Radiant floor heating help

Mark Custis

so I do not have all my toys. My laptop is striped for travel and I did not bring my scanner.

ME's essay is on the money, so is MPF.

I have a question, what is the heat loss on the home under design conditions?

I am starting to think like the brick under the gas peddle image we may not be able to get the entire load through the threeway at what looks to me to be 3/4". Adding to ME's thinking and other's comments, it might be easier to start over on the near boiler piping. I am thinking about getting rid of some of the pumps, and adding a mixer for each zone.

In a perfect world I would:

Dump the pumps.

Put a zone valve and three way on each radiant loop. Add check valves to insure flow dirrection. An alternate configuration could be done with injection pumping to the loops, through the threeways.

Pipe P/S or hydro seperator and insure you are not condensing the boiler.

I think one could control the entire system with a relay package, then add the dial up remote later.

Just my thinking today

Merry Christmas

stuart_9

Radiant floor heating help!!

Hi,

I have just purchased a 3 story house with radiant floor heating.

The system just doesn't seem to heat that well. I’m having difficulty trying to figure out the problem. I've attached a schematic of the system.

Is it "OK" to have the basement loops "T'd" off the main loop?

Any help would be graciously appreciated.

Thanks, Stuart

Mark Eatherton

WAY too many pumps...

and some in series.

What controls the 3 way valve?

What brand and model of boiler ?

Are there check valves on all pumps to avoid back flow?

Was this installation done by a professional?

Got pictures?

Answering these questions will cause us to ask more, but we will eventually get you some heat.

ME

Mark Eatherton

Although not piped really well...

it OUGHT to put out heat. Heat is but one component of comfort, but it should put out heat.

Is it staple up, staple down, in plates, without plates or what, and what is the supply temperature and return temperatures doing?

Are you interested in saving some electricity?

How large an area are the manifolds serving, and how many ports per floor?

ME

heatboy_13

Not only......

..... too many pumps, but what starts them. It looks like the boiler return pump needs to run when anything else is running. What starts it? It's possible, depending on pump sizes and pressure drop inside the secondary loops, the primary loop may have so much flow that any of the secondary loops, including the basement, may not get water back into the primary. Or, it might also just throw water everywhere and not have any thermostat zone control.

Unknown

Are you sure,,,

that the diagram is 100% correct? Like ME said, show us valves and check valves please.

stuart_9

More data

I've updated the drawing to show the thermostats and ball valves:

To answer some of the questions:

1) Circulator 2 is switched on when the furnace fires
2) Circulator 1 is switched on when any of the thermostats ask for heat
3) There are no check valves in the system I can see (one of the pumps, the top floor, has a sticker on saying "check valve removed"). The other circulators i cannot tell.
4) The three way valve is a manual mixing valve
5) The boiler is a gas fired Utica series MGB boiler
6) The PEX tubing is stapled to the floor and covered with about 1 inch of concrete
7) Each floor is about 670 sqft, each floor has a separate manifold with 3 heating loops.
8) I will take pictures tomorrow when I am back there (its a vacation home i normally use at weekends).

The reason I started looking at the systems was that it didn’t seem to be getting as warm as it should where it should i.e. some floors would get warm when the thermostat was satisfied.

Thanks for all the replies.

Stu

stuart_9

Another quick question:

How do you know if the circulator pumps are working correctly? I feel them spinning but do not know if they are pumping.

The return circulator (#2 in my updated sketch) is a bell and gossett SLC - 30

Thanks, Stu

Brian R

Weekend Usage

I'll let the pros here digest the pumps, valves, etc. I picked up on your comment about only weekend use. Do you turn down the thermostats to a very low setting when the building is not occupied? Are you familiar with the time lag for a radiant system to warm-up? I'm wondering if the response time is being confused with lack of heat output? Just my 2¢ from a homeowner with radiant heat.

Unknown

Excellent point Brian...

So we'll be looking for an answer to that question. Off the top o my head I'd remove circ #1 and just connect the pipes in it's place. Put a valve or some other means of stopping circulation through that loop you show above and around the 1st and 2nd floor loops and make sure there are check valves in all circuits or add them if there is not. If they are Taco IFC pumps you may find it necessary to pull each one to ensure there are checks still in them and replace the one where it indicates the check has been removed. However that's just my $0.02, wait and see what others have to say about it.

stuart_9

Weekend use

Thanks for the comments:

I do set the thermostats to 55F when I’m not there. How long does a radiant system take to warm up?

The reason I think there's an issue is the fact that the thermostat settings have no effect on zone temp and that the top two floors of the house have relatively cold inlet temps to the heating loops (significantly lower than furnace temp). You can barely feel the heat on the tiled floor in bear feet on the top floor. I think the top floor gets most of the heat by convection from the lower levels.

Thanks, Stu

Unknown

It can take...

a couple days.

But you SHOULD feel heat going up to the floors almost immediately.

Mark Eatherton

Confirming flow

You can close a ball valve near the pump quickly while it is running. This SHOULD create a Swoooshing sound, if not, you have little to no flow.

THe upper floors could be air bound. Take lots of pictures of the near boiler piping when you go back.

Ther should be a difference in temperature across the supply and return maniflds. If not, you are probably air bound and the system will need a good purging, but you still have some "issues" to deal with, especially in light of your no check valve comment...

ME

Jamie_5

supply temperatures

Disclaimer: non-professional input follows.

Your system is apparently designed to supply water to the zones lower in temperature than what the boiler supplies. That is the function of the mixing valve, which is piped before the split into two parallel branches, each of which splits into two parallel zones. Thus, the heating loop inlets should be cooler than the boiler outlet. However, the inlets to each manifold should see the same water temperatures. If the supply temperatures to the upper floors are cooler than the supply temperatures (at the inlet to the manifold) for the basement and Florida room, then you have a problem specific to those upper floors.

As piped, neither pump 1 nor 2 seems necessary, assuming each zone pump is properly sized. What seems to be missing is some way to protect your dry-base, atmospheric, cast iron boiler from low temperature return water (and condensation).

I think I would hydraulically separate the zones from the boiler, either by primary-secondary piping or by using a tank, add a thermostatic valve to the boiler loop to protect it, and replace the manual mixing valve with a thermostatic valve or motorized valve set according to outside air temperature. It may also make sense to reduce the number of zone pumps and use zone valves to a certain extent. Otherwise, put a check valve on each zone.

Mark Eatherton

Not bad for a rookie :-)

Jamie hit most of your issues squarely on the head.

Boiler is not meant to operate with a return temperature of less than 140 degrees F. If exposed to this potential on a continuous basis, it will condense and will dissolve. Short of replacing this appliance with a high efficiency modulating/condensing appliance, (Big $'s) you need to change the near boiler piping to protect the appliance AND the floor. Speaking of which, what types of floor finish covering was used? It makes a BIG difference and may require numerous operating temperatures to overcome. Not easy to do in a finished home.

With this being a weekend use home, it WILL be exposed to long term condensation production because it spend most of its time recovering from a deep cold set back condition. The frustrating part of this is that just about the time the house starts feeling half way comfy, it's time for you to leave and go back to the city! There is a solution to this problem and we'll go over it after we address all the other "issues" your system has. Right now, you have some decisions to make before we start redesigning your system.

Do you want to reduce your energy consumption by a minimum of 30%?

Do you want to simplify the installation from a hardware stand point?

To put it bluntly, your system is a mechanical debacle and if you ignore it, it WILL go away...

Bring us pictures of the near boiler piping, answer these questions and we can move forward from there.

ME

Unknown

Ah!

I see what you're getting at with the boiler return temp. My thought was pump 2 was necessary to circulate through the boiler and mixing valve and then return temp would be governed by the mix of whatever passes through the mixing valve and water returning from the loops. I figured it would maintain at least 140* but I can see where that might not necessarily be true, especially under heavier load conditions. I also see where unless closely spaced tees were used there would most likely be ghost flow through all off zones generated by pump #2. And we can't use close tees because the downstream zones would see much cooler water than the upstream ones. I'm interested in seeing what your solution is.

Charlie from wmass

Radiant temps are usually under 100 degrees F on the supply side. This is why there is a mixing valve as boilers usually like warmer water going through them. This is also why set backs can be trouble some as water temps delta t's do not allow high speed recovery. It is like driving with a block under your gas pedal. You can get up to highway speed and you will save fuel but you may not get up to highway speed as fast as you would like.

stuart_9

Thank you for all the replies!!

I have been working on the system tonight, I have found that three of the circulating pumps has failed (they were powered but had sized motors; all of them Bell and Gossett SLC-30's). I removed pump 1 and used it as a replacement for the B&G pump 2 (pump 1 is a Taco 007). I now have heat on the 1st floor (the zone pump on the 1st floor is a 3 position Grundfos)

What would be a good replacement for the B&G SLC-30’s?

Thanks again, Stu

stuart_9

Thinking about this some more, if I have to replace the circulators, is there a better/more effective way to control the zones?

To answer a question from earlier, pump 2 is switched when any of the thermostats request heat (pump 1 was wired the same, it is now removed, see previous post). The zone pumps are switched when the local thermostat requests heat.

I will post pictures of the near boiler plumbing tomorrow.

Stu

Unknown

4:30 AM revelation...

The kind where you wake up and say, hmmmm...

I have to wonder about the way the mixing valve is plumbed in. I'm assuming when you use the term "manual mixing valve" you mean it's adjusted manually but then operates thermostatically to maintain the set outlet temp. Normally a mixing valve will have a hot inlet, a cold inlet and a mixed outlet. It is not plumbed that way according to your diagram. There is a hot inlet, an outlet to the system and an outlet to return to the boiler where normally you would expect a cold inlet. In the case of floor loops the loop returns would be plumbed to the cold inlet and a pump used to circulate from the mixed outlet to the cold water inlet. I can't see how this would ever provide anything but boiler outlet temp water to the system and have to wonder if it isn't mostly shut to the system because it's trying to reduce it's outlet temp but isn't able to. Please include a good closeup of the valve when you take your pics and show us how the connections are labeled on the valve itself and which one is plumbed to what. Now that you have some heat going I think we'd also like to know what temp water is actually going to the floors. ME is right when he says debacle. Sad to see such a potentially great system so fouled up in such a manner.

Jamie_5

non-directional pipe

I believe the pipe you're calling the "outlet to return" acts as the cold inlet to the valve. Just imagine the flow going into the mixing valve rather than out for that section. The pumps are on the mixed side of the mixing valve and the cold inlet to the mixing valve pulls from the return side of the system. So the system should provide mixed temperature water to the zones. Or so I believe.

Brian R

Remote Control Thermostats

After you Pros get done fixing & optimizing the hydronic end of this guys system, I think he could use a remote controlled thermostat. I remember reading about such a while ago here, where you could call the thermostat by phone to change the setpoint to occupied mode. This could really help this guy, by calling ahead a day or two in advance to warm up the building. If his system has enough mass, he could probably set back to un-occupied mode on his last day at the weekend home, and the radiant mass will carry-over until he leaves the place. Maybe someone could post their suggestion for something like this.

[Deleted User]

The professor from Gilligans Isle has the answer....

Freeze Alarm.

http://www.safemart.com/Freeze-Alarms/Advanced-Freeze-Alarm-FA-I-CCA.htm?CAWELAID=63025791

ME

stuart_9

I've attached pictures of the near boiler piping and thermostatic valve, thanks, Stu

Rich L.

Taco

The Taco 007 has very similar charistics to your B&G pump. They make a great pump.

That said you could eliminate your zone pumps and switch to zone valves for control and you would likely see an energy savings. This would of course require some piping changes, which it looks like you need anyway to protect your boiler.

Mark Custis

Stu

It looks like the mixer is plumbed right, but I am only on my secound cup of coffe after diner out.

Do you know what the mixed temperature is?

I like the call ahead for heat idea.

stuart_9

I think the mixing valve is plumbed wrong, see attached sketch.

I think the cold supply to the mixing valve is the wrong side of the circulator. The way it is plumbed you would never get any cold flow to the mixing valve. What do you think about the proposed line (in red on the sketch)?

Thanks, Stu

Unknown

Appears to be wrong to me...

Normally you'd have two inlets and one outlet (mix) and you have water entering in two places, doesn't seem right unless it's a style of mixer I'm not familiar with. I don't think the solution on your redraw is the best way to go about it even though it looks like it might work better than what you have. My brain's a little fried right now too so I'm having trouble thinking through the circuit myself. I'll look at it again later. I like the dial up t'stat idea as well. Maybe ME will jump in and tell us what he has in mind in the meantime.

Unknown

Looking at it some more...

it gets even more confusing. According to the original diagram not only is it piped to the inlet of circ#2 it is also piped to the outlets of all other circs and if the sequence in piping connections is correct it looks like the water from the loop pumps would have to go backwards against circ#2 flow direction to make it to the cold water inlet of the mixing valve. Maybe circ 1 got added to win this battle? Very strange configuration.

Unknown

beat me by a click...



Exactly what I was thinking. The mixing valve is absolutely too small to start with so really there is no quick fix here. The simplest way I can see to use most of what he's got and not add a tank is to establish each loop with it's own mixing valve and go primary/secondary with an 1 1/4" primary loop around the boiler and 1 1/4" x 3/4" manifolds close teed into the primary loop after the primary pump. Then the upstream manifold would feed the hot side of each mixer and the cold inlet to the each mixer teed off into the downstream manifold. Simple P/S. Then each loop pump operates on a call from it's respective stat and the primary pump runs when any secondary pump runs. Boiler set to operate on it's limit. I suggest he purchases Dan's "Pumping Away" and (in my copy anyways, fourth printing 1997)) turn to page 85 for a basic diagram of P/S piping and then read further for several variations on the original diagram. Unless someone wants to draw and post a diagram here. I can but not now and it would only be a pic of my chicken scratches on paper. I never did buy the capapbilty to draw these on the computer myself. I may break down one day.

I'm still interested as to what approach ME has in mind as I'm all for reducing pumps to a minimum. I THINK he's leaning mod/con but I'm not sure.

Mark Custis

Thanks

While we are at it and it is Christmas. Lets pump the secondary loop with and ECM pump or other delt smart pump. Set the pump to do all the zones and have it throttle back for a lesser load. We could also use an actuated manifold instead of zone valves.

Adding t-stats and flow meters will let us see what the system is doing and plan control strategies accordingly.

I will learn cad, but if I could scan I would include a drawing or two.

Do we know if this is an atmospheric CI boiler?

Unknown

Merry Christmas back!



he said it was a gas fired utica MGB. I'm assuming it's atmospheric CI but I don't know Utica that well.

Good ideas on the ECM etc. I'm still slightly stuck in the pump for each zone world. I did look into doing a single delta variable for all zones like that on a project I'm doing right now but the only pump I found to be capable that also reduced power consumption per load as well was very expensive, so I stayed with one pump per zone on it for economy reasons. I do plan to use delta variables for the floor zones though. My goal was to vary power consumption in relation to load changes and noone (that I know of) is really making smaller pumps that have that capability so my next choice was to use one pump per zone and compartmentalize power consumption that way. Each zone should draw no more that .25 to .50 amps each when running and I don't expect them to all run simultaneously very often. If all goes well I expect the system to draw no more than maybe 8 amps for both heating and cooling when all motors are running, including 2 air handler fan motors.

stuart_9

Thank you for all your help.

There were a number of things wrong with the original sketch; it took removing some drywall to figure out the piping route!! I have attached an update system schematic.

I have two circulators operational, the top floor and bottom floor. This gives me enough heat to get by, now I want to make the system more effective/efficient.

I have purchased Don's book, Pumping Away. Will this provide the necessary system schematics and system trade off's?

Thanks again, Stu

Mark Eatherton

That'll work, but...

Make certain there is a check valve on every pump or you WILL have problems.
As for increases in efficiency, you are going to have to change your heat source to a modulating/condensing heat source if you REALLY want to save. Also, you still need to address the long term condensation production potential. If you go mod con, you won't have to worry about that, but it will require some modifications.

ME

Unknown

I'd have to,,,

purchase an updated version myself and see what might be new in there. The basic P/S diagram will get you by unless you want the added efficiency (~30%) that ME is talking about. I thought the mixer was 3/4" but now I see it's 1". I still don't think it's big enough but it's better than 3/4". Your diagram does not show the main loop circ now. Is that an oversight? Somewhere in Dan's book he mentions a Danfoss ESBE valve that will protect the boiler from condensation issues if you want to stay with the boiler you have. I can't find it in mine, maybe it's in another of his books but here is a link:

http://na.heating.danfoss.com/xxTypex/156252_MNU17421643_SIT209.html

It bypasses the supply past the system and back to the return to ensure at least 140* return temp at all times.

stuart_9

The main loop circulator has been removed (I found that it was never connected the control panel). Also, the way it was plumbed the mixing valve would never receive cold water. I actually put a working circulator in it place before I knew how the mixing valve worked, once the boiler was up to temp the system didn’t flow at all....I assume the mixing valve was trying to add cool water to drop the temp but it was been pumped away by the circulator.

The only circulators I have are on each radiant loop.

Thanks, Stu

Unknown

Hmmm,

a bit unorthodox but it at least looks like it might heat. Condensation is still an issue and without major repiping work I can't see any way of correcting that aspect. Take the return water temp just before the boiler, downstream of all returns with all pumps running and see if it's at least 135*F. If it's lower than that the boiler will condense flue gas on the fireside and the resulting acidic solution will eat the cast iron away in short order. ME's mod con suggestion w2ould alleviate that situation but either way major reconfiguration is in order I'm afraid.

stuart_9

I have just measured some temps:

Out of boiler - 170F
Out of mixing valve - 115F
Return to boiler - 98F

Thanks, Stu

Mark Eatherton

Sounds reasonable...

except for the low return temp going back to the boiler, which will cause it to condense. Every boiler condenses at one point or another, it's a matter of how long it condenses that causes concern. Based on your usage pattern (weekend use) it probably condenses pretty much all the time.

A bypass won't help you because you don't have a boiler primary pump.

Also, you are exposing the cast iron sections to a pretty strong temperature differential, which will cause thermal stress across the sections due to temperature differentials, and can cause early failure of the cast iron sections.

How long will it last? No one knows for sure, but it is not operating under the best of conditions...

ME

Unknown

Right...

That 98* return is not good!

Jamie_5

4-way valve

I'm revising my earlier suggestion. This looks to me like a good application for a four-way mixing valve. If he replaces the thermostatic valve with a 4-way and appropriate controller, like a Tekmar 360, he could gain outdoor reset, boiler protection, and the ability to replace pumps with zone valves all in one fell swoop, no?

Of course, he would still need the freeze protection alarm or something similar.