual for the water heater and the pumps.

I'm so glad to get some responses to this, As i said just purchasing a new house i'm not ready to have to totally replace the heating system, and hopefully this can be salvagable.

Thanks again,

Scott

PS... Diagram to follow

Constantin

Some thoughts...

A radiant floor system can take a very long time to recover, particularly if the layer above is thick or the insulation below is insufficient. It also depends on your heat loss, as you can imagine.

One thing that might help a bit is having the system on an outdoor reset controller and a mixing valve or injection system that allows multiple water temperatures, one high for the baseboard system, one lower for the radiant loops. The current water temperatures for the radiant seem to be set by manual mixing valves, which may account in part for the long response time.

I also wonder to what extent the boiler you have may or may not be short cycling. Thus, you may want to compare a heat loss calculation to the capacity of the boiler.

ScottFP

The PEX does all have the oxy barrier.

mixing valves - watts 70A

Fill valve - Sparco F449

Expansion Tank - Honeywell 4.4 gal

All Pumps - Taco 007

Switching relay - Taco SR 504

air seperator - Sparco model ??


The previous owner did leave a copy of the heat loss calculations for the house in with the instruction manual, I had our code and zoning office review them and also called Takagi, both said the water heater was large enough to space heat the whole house.

I have attached a diagram, sorry for the crudeness of it all but i don't really have a software package that will allow me to diagram a heating system. If there is anything that is unclear or needs more explanation; let me know.


Thank You

Tony_23

Yup

It's definitely screwed up.

70A valves are for domestic hot water and usually aren't suitable for space heating.

Your piping is neither primary/secondary nor direct or reverse return. If I am seeing and understanding your sketch correctly, there is 1/2" PEX after the "primary" circ back to the return. That will cause the Takagi to short-cycle for lack of flow.

Central PA ? Be more specific.

ScottFP

State College, PA

ScottFP

Should that 1/2" be 3/4"? or is the primary pump not even in the correct place at all?

ScottFP

According to the manual sheet for the Watts 70A it is the only mixing valve that they make that is suitable for radiant heating, and is not suitable as an anti-scald valve.

ScottFP

One more thing

As Tony pointed out my system is not a primary / secondary system according to any of the information that i have been able to find on these types of systems, i haven't been able to identify it with any other types that i have been able to find information about.

I have however in my research found a company that sells premade systems that are very similar in design:

http://www.aimradiantheating.com/

and

http://www.aimradiantheating.com/store/howitworks.html

Any ideas?

Thanks

Scott

bob_50

Scott

does the primary pump run whenever ANY zone is calling for heat? It should. I am not familiar with your boiler. Many instantanious water heaters firing rate is determined by flow rate through the boiler. You say the baseboard zone works fine. All the flow of that zone goes through the boiler all the time plus the flow through the primary. On the radiant zones only the quantity of supply water required to mix with return to maintain mixing valve set point is circulated through the boiler plus the primary if it's running. Just some ideas. bob

ScottFP

RE: Scott

The primary pump does run when any of the zones is turned on, the TK-Jr water heater is automatically fired when the flow rate reaches a certain point as you said; i think its .6 GPM or something like that.

Also i checked the mixing valves, and the thermostat assembly has been removed, so the full 180 degree water is flowing through the radiant loops. Don't know why they did this, but thats how things are running right now.

The system does heat the house, it just doesn't seem like it is working right on the radiant loops as i said the return temperature of the radiant loops is probably only 1/4 the temperature of the supply. Not enough flow? not sure whats going on here.

Thanks

Scott

Weezbo

it is sad that you have no zone valves...

however with 4 pumps, you could make a lot of header and a real sharp system i like the idea of oversizing the supply and return header.. and i like zone valves 4 way mixers and variable speed injection, any way ,...the designe is either heating the great outdoors or there are things which dont meet the eye. let me say that lash up is something i would re do .i like cool supply temps and i have to say that i dont see that
'switchy changing' these pieces about, as the ticket.

Tony_23

That's because

It's not any one type of piping design They've mixed up a combo of types and apparently didn't/don't understand what they've done, hence no understanding of what to do to correct it.

The elements being gone out of the 70A's doesn't make them work better, it makes a short circuit and bypasses much of the heated water from getting to the loop. If they weren't there and it was just a tee you would see what I mean. Sometimes, the fact that you're looking at the BODY of a mixing valve, won't let your mind see that it's no longer a mixing valve

Plumdog_2

Return temps are 1/4?

It seems almost impossible that the returns are 1/4 the temp of the supplies. If your supply is 180 then the return would be 45. I have attempted to help out a DIY installer who has copied an internet-sold system similar to what you have there; and I would stick my neck out here at risk of being offensive to some parties with underlying interests and just say that the design lacks in engineering soundness what it makes up for in cost of installation.
If you have tubing stapled under the floors, and carpet or hardwood above, then that is the biggest problem you have. The tankless heater senses both delta T and flow volume and ramps up the burner in an attempt to achieve the output temperature selected by the dipswitches or control. If I had to make one modification; I would add a fairly sizable buffer tank and mix/pump from it to the zones. I would use a fairly big pump thru the tankless (0011) and use a wide temperature differential on the tank control.

ScottFP

The components inside the 70A are gone, however the cup at the bottom is always closed without these so effectively it is only letting the hot supply through.

I was estimating the return temp at 1/4, they are definately way below 15 - 20 degrees different.

So since this system doesn't even really meet any type of system design criteria, should i disassemble some and try to make it a true primary / secondary system? or is there some other way that things can be repaired?

And as for the pumps, in your opinions are the 007's large enough to handle each loop and work as the primary?

The primary pump as it seems to me should be pumping through the system towards the secondary loops, not pulling whatever water is remaining in the supply, as i said i'm not a pro, these are just some questions that have come up while i have been looking at lots of info on designing and installing these types of systems.

As for the piece of 1/2" PEX running between the primary pump and the return, shouldn't that be 3/4" in diameter as not to limit the flow?


Thanks for the input from everyone,

Scott

Wayco Wayne_2

I lke

the buffer tank idea. To size the pipes correctly we would need to know the heat loads for each zone. 3/4 seems small for most primary loops. 3/4 can only handle 40,000 btuh at a 20 degree delta T. Look on the pex for numbers that tell you the length of tubing. 1/2 in pex should only go about 330 ft. Radiant floors are often designed at a 10 degree delta to keep hot and cold spots from happening. I'd pipe to a buffer tank and keep it at 160 and draw the other zones off the tank. Good luck. WW

To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"

Tony_23

Books & More

On the left of the page is a link. Go there, order "Primary/Secondary Pumping made Easy", read it. Then you will understand (I hope).

ScottFP

There are four loops per zone and they are all between 250 and 275 feet.

Dan_15

The primary pump should be just after the expansion tank pumping toward the zones. The return and primary loop is way to small. You need to figure out how many gallons per min. you need to supply the right btu's to the loops. take the longest loop of 1/2 in pex and multiply it by 0.0373 to get the head. then multiply the gpm by the amount of loops in that zone. to figure out what pump you need. example: 275 feet of 1/2 pen x 0.0373 = 10.25 so call it 11 feet of head. 4 loops @ 1 gpm = 4 pgm with 11 feet of head. I don't know what the 007 is good for. you could contact taco to find out. or check online. sounds to me like your water is to hot but, not enough gpm through the tubing. I'd make my primary loop out of black pipe 1 1/4 and have the zone supply header and return come off that. get the book pumping away it will help you alot. You might first try getting that primary pump moved over just after the expantion tank.
mike

ScottFP

If i calculate the head loss like you say,

268' longest loop * .0373 = about 10 with 4 loops @ 1gpm

and

236' longest loop * .0373 = about 9 with 4 loops @ 1gpm

and

100' total baseboard loop (3/4"PEX) * .00729 = about 1 with 1 loop @ 1 gpm

the 007's performance curve says its good for between 9 and 10 feet at 4 gpm, and between 11 and 12 for 1 gpm. So those pumps should be sufficient.

Wouldn't changing the size of the primary supply and return piping be limited right back to where it was with the boiler only having 3/4" inlet / outlet? and being a flash water heater, the tubing inside is even smaller.

Also, if i move the primary pump to right after the expantion tank, I should just bridge the supply and return lines there with the pump, or should i place the pump inline on the supply and where the pump was located; connect the supply and return lines?

I might not be calculating correctly, but the head loss of the system piping before the zones split off comes out to be about 3', about 70' of 3/4" type M copper (allowing for all the fittings) which the 007 is adequate for at 5 or so gpm. which should supply enough flow to the other 3 zones? This isn't counting any of the tubing inside the water heater though, and as i mentioned is much smaller than the 3/4" inlet or outlet. these are the parts that confus me the most i think. These calculations were done with the information from:
http://www.pmengineer.com/CDA/ArticleInformation/features/BNP__Features__Item/0,2732,116233,00.html


BTW i have ordered the book you suggested, maybe it will shed a little more light on what i'm apparently just not grasping.

Dan_15

My thought is that the primary pump should pump towards the zone pump. If the primary pump is after the zones I would think they would be competing for the water. You'll love the book. good luck.
Mike

Dice

heat problems

How many thermostats do u have?

ScottFP

Thermostats

i have 3 thermostats, it is a split level home with 3 levels and a basement, 1 thermostat per heated level.

Jamie_5

I am only a DIYer, so take what I say with at least a grain of salt.

If I am understanding correctly, you need two 4GPM zones of 160F water for the radiant (it seems you are saying you have 4 loops in each of two zones, each loop of which requires 1 GPM) and one 1GPM zone for the baseboard, presumably using 180F water. I am guessing that the head loss of trying to push 9GPM through the water heater would be very high.

However, allowing for only a 10F drop across the 160F zones, I calculate you would need only 3.6GPM of 180F water if your system were piped primary-secondary with parallel crossover bridges for the secondary circuits. (2.7GPM of 150F water combined with 1.3GPM of 180F water is 4GPM of 160F water. You would need two such supplies of 180F water for the radiant zones (2 * 1.3 GPM = 2.6GPM) and a 1GPM supply for the baseboard (2.6GPM + 1GPM = 3.6GPM).) A 3/4" piped primary circuit with three 1/2" cross-over bridges would handle those flows, and I think a reasonable pump would be able to push that much water through the water heater's heat exchanger. (If your radiant zones are designed with a 20F drop across them, each radiant zone would need 2GPM of 180F water, and those cross-over bridges would probably have to be 3/4" pipe. The primary loop flow would then rise to 5GPM, which might be better handled by 1" pipe.)

Looking briefly at your diagram and the photo, it doesn't look like it would be too hard to change it to a parallel primary-secondary system. The pump currently marked as "primary" would become a true primary pump, and the return portion which is currently 1/2" PEX would need to be changed to 3/4" copper. Then the supplies and returns between the zones and the pseudo-primary would have to be replaced by 1/2" cross-over bridges. (NB: These sizes are based on the assumption of a 10F temperature drop across the radiant zones.) The zone supply and returns would be installed as closely-spaced tees on the new bridges. You could go with either manually set or thermostatic mixing valves on the two radiant zones. You would probably have to include some way to balance the flow across the three cross-over bridges, though that might be do-able with just ball valves. Judicious placement of thermometers to monitor the return temperatures would let you adjust manual valves.

The professionals here may point out serious flaws with my reasoning, but that's my attempt at fixing your piping.

I would have two remaining concerns. One is that I don't know if you can use that water heater to produce high temperature water with a relatively small temperature drop of less than 30F. Another is that you appear to have suspended tube radiant with an unknown quality of insulation to keep the heat in the joist bays, since I am not sure what you mean by "foil insulation." The heat from the tubes may not be going where you need it to, i.e. through the subfloor and flooring above the tubes.

I hope this helps. The best of luck to you.

ScottFP

Very high head through water heater

You are correct about the flow through the water heater causing a lot of head loss, i just found some specs on the water heater and it looks like the max flow is 6 gpm, with around 45' head at this rate. The 007 pump is only going to handle about 9' head at 6 gpm. I have attached a photo. If the flow were lowered to around 3.6 gpm then it looks like the 007 still isn't quite large enough?

Does anyone else have any opinons on Jamie's suggestions on how to fix the system? I'm very willing to change whatever needs to be changed to get the system running properly, i just need some input on what those changes should be specifically. I really appreciate all the great information that i have gotten so far, this is a wonderful messageboard for people in a situation like mine. I'm greatful that there are so many people out there willing to help others out with some of these issues.

Thanks again,

Scott

Jamie_5

pump

Hopefully one of the professionals will chime in here. I think you are on the right path to be worrying about flow rate in the radiant zones, given how restrictive that heat exchanger is. However, I am a tyro. I have the luxury of being able to back up and try again if I am wrong.

You said the previous owners left the heat loss calculations. What are they? Basically, you have to know what you need in the way of output per square foot for the radiant zones in order to know if there's any point to trying to salvage what you've got. As one other poster put it, if you've got carpet above and any other difficulties, such as too wide a tubing spacing or insufficient insulation of the joist cavities, trying to fix what you have may well be for naught. If you can provide more information, I can look in the Wirsbo design manual I have to see if there's hope. Do you have tubing 8" on center, i.e. two tubes per joist bay? What's the insulation below? And what is the flooring above?

If, on the other hand, the required output is low enough so that you can get away with flow below 4GPM in the primary loop, it looks to me like a Taco 009 would do the trick for a primary pump. (My Wirsbo design book says that 275 feet of 1/2" tubing running 160F water at 1GPM would, by itself, produce nearly 9 feet of head. A 007 looks like it moves about 5GPM at that head, so the existing zone pumps should work.)

ScottFP

Today was the coldest day that we've had so far this year, and the system has been running almost non-stop. I don't know how the previous owner even made it through one season without having to change something. I have to get this thing in some kind of working order very soon.

I do have 2 tubes per joist (1 loop of tubing), with some type of foil insulation a few inches from the subfloor, and a bat of R19 insulation directly below that. The one level of the house is all carpet, and the other is mostly laminate (fake hardwood).

I have attached a copy of the heat loss summary page.

It would be easy enough to connect the zone returns right next to the zone supplies making the T's closely spaced as in a true primary secondary system, and moving the primary pump would require a little reworking but i don't think that would be that big of a problem either. We've pretty much established that the primary pump is not adequate to push through the water heater, so that will need to be changed; exactly what model i'm not quite sure yet as i think it would depend on exactly how the rest of the system has to be replumbed.

If any of my statements here are incorrect please let me know, and if anyone has any other suggestions they would be greatly appreciated. As i said the days here are getting much colder much quicker, and now is probably the time that things need to be changed before all on winter hits, and i find out that the system is not going to adequately heat the house.

How exactly do i control the flow rate that i want in the system? if everything were to be reworked to these stipulations, how do i know / control what the flow rate is?


If any other information is needed, please let me know.

Thanks,

Scott

Jamie_5

heat output

Well, if your carpeted area is also the living room area, I think you're going to be out of luck. That area needs 23 BTU/hr./sq. ft., and that's assuming every square foot of floor is available radiant space. The Wirsbo design manual I have says that in a joist heating application like yours, 23 is the maximum output you can get at 160F supply temperature and enough flow to only have a 10F drop across the circuits when you have a 3/4" plywood subfloor and a floor covering of only R=0.5. That is about the R value of 1/2" wood floor. If the carpeted area is the bedrooms, you *might* be able to do it, *if* you have a thin enough carpet and solid rubber pad. Otherwise, I think you're looking at having to start again with a better design.

As for moving the supply and return lines, my proposed solution was based on having 180F water at all three zones. If you simply move the returns next to the supplies, each zone will be in series with the last across the primary loop and will impose a temperature drop on the supply water to the next zone. That is why I suggested crossover bridges: the zones would be in parallel and all see the same supply temperatures.

And all of this is assuming you redesign the fake primary loop to act as a true primary loop and can get adequate flow through it.

On flow control, it is a combination of pipe sizing and valving. Balancing valves, globe valves, or even ball valves might be appropriate, depending on the rest of the design.

It looks like it might be time to pay for some design help. There are heating professionals who will do consulting and/or design for a fee, perhaps someone on this board. I'm installing radiant ceilings and floors in our house, and I paid a person in Wisconsin to check over my numbers and paid someone else (local) to install the boiler, near piping, and indirect. It isn't inexpensive, but probably better than the alternatives.

I hope this helps.

ScottFP

The carpeted area is the bedroom level, and the way it is now the mixing valves aren't really doing anything; so the water temp at each zone is the full 180 degrees. Would placing the return lines on the primary loop right next to the supply lines cause too much of temp drop between the zones? the baseboard (180 degree) is the first in the flow order. It would be a lot easier to make these changes than some other possibilities i have found. As it would only involve for each zone - 2 cuts and a T and 45 installed. Along with moving the primary pump.

Would i still be able to keep the primary loop at 3/4" copper? There are ball valves where the secondary loops come off the primary, i'm guessing to help with the balancing / flow control, but no way to indicate what the flow is (some sort of gauge that can be installed somewhere maybe?).

I did check with all of my local HVAC companies, and no one wants to have anything to do with this system, probably as someone else pointed out earlier that it appears to be a DIY homeowner install. The find a pro closest match was about 3 hours away.

I'm not really sure what to do.


Thanks,

Scott

Jamie_5

what to do

Scott:

I'm sympathetic. I inherited a terribly messed-up baseboard system in the house where I live. Massively oversized boiler, fintube convectors at ceiling level, direct return system with supply and return tees at the end placed a few inches apart, etc., ad nauseum. I ended up ripping it out and starting over (along with almost everything else in the house, as well). So I'll post a diagram or two and an attempted explanation of what might work on what you've got this weekend.

However, what I was suggesting before was that, if you're willing to do the physical work and willing to absolve a contractor of responsibility for whether changing the system will make it work, maybe someone would be willing---for a fee---to provide a detailed schematic of their best attempt at a design fix. Then the distance to one of the professionals listed at this site wouldn't matter, since you could do the consulting electronically, either by phone or e-mail.

You could also try the Radiant Panel Association website for a contractor lead.

Jamie_5

proposed fix

If I've done this right, I've attached a picture that shows how I would modify your set-up.

The primary loop is 3/4" copper, with three parallel bridges piped in 1/2" copper. I've moved the pump to pump away from the point of no pressure change (the expansion tank), and placed the pressure reducing valve to feed there also. The ball valve and boiler drain are there to help you purge air from the primary loop. I have not illustrated an air elimination device, but you probably want to install one. If you do, most have a tap for piping to the expansion tank line.

I would use a Taco 009 for the primary pump. Based on the chart you posted about the pressure drop through the heat exchanger, a 009 should provide sufficient flow of about 4GPM. If you make the cross-over bridges the same lengths, the flow should be about equal across them as well, giving 1.3 GPM through each bridge. However, the return side thermometers will tell the story. The ball valves on the bridges are there to help with purging, allowing you to isolate each bridge, and in case you need to balance the flow across those three bridges.

The baseboard zone supposedly needs 1GPM flow of 180F water. I assume it's the basement zone, which means it will be short: that zone has a heat loss of about 5500 BTU/hr., and the baseboard should be good for about 500 BTU/hr. per foot of baseboard at an average water temperature of 170F. You will use the globe valve on that zone to give the circulator something to "push" against and reduce the flow in the zone to 1GPM. You can also close the globe valve completely when you have to purge that zone. The thermometer will tell you when you have the right flow. Working on the assumption of 180F supply water and 170F average water temperature, there should be a 20F drop from supply to return on this proposal, so you should see 160F water coming back.

The two radiant zones should be piped identical to one another. Based on the calculated head of those zones, the Taco 007s should move just about 4GPM, which is what you say you need. As previously calculated, 1.3GPM of 180F water combined with 2.7GPM of 150F water should give you 4GPM of 160F water. The thermostatic mixing valve should take care of that automatically, assuming you can find one that has 160F in its mixed output range. (If not, you can put a ball valve on the bypass, illustrated as going into the mixing valve, and another ball valve on the return to the cross-over bridge, plus a thermometer after the pump, then fiddle with the pair until you get the mixed temperature you need.)

The 1/2" x 1/2" x 3/4" tees on the cross-overs that feed the secondary zones need to be closely spaced, less than 6" between the side port centerlines. This produces a very low pressure drop between the two tees and means that changes in flow in the secondary circuits (here, the zones) won't produce pressure changes in the primary circuit. When the pump for that zone is off, the water will just flow across the two tees. When the pump is on, the water will go to the lower pressure area created by the pump, and enter the secondary circuit. On the radiant zones, the thermostatic valve will allow 1.3GPM of 180F water to enter the secondary and it will return 1.3GPM through the downstream tee. There should be no flow across the common piping, so leaving it 1/2" should be fine. On the baseboard zone, the throttling of the pump by the globe valve should allow only 1GPM to enter the secondary circuit, so 0.3GPM will flow across the common piping to join the 1GPM coming out of the secondary circuit.

The return temperatures should be 1.3GPM of 150F water for each baseboard zone and 1.3GPM of 165F water from the baseboard zone (1GPM of 160F water and 0.3GPM of 180F water). The water heater should thus see 3.9GPM of about 155F water. However, this implies that the system sheds about 49,000 BTU/hr. rather than the 37,500 that is your calculated heat loss. I therefore suspect that the assumed flow rates for the radiant zones are higher than they actually need to be and that you will see less of a temperature drop across those zones than calculated.

I hope this helps and that you can make the system work. Best of luck to you.

ScottFP

How about this

Would this new design be correct for converting the existing system to a primary / secondary system? I think i have done the calculations correctly. This seems like it would be an easy conversion from what is already installed. There is also another post above for a possible solution using parallel secondaries, of the two solutions assuming mine is even correct, which would work better?

PS. Thank you Jamie Rucker for the other solution.

Thank You,

Scott

Jamie_5

nomenclature

Scott:

I ran the numbers, and your set-up should work, though by the time one gets to the last zone, the supply temperature is around 165F, which means that almost all of the primary loop flow has to go to the secondary. That's not a problem, it just means the mixing valve has to be sufficiently unrestrictive to let the amount flow in through the hot port. Also, make sure you include the resistance of the mixing valve in your head calculation.

A few things. I was using PRV to mean pressure reducing valve, i.e. the auto feeder, while I now see you meant pressure relief valve. The auto feeder should come in at the expansion tank, and the tank pressure and feeder pressure should be the same. Tee it in between the expansion tank and air separator. You should also put in some purge valves to make your life easier when you fill the system and try to get most of the air out. Alternatively, put ball valves between the supply and return of each secondary in the primary loop, so you can force the water to flow through the secondary when you're purging. And, if you don't include some way to reduce the flow in the baseboard circuit to limit it to 1GPM, the circulator on that zone will take as much water from the primary circuit as it can push around the baseboard loop (at the head you calculate for that loop). That will raise the average water temperature to that loop. I don't know whether it will lower the primary loop water temperature and give you a problem downstream. I assume it will not, unless your heat loss through the baseboard zone is higher than calculated. And, if the head is as low as you calculate, the flow in that loop will be quite high, high enough for velocity noise and erosion. I'd go to a smaller pump for that zone or include a throttling device.

I'd give it a try. It sounds like you've got little to lose.

ScottFP

Would the ball valve at the beginning of the baseboard zone be sufficient to limit the flow to 1 GPM (IE: semi-open)? or should i do as you suggested before and put a gate valve after the pump to limit the flow?


Thanks,

Scott

Jamie_5

valves

My hands-on knowledge is very limited, so I don't have the experience to know whether the ball valve will work to limit flow sufficiently in your application. A ball valve is probably sufficient. However, I think it would be better to put the restriction after the pump so as not to create a pressure drop right before the pump, which could conceivably lead to cavitation.

It is my understanding that you should *not* use a gate valve, which is different from a globe valve, as a flow limiting device. I believe this is due to the "shutter" design of a gate valve. A partially closed gate valve will present a slightly free disc as an obstruction to flow and that disc will get vibrated against the channel it slides in, causing noise and degrading the disc. At least that is my understanding of the reason.

Tony_23

Bbd loop

I think you must have more than one foot of head loss in that loop. You have two checks and 100' of tube plus ftgs. If you move the water faster than required, you won't get the deltaT, but it will heat alright and the return will be hotter at 4 GPM than 1 GPM.
Now, you could just install a smaller circ more suited to your requirements. A 005 or a B&G NRF9 might do you better, or a 3 speed Grundfos on low. Do not use a gate or ball valve to regulate flow, use a globe valve or a circuit setter.

You're making progress