Radiant Not Warm Enough in 1 Room

Too Hot in Cold_2

As I discussed a month or two ago, my radiant system was plumbed with a conventional boiler, 5 zones, but no primary-secondary loop.

I have started a new thread on this because I got some more detailed information about the system temps. Also, I think that my posts were lost in the original thread, which concerend an overheating problem that has been solved.

The system uses 5 zone valves (one for each zone), a single Taco 007 circulator, and it is a Utica MGB furnace with input=125,000 BTU/hr. Tubing is Onix, embedded in mud, with tile on top. Family room has tubing sandwiched between floor and subfloor, with wooden spacers holding floor off subfloor. Kitchen and family room are over finished, heated basement. No insulation was put in floors under kitchen and family room. The plumber set the boiler to 140 degrees.

1. I am concerned that the water temperature is too low and the boiler will be damaged by corrosion due to condensation. I measured using an infrared thermometer and the temp of the supply line ranges 122 - 142 as the burners cycle on and off. The boiler is probably a little hotter.The temps. of the individual zone return pipes range from 110 - 130. The actual return into the boiler measures 115 - 120. Note that I put a piece of masking tape on each pipe as suggested in the thermometer manual. I don't know how accurate the temperatures are, but copper is a good conductor. Also, I did not measure all zones for long periods of time. Some returns may get hotter as the floor warms up.

2. My 4 zones that are under tile work ok so far as heating.

3. I have one zone of radiant in a sandwich under hardwood floor. Now that the cold weather has hit, I have found that the heat is not adequate. I have the thermostat set to 71. One day so far, the outside temperature went into the teens the temperature reading on the thermostat dropped to 66 and that zone ran continuously. The next day, the outside temp went up to 30 and the inside temp of that room crept up to 69. The day after, it got up to 71.

Using the infrared thermometer, I have found that the floor temperature in this room only gets as high as 78 - 80, depending on where I measure. I measured the temp. of the copper supply pipe and it ranges from 122 - 142. The return ranges from 118 - 128, so it seems like the flow is ok.

My tiled radiant floor areas get up to 85 - 90 degrees. The returns from the tiled areas are cooler than from the hardwood floor area. My kitchen, which is the biggest zone, has a return temp of about 112 degrees, and does not fluctuate much as the burners turn on and off. I have some big bathrooms which have 115 - 119 degree return temps, and one tiny bathroom, which has 120 - 128 degree return temp.

These numbers seem to make sens to me. The tile is a better conductor than wood, so the tile floors get hotter and the water flowing through them gets cooler. Not much heat conducts through the wood floor so it does not get so warm, and the water does not cool so much.

It seems therefore, that the water needs to be hotter for the zone on the hardwood floor.

Someone here suggested raising the furnace temp 10 degrees to giv e a little more heat. This would help with the return temps also.

My estimate of the heat loss for this room is 12,000 btus when it is 0 degrees outside. I live in NJ. The floor area of the room is approx. 400 sq. feet. If I figure 30 btu/sq. ft. then the system should be adequate. The tubing looked like sections were less than 12 inches apart. There are 4 or 5 runs of tubing coming off the manifold, so I do not think the lengths are too long. The individual tubing returns are all similar temperature, 109 degrees or so. This is lower than what I measure for the copper return, but the Onix tubing is probably not as good a conductor as the copper.

It seems like I need either a primary-secondary system or one or more boiler bypasses to protect my boiler and provide different water temps to the hardwood and tiled zones.

5. I have attached pictures of my furnace. In the picture with two furnaces, the radiant system is on the right. In the picture showing only the furnace for the radiant, the zone for the hardwood floor is on the right, obscured by the gas pipe. The expansion tank is on the supply line, which goes to the right. The return for the HW floor zone is larger in diameter than the returns for most of the other zones. The leftmost return is for the kitchen. It is a larger one also.

Estimate heat loads are: kitchen - 17,000 btu, room with hardwood floor - 12,000 btu, the other three zones are tiled bathrooms, two are about 5,000 btu, one is 1,600 btu. These are all my estimates, not the plumber's. The boiler is UTICA MGB -125,000 btu/hr.


6. Any suggestions for adding a primary-secondary loop or boiler bypass(es)?

7. Assuming that it is the way to go, how should I squeeze in the primary-secondary loop? I would like to alter the system as little as possible in order to keep the cost down. There is some room on the wall to the right of the boiler, however the zones all come up directly over the boiler.

8. The system was installed as part of a renovation. I appreciate the suggestions to get someone who knows what they are doing to fix up the system, but, since the general contractor is not finished with my house I still have to go through some discussions with his plumber to see if he can solve the problem. If he fails then I can try to get the GC to bring in someone else to fix it.

Weezbo

i was going to send you an e mail with a picture,

just to give you a change of view on how the header might be placed.look on top right photo. thats my buddy Fred this is the actual best and only place for the header.

Mike T., Swampeast MO

Your "sandwich" floor is relying heavily on convection while the embedded works via conduction. You're feeling the difference with cold weather.

In my mind at least you have a "two temperature" system.

You need primary-secondary and a mixing valve. Temps will be higher for the primary and family room (thus lessening the concern of low temps and condensation) with the rest of the system served via the mixed (lower) temp.

S Davis

Wood Floors

You should not run a wood floor over 80 degrees surface temp so you might be at the max output in this area. Was a heatloss done?

S Davis


Apex Radiant Heating

Mitch_4

correct me if I am wrong

but I added your total estimated heat loss and came up with 40,600 btu/hr.

Your boiler will put out 100,000 btu/hr @80% efficient.

A tad oversized maybe? WHY???? I would be asking the GC / plumber that question first. (If the answer is insurance if it gets cold, you should be good for the next ICE AGE!)...

I would run away from that plumber and get a...

HEATING PROFESSIONAL!

you could have used the MGB 50 or 75 and saved yourself some $$$. I imagine you short cycle a bit.

Utica recommends boiler protection, and I dont see it.

Get the manual and read up on the recommended piping. INSIST it is done per mfg specs and recommendations, and find out why the frig he used such a HUGE boiler.

I know this doesn't address the zone issue, but that has been done.

Mitch

Kal Row

\"a single Taco 007 \" stop right there...

you are most likely out of "push" or what we call "head" against the friction of all your zones - you might need a pump per zone, and the expansion tank preceding the pump ("pump away" as we call it) – better would be to have primary/secondary/injection…

for a quick and dirty fix you could try a Grundfos 26-97 or 99, it draws 3 amps but has some serious push

we need to know your closest guess of tubing type/size/lengths to even begin to give you a closely sized pump - a 26-99 is overkill (and will need a pressure differential bypass) and a 007 wont cut it –

the major problem in your system is that for the boiler’s sake you really need to move the water – eg if the boiler is 100,000 then at least, you need to move at least 10gpm through it, to get the heat out, and it will take 1-1/4 pipe, whereas the loops need a lot more push and a lot less flow cause the you want a bigger temp difference from supply to return – LOL

as the robot said in “short circuit” “need more input…”

too hot in cold

Heat Loss

I had thought that 85 degrees was the limit for hardwood floors?

I am not sure what kind of heat loss calc. was done. This was all part of an addition to my house.

My own estimate is approx. 11,000 btu/hr heat loss at 0 degrees. Floor area is about 400 sq. ft. I thought that 85 degree temp will give 30 btu/hr/sq. ft, which would give 12,000 btu/hr.

If floor temp is max 80 degrees then this is probably 8,000 btu/hr.

too hot in cold

At one point I had been talking with the plumber about putting an indrect HW heater on the system for the master bath in case the main water heater was not sufficient for the entire house. The plumber said therefore that he had oversized that one somewhat.

I will check what the cycle time is. Some of the time, it runs for 3 minutes on and 3 off. I thought that some of the time it runs longer, but I will have to wait for more cold weather to check.

too hot in cold

Pump

The main supply pipe to all zones is 1 1/8 OD copper. The returns from the two big zones are also 1 1/8 OD copper. The returns from the small zones are 7/8" OD copper.

The HW floor zone comes off the supply close to the boiler.

Supply temp cycles from 122 to 140, with an average of 130. Returns are between 110 and 130. I thought that the temp. drops between supply and returns are therefore 10 - 20 degrees on average, which is OK, so I thought that the flows are adequate, especially for the hardwood floor zone.

What will I gain from a bigger pump? It seems like it might help the tile zones heat a little more uniformly, but will not help the hardwood floor zone.

too hot in cold

How Pipe 1 Room on Primary, Rest on Secondary?

How would 1 room be piped on the primary and the rest on the secondary? All the primary could not go through the family room because there would not be enough flow for the secondary.

Kal Row

you guessed it \"eveness\"...

even though in a the radiant world, you want larger delta-T's, however for the sake of evenness - you want more flow - that’s why we use closely spaced T injection loops profusely in radiant - so that the heat supplied to the loop, and the flow rates between the loops are independent,

When you pipe a supply and return, off the main, with less than six pipe diameters apart, then, when a pump goes on or off or changes speed, it doesn’t change the water speed in the other, so it is like taking water from one pot of hot water to another as needed,
of course a change in speed in the primary affects how much heat you can get out of it, and it’s usually set to the total btu avail using the 1 GPM for every 10,000 btu rule (which itself is based on the specific heat capacity of water at an avg temp of 170 (delta-T of 20 180-160))

To get this right, you need, a boiler loop, system loop, injection loop, and zone loops – and have tekmar boiler control with outdoor reset providing the correct total supply temp based on outside temp, also protects boiler from condensing and short cycling – and runs pumps 1min a day during the summer to keep them from sticking.

So you have the boiler with it’s taco 007 running the water around in it’s loop only getting to 155-180 depending on the outside temp, and another 007 variable speed driven by the tekmar control, to inject just the right of head into the system loop, again with it own 007 – the loop will run 120-170 depending on the outdoor temp (I happen to use the 007ifc or the Grundfos 15-58fc – I like the built-in spring checks in these pumps – less parts and better flow rates).

Now, each zone, “close-T’s” into the system loop, and keeps it’s zone, at the correct temp, – in fact the zone pumps, can run all the time, or at least, be turned on by the first stage, of a two stage tstat.

Now if the zones are using variable speed pumps, or thermostatic 3way valves, to control it’s temp, – then, you can even run the boiler loop at a higher temp, and pipe the different temp type loads, all off the same system loop

This is what I would have done :

You have two boilers – ok let both boilers pipe into one primary loop with a tekmar two boiler controller,(also gives you boiler redundancy), from there, inject into the system loop with a tekmar 362(also provides for domestic hot water) using outdoor reset, and set the tekmar for the highest temp load type which in your case is the fin tube baseboard,
then have each zone close-T’d off the system loop, each with it’s own pump at full speed, using a thermostatic 3way mixer for the high temp zones, and electronic 3way mixing ah’la’tekmar for the radiant zones - this way you are getting maximum flow everywhere for temp evenness in that zone, and you are putting the correct amount of btu’s exactly where needed

the good news is that you can do all this in just in the boiler room, the bad news is, it will cost about as much as the new boiler installs

DIYPeter

close Ts

kal, I'm finishing my install, and have 9" between Ts on 1" secondary loop. 5 3-way mixing valves w/6 pumps/zones. Should I do over/ (it looks so pretty now)

Kal Row

that depends on the secondary size feeding the zones...

it's hard to get a 3-way setup closer without some pipe games, you did not give me the pipe size of the secondary, since i think that you have to be closer than 4 times the diameter of the straight run, so that the flow in the close-T takeoff doesn’t affect the straight run, so you see the larger the diameter of the supply header the easier this is to do, CALIFFEI makes a box with baffles that just perfect for this, in fact they always bring it to the trade show on a board with a bunch of pump loops and pin-wheel flow indicators on each and demonstrate how turning on any pump doesn’t make anyone of the others turn – they call it “hydraulic separator manifold” www.caleffi.com/?q=node/cb/8/calew/ser/cat/1/178/547/51 ,
taco also makes a close T fitting for larger pipe sizes – it’s basically a 2” or larger couplings with a length-wise baffle down the middle, and two ¾ openings on top on either side of the baffle, makes it so easy and you don’t have to worry about direction either - www.taco-hvac.com/uploads/File_Library_NEW/LM_TwinTee_Brochure.pdf ,
it’s part of their “LoadMatch” system – (have fun trying to buy them though – I just love it when manufactures make a big splash and then don’t supply the splashed product – but that’s another rant)

basically it’s a good idea to make the boiler loop and system loop extra large, for one this guarantees good flow through boiler, as for the system loop if the pipe is really large, then it act’s as a better pot of hot water, ie the temps are more even across the load close-T pairs. Also rule for injection between loops is, that in order to mitigate gravity ghost flow you need to have the injection loop at least on pipe size smaller, and be pumping downward from hot to cold apx 2ft against the natural flow of gravity, so that when the injection pump unseats it’s spring check, the water doesn’t just take off like a rocket even though the injecting pump is only set for 10% by it’s controller – ps, with the larger flow requirements – it’s usually a better idea to but two pumps in parallel instead of one large one – this way you don’t have all your eggs in one basket

as for your system – your bigger problem might be too low a temperature at the last close-T, again that why I like large supply rails with parallel pumps for such a show – the idea is to keep the pressure differential between the t’s as low as possible

Bob Forand

Caleffi makes a primary/secondary loop in a box. It is small enough that you should be able to fit is in your system. This will allow for control of the system and boiler protection...Good luck.

Kal Row

somewere in my whole megila above, is the caleffi link