Questions about radiant heat system

[GLyford]

Dad's controller had a separate controller input and pump for the stove, at least, so didn't need to wait for the panels to warm up. There was a thermal sensor at the stove.

The exchanger was of the stainless box variety, and was actually his 2nd, the first wasn't stainless and eventually leaked.

It all got ripped out when they redid the kitchen, and the house's new owner has taken down the panels as well.

 

[mx842]

Raspy..........you posted a suggestion over on I think it was #47 about how to pipe the HX and water tank. I think I'm a little confused about how to do this. I know I have stated that my system is a two piece setup and may be part of my own confusion. I guess it would be better to say this is a three piece setup rather than just two, the HX, the old water heater tank, then the supply loops to the manifold. I guess I was thinking of the HX and the water tank as one piece because I can isolate those two parts by full flow valves from the rest of the system. My original plan was to somehow get the water that was in the 30 gal water heater up to around 175 to 180 degrees which I still think I can do then somehow mix my return water from the loops with the hot water in the water heater tank to get it down to a more desirable temp for the floor.

I figured out that the HX holds just a tad over 6 gals and the storage is 30 gals.

My confusion with your suggestion is where do I tie in the water in and out of the HX. I like the idea of the top of the tank being used as an air and pressure tank but I already have one of those in the system along with a couple air bleeders at different points in the system. I'll probably be able to get back on this by Wed or Thurs so hopefully I'll have a better understanding of what to do when I get back at it. Thanks!!!

 

[mx842]

I haven't had time to post a lot for the past week or so but I think Im getting close to getting this thing up and running. I spent two days last week re-piping the system only to come up short and having to do it all over again. I had rebuilt my Taco 100 series in line pump that was leaking and had it plumbed in at the manifolds. I guess I should have checked it before I fired up the boiler but I didn't. Anyway, I fired it up and waited for the temp to come up to 150 coming out of the HX and turned on the taco 007e I had just bought to circulate water through the tempering tank and the HX. By then hot water had already been flowing into the tank through the pump so the tank had already warmed some. Everything looked ok so I threw in some more wood and once the tank had gotten to about 160 at the top of the tank and I had 135 degree water coming out the bottom line going back to the HX I went over to the manifold room to cut on the circulating pump for the loops. By then I already had warm water at the pump.

My plan was to cut on this pump and let it run for about 10 minutes just to see how much how much that cooled the water in the storage tank then go from there. It all went down hill from there. When I plugged in the pump water went ever where. Not only was water shooting out the weep hole it was also leaking out of one of the flange gaskets and the flange where the pump housing bolts up to the bearing housing. I fixed the flange gasket pretty quick and adjusted the spring coupler which slowed down the leak out of the weep hole. water was still pouring out of the pump housing and there was nothing I could do but pull it apart and try to fix it. I pulled it apart and everything looked ok, nothing out of place so I put it back together and tried it again but it still leaked like h@ll. By now the water in the boiler room was flirting with 220 degrees and there were some weird noises starting to come out of the storage tank. I ran back over to the manifold room and decided to put that old 100 series pump that i had replaced with this newer leaking pump and once I changed out the pumps and turned it on nothing happened. The leaking pump had tripped a breaker and by the time I figured that out and got the pump running the storage tank temp gauge was passing 240 and was really clanging and banging. I had put a mixing valve on between the water tank outlet and the supply pipe to the manifold and no matter how I adjusted it nothing would come out. I guess I had an air lock somewhere that was blocking either the supply or return line from the manifold. At that point I tried to get as much of the burning wood out of the stove as I could but it was too hot so took the water hose I already had laying on the floor and misted the fire until it had cooled down enough to be not putting any more heat into the water as there was already there. I cracked the fill valve that I had put in the top of the tank to vent air as I was filling the system. This let out let out some of the steam and pressure that was in the top of the tank and then left. The tank was making all sorts of different noises and it did this for about an hour as it was cooling down.

I removed the mixing valve from the system and changed a couple other things since and hopefully will be able to try again soon. I had flushed the whole system and when I filled the system the last time after I had checked for leaks and added conditioner the did a dry run to see how things looked. I had a hard time getting all the air out of the system and even then I still cant get the flow I want through the manifolds. The flow gauges hardly move at all and are reading only .2 gpm. I was hoping for at least 1 gpm through the loops but it don't look like either of these 3 piece pumps has the b@lls to do it or I still have air somewhere in the system that holding everything up.

 

[Raspy]

Exciting times. Good lessons.

Get that tempering valve out of there. A simple bypass will serve you better because it's not so restrictive. You need flow! Find out why the spring drive pump is leaking at the mechanical seal (is it an old worn out pump?).

Look at the manifold and figure out how to adjust the flow. This is VERY important. Don't assume anything, look at it carefully and figure it out. You must have the loops open and flowing correctly. You may not have the adjusters open as the flow is too low and your loop lengths seem right.

Don't buy any more pumps until you know what you are getting and why. The 007 should be fine on the HX, but it's a very low head pump. The floor needs a higher head than the old spring drive model Taco. It's probably OK, but a Grundfos 15-58 or a Grundfos 26-64 would be nice. A Grundfos 26-99 three speed is the Cadillac, but it's a very expensive pump.

Carry on. You've seen the effects of panic overheating with a wood fired system. Now, wait till you have a power failure and a big fire , while you are not paying much attention to the system. Eeek!.

Looking forward to your next installment.

 

[Raspy]

mx,

Regarding plumbing the HX to the water heater: The HX is horizontal and it doesn't matter which line is in or out, but plumb the out to the top of the tank with as little restriction as possible and an up-slope all the way. This will help the thermosyphon affect and, during a power failure, will circulate some water. Assist it with a pump to get the maximum BTUs to your tank. The 007 should be fine for this duty. Use the 3/4" port on the side of the tank near the top, typically, this is the pressure relief port, as the return from the HX.

TEE the output to your floor pump at the same upper side outlet of the tank, where the HX line ties in. Simply screw a brass or galvy tee on a short nipple sticking out of the tank fitting and plumb each line to it. Return the floor loop to the cold inlet on top of the tank. Pull the dip tube out of the cold inlet fitting and set it aside. Put an automatic air vent on the hot outlet. If you insist on an auto fill system, tie it in at the cold tank inlet with another tee.

I suppose the best flow pattern would be to place the HX pump on the bottom line pumping away from the tank toward the HX. Place the floor on the line from the upper side fitting tee, pumping toward the floor. Return it to the cold top fitting as mentioned.

All of this will tend to move air to the tank and get it out of the system. It will not inhibit thermosyphon.

Install tees on the supply and return lines to the floor. One just before the pump, which is pumping toward the manifold. And one after the manifold on the return line to the tank. Connect these with a 3/4 pipe and a ball valve. This will allow you to throttle a bypass and control the supply temperature somewhat. It also reduces restriction in the floor loop. You'll still get all the available energy to the floor, but you'll avoid scalding temperatures and you'll help the pump purge the air. A bit of manual adjusting of this ball valve will get you find the right setting. Probably 1/2 open for starters.

Install a ball valve on the return line to the tank from the floor. Preferably, near the tank. Then tee in a hose bib right before this valve (on the manifold side). You can then close the valve and open the bib to manually purge the floor loops. This will help you get it all started and force out any air.

Again, look carefully at the manifold to make sure you have the loop restrictors fully open. Very important.

 

[quicksandfarmer]

You're on the right tract with experimenting and observing, it's too bad your experiment was ruined by factors that really didn't have anything to do with it -- although it sounds like exciting times!

The reason you need to experiment is you'll only know if you need a higher capacity pump if your return water is coming back stone cold and your boiler water isn't getting too cold. If your return water is coming back warm, more flow won't do much for you. If your boiler water is dropping below an acceptable level, more flow will just take more heat out of the boiler. Or put another way, you want to increase the flow until you hit one of two limits of your system: the ability of the floor to disperse heat, or the ability of the boiler to provide heat. Once you've hit either of those limits more flow does you no good.

 

[Raspy]

you want to increase the flow until you hit one of two limits of your system: the ability of the floor to disperse heat, or the ability of the boiler to provide heat. Once you've hit either of those limits more flow does you no good.

Well, and a couple of other factors: He needs enough flow to evenly heat the entire floor such that it feels warm everywhere. This is why the floor circulation is entirely separate from the heat exchanger circulation. Heating the floor is different than the ability of the boiler to provide heat. Increased flow evens out the experience and it increases the floor's capacity to accept the energy, but it has little to do with how many BTUs are being delivered from the heat source. He needs to EXCEED the ability of the floor to disperse heat. This means the floor will warm up because it is past it's equilibrium with the surroundings. This is what makes floors "greet' us as we step on them, we have added more energy than they are delivering. Then, after the heat source is cut off, they coast down and deliver the excess over time.

These factors give maximum comfort, maximum stored energy, the best stability and the most even heating to the room. Hopefully, the BTU output of the heat source is sufficient to not only carry the heating load, but make the floor feel comfortable in the process. Even better is having enough power to warm it up in a fairly short time. This allows us to program the heating which brings in other desirable features. Of course, in this case, where it's wood heat, the fireplace needs to be as small as possible to do the job and programming is not really an option. That's why, again, that design is fun and always different. So many things to consider in each case.

 

[quicksandfarmer]

Of course, in this case, where it's wood heat, the fireplace needs to be as small as possible to do the job and programming is not really an option.

The other way to look at it is the way you program a wood stove is to put another log on!

 

[Raspy]

Yeah, we need a thermostat that throws another log on the fire in the middle of the night. I'd love that!

 

[CalG]

Call me stubborn, but in reading, I still suggest to shut down all floor circuits but one, and try to work out the system that way.

Why> Now the flow is too low. Shut down the size of the system, and perhaps the flow can be managed by valving.

Simplify,simplify,simplify!

 

[mx842]

mx,

Regarding plumbing the HX to the water heater: The HX is horizontal and it doesn't matter which line is in or out, but plumb the out to the top of the tank with as little restriction as possible and an up-slope all the way. This will help the thermosyphon affect and, during a power failure, will circulate some water. Assist it with a pump to get the maximum BTUs to your tank. The 007 should be fine for this duty. Use the 3/4" port on the side of the tank near the top, typically, this is the pressure relief port, as the return from the HX.

TEE the output to your floor pump at the same upper side outlet of the tank, where the HX line ties in. Simply screw a brass or galvy tee on a short nipple sticking out of the tank fitting and plumb each line to it. Return the floor loop to the cold inlet on top of the tank. Pull the dip tube out of the cold inlet fitting and set it aside. Put an automatic air vent on the hot outlet. If you insist on an auto fill system, tie it in at the cold tank inlet with another tee.

I suppose the best flow pattern would be to place the HX pump on the bottom line pumping away from the tank toward the HX. Place the floor on the line from the upper side fitting tee, pumping toward the floor. Return it to the cold top fitting as mentioned.

All of this will tend to move air to the tank and get it out of the system. It will not inhibit thermosyphon.

Install tees on the supply and return lines to the floor. One just before the pump, which is pumping toward the manifold. And one after the manifold on the return line to the tank. Connect these with a 3/4 pipe and a ball valve. This will allow you to throttle a bypass and control the supply temperature somewhat. It also reduces restriction in the floor loop. You'll still get all the available energy to the floor, but you'll avoid scalding temperatures and you'll help the pump purge the air. A bit of manual adjusting of this ball valve will get you find the right setting. Probably 1/2 open for starters.

Install a ball valve on the return line to the tank from the floor. Preferably, near the tank. Then tee in a hose bib right before this valve (on the manifold side). You can then close the valve and open the bib to manually purge the floor loops. This will help you get it all started and force out any air.

Again, look carefully at the manifold to make sure you have the loop restrictors fully open. Very important.

Man!! I wish I had seen this before I piped this thing up. Sounds like I still have some work to do. When I asked this question I had two 20 ft sticks of copper tube, one 1" and one 3/4 and a bunch of fittings. I used that all up the first time then when I re-piped it. Then when I took the mixing valve out I re-arranged the whole setup re-using what I could save. I was thinking about teeing off the top port fitting like you just suggested but I wasn't sure if that was right or not. I instead used the two side ports on the tank where the old heating elements were. I plumbed the top heating element port out to the supply for the loops and the bottom one I used for the return to the HX. Thought being that these are 1" ports and I figured they would be less restrictive than using the 3/4 ports on the tank. I have the pump on the hot side of the HX and it goes to the top port where the T&P valve was. I moved the T&P valve to the hot top port of the tank.

I have a 4 way cross tee piped into the bottom drain on the tank. I came out of that with a boiler drain valve in one port and I sent the return water from the loops in to one of the other ports. I had made up this contraption out of 3" copper I had laying around that I was going to use the other port of the 4 way that I could use to add chemicals to the system without having to shut everything down. Right now that is plugged off, I installed it the first time I re-piped it but removed it when I took the mixing valve out. It looks like I need to go back to the supply house for another stick or two of copper and another box of fittings. I did fire it up last night and it seemed to work pretty good though the way it is.

I found a couple of old aqua stats that I took off an old boiler a few years ago and I put one in the HX to control the 007e. this one has a differential lever on it that you can set. I set the control to cut the pump on at 180 degrees but with a 30 degree differential and that worked pretty cool. Of-course by the time it got up to the 180 mark I had hot water already passing through the pump and going into the tank but it did speed things up once the pump turned on. The biggest problem I still have is that I'm not getting the flow I need through the loops.

The Taco 1/8 HP pump I figured would do the job but I guess it's just not enough. I finally fixed that pump and got it to stop leaking. I took it apart three or four times and put it back together and finally I got it to stop leaking. I had put a new seal in it with new gaskets and I still don't know what fixed it but it seems to be ok now. However when it starts it still squirts a little water out of the oil hole but stops pretty quickly once the pump has come up to speed. I'm still thinking I have some air somewhere that needs to work it's way out.

Another thing I noticed is that one of my air bleeders gets water logged and dumps water all over the floor. It's one of those copper looking things that looks like a small water tower and I just bought it. What would cause one of those to get water logged like that?

I may leave it like I have it now and re-do it like you suggested later just to see what it does. There are a couple other things I want to change but it's starting to cool down now I probably ought to leave it as it is for now and do it all over again next summer.

 

[Raspy]

I stil think you have not opened the loop restrictors properly. Go back and be absolutely sure how the manifold works and get the loops opened up. There are various ways these restrictors work, but they usually have a shut off valve on one side of the tubing loop that is just a screw down cap. Screwed down shuts it off. The other side has an adjuster that can be opened with an allen wrench or a square drive key. I'm talking about where the PEX loops connect to the manifold. Find out how your works and get it opened up on each loop. While the spring drive Taco is not the pump I'd pick in a perfect world, it should do much more than you are getting. The purging system with ball valve and hose bib on the return line from the manifold will purge any air out.

Once the air is out and the loops are opened up, you should get good flow to the floor. This is the area for you to fix next. Then you can do something to the rest of it if needed. Right now you need delivery to the floor and all pieces of the puzzle seem to be in place except proper flow.

If you cannot get flow through the loops here's what you have to do: Disconnect the loops and blow through them one by one. Make sure that each one is clear and make sure that each loop is connected to the supply on one end and the return on the other. If you accidentally connect each end of a given loop to the supply side, you get no flow. It's obvious, but it is an easy mistake to make and I've done it more than once and it's sometimes hard to spot once you are trying to start the system.

Getting the floor loops flowing is your next problem to solve. You should be seeing 1 gpm or more in those loops with the setup you have.

 

[Raspy]

Call me stubborn, but in reading, I still suggest to shut down all floor circuits but one, and try to work out the system that way.

Why> Now the flow is too low. Shut down the size of the system, and perhaps the flow can be managed by valving.

Simplify,simplify,simplify!

Well, OK, I'll call you stubborn if you wish. I'd agree if the pump was seriously undersized, but it's not.

I don't see how shutting off all loops but one will help. The flow is already restrictive and shutting more loops off adds more restriction. The pump has enough power to feed more loops with about the same output head as if it was feeding one. I think there is another problem with the loops or the restrictors. Clearly, whatever is happening, it seems the delivery problem must be solved next.

This is the kind of thing that makes hydronics interesting. There is always more to it than first expected and lots of room to fine tune.

You're right about simplifying all systems. It's been my mantra for years. And it's not an easy thing to do!

 

[dadster4]

Using the system pump to try to remove all air from the loops often doesn't work...not enough pressure and volume to force air out of the lines.
This always works: 1-Attach a garden hose (under house pressure) to the return manifold drain valve. 2-Attach another garden hose to the supply manifold drain valve with the other end running to a drain. 3- Isolate both manifolds from the system using the red and blue valves. 4- Close ALL the balancing valves. 5- Turn on the water supply to the first garden hose and open the drain valve on the both manifolds. 6- Open each balancing valve one at a time until no air is present coming out the drain hose(you can usually feel and hear this happening also). 7- Close the drain valves at the manifolds and remove the hoses. What this does is use city or well water pressure to force out any air blockages in the loops.
I'm glad you weren't hurt when your system overheated, I have designed and installed many in-floor, floor-warming, and hydronic systems, and know you are doing the best you can with what you have to work with, please be careful

 

[Raspy]

Using the system pump to try to remove all air from the loops often doesn't work...not enough pressure and volume to force air out of the lines.
This always works: 1-Attach a garden hose (under house pressure) to the return manifold drain valve. 2-Attach another garden hose to the supply manifold drain valve with the other end running to a drain. 3- Isolate both manifolds from the system using the red and blue valves. 4- Close ALL the balancing valves. 5- Turn on the water supply to the first garden hose and open the drain valve on the both manifolds. 6- Open each balancing valve one at a time until no air is present coming out the drain hose(you can usually feel and hear this happening also). 7- Close the drain valves at the manifolds and remove the hoses. What this does is use city or well water pressure to force out any air blockages in the loops.
I'm glad you weren't hurt when your system overheated, I have designed and installed many in-floor, floor-warming, and hydronic systems, and know you are doing the best you can with what you have to work with, please be careful

Yours is a good way to get some air out. Although it cannot be used on other, older systems with weak tubing that cannot stand city water pressure. And not every system has isolation valves on the manifolds. Plus the air in the connecting piping must be removed too, otherwise the pump will still be air-locked and air will get to the manifold loops from the piping. There is no need to reverse flush the air out. Reverse flushing is only needed for removing junk in contaminated systems and it cannot be used if there are check valves in the system. Flow meters also have a hard time with reverse flushing as they tend to slam closed. Forward flushing with the correctly installed valving will move all air to either a hose bib or an air vent.

Every properly designed system has a valve arrangement that will allow flushing of the entire system, including the pump and all lines, simply by turning one valve and opening a bib. Then opening each balancing valve as needed to flush the loops one by one.

A simpler way is to use the ball valve and hose bib on the return line as I recommended. Then use the auto fill system already installed to flush out each line simply by opening the hose bib and closing the ball valve. This also works to assist the pump in a startup by flushing the pump as well as the loops. It takes very little pressure to move air out. The final bit of air always in the system will find it's way back to the tank and get out through the air vent on top.

In this case we don't know if there is air in the system or if the manifold loop restrictors are closed too far or if there is simply air in the pump itself.

 

[mx842]

Yours is a good way to get some air out. Although it cannot be used on other, older systems with weak tubing that cannot stand city water pressure. And not every system has isolation valves on the manifolds. Plus the air in the connecting piping must be removed too, otherwise the pump will still be air-locked and air will get to the manifold loops from the piping. There is no need to reverse flush the air out. Reverse flushing is only needed for removing junk in contaminated systems and it cannot be used if there are check valves in the system. Flow meters also have a hard time with reverse flushing as they tend to slam closed. Forward flushing with the correctly installed valving will move all air to either a hose bib or an air vent.

Every properly designed system has a valve arrangement that will allow flushing of the entire system, including the pump and all lines, simply by turning one valve and opening a bib. Then opening each balancing valve as needed to flush the loops one by one.

A simpler way is to use the ball valve and hose bib on the return line as I recommended. Then use the auto fill system already installed to flush out each line simply by opening the hose bib and closing the ball valve. This also works to assist the pump in a startup by flushing the pump as well as the loops. It takes very little pressure to move air out. The final bit of air always in the system will find it's way back to the tank and get out through the air vent on top.

In this case we don't know if there is air in the system or if the manifold loop restrictors are closed too far or if there is simply air in the pump itself.

The more I think about it the more I think I have air in the system. I can tell the flow meter valves and the flow restrictors on the return manifold are working because I can loosen the loop lines where they feed into the manifolds and open and close each valve and see water flowing and stopping as I turn each valve. About a week ago I was messing around with the valves I would close the restrictor and loosen the loop connection on the return manifold and with the pump running I could bleed that loop. Some of them did have air trapped in the loop some did not but once I did this to all of them awhile later all the flow meters were reading like they should accept two of them on the far end of the manifold. I don't remember exactly they were reading but they were down in the mid to lower part of the flow meter probably a little more than a gallon per min. I played with the two that would not flow for a couple hours and never could get the flow meters to show very much flow at all.

I made a few changes that required me to drain the system and I haven't been able to get that flow back. I'm still trying to figure out why that one air valve gets water logged. I don't know if it is just bad or if something else is causing this.

I think that once the thing cools down I'm going to add that purge valve before the main return valve and see how that works. Although I do hate draining it again and loosing all my conditioner but I guess that's better than what I have going on now, it's only money........Right!

 

[DaveK OR]

If your building is properly insulated, you don't need very much "heat" at all in the floor slab. You should not even notice that the floor feels warm even to bare feet.

On the other hand, in an old, drafty, and poorly insulated building, you may have to pump that floor temperature up to maybe 85F or higher to maintain room air temperature at a comfortable level. That nice toasty-warm floor in a comfortable-temperature room means that you are wasting a lot of energy.

 

[Raspy]

mx,

It sounds like you are fighting air.

Install the purge valve and hose bib I suggested and purge out each loop separately. This will purge the entire system easily and get you started. It will also for air out of the pump. If another unexpected batch of air gets to the pump later, just leave the system running, close the ball valve and open the bib to get it on down the line and out of the pump. Then shut the bib and open the valve so the pump can take over. This little help will only take a few seconds and is very handy. Hopefully, the air will separate at the tank and go up to the air vent. I wanted you to plumb the tank so that it could act as an air separator. Tanks are wonderful for this. For instance, in my hydronic solar system I have about 20 gallons of air all the time. This remains in the tank while collecting and gravity flows to the collectors when shut off to prevent freezing. Air management is one of the most important aspects of hydronic systems. It can help you or hurt you. In your system it's important to have it all eventually gone.

Make sure you have flow in all loops and the air is down to a minimal amount that can get out at the vent. The trick to balancing the manifold is to get everything going and see what the flows are. Start with all restrictors wide open. Some loops will be flowing less than others. Gradually reduce the fastest flows with the restrictors until the slower flows come up to match and everything is even. Done. Then, if needed, you can reduce the flow to certain loops to reduce the heat in those areas a bit. You might want the most heat at the workbench and less near the door, or less heat in the bedroom and more in the bath, for instance, to maximize your comfort. The air temp in a given area will be about the same, but the floor temp could be warmer where you stand than where the car sits, for instance. This is how you maximize the comfort and tailor it to your needs.

Don't add any conditioner until you have the system completely dialed in, leak checked, air out and heating as you'd like. All done. Then add the conditioner as a last step. Don't do it until you are sure you're done. It won't hurt a thing for it to run a while with no conditioner and prove itself.

When you say "waterlogged" I assume you mean the auto-vent is dribbling out the top. They are always full of water, but should not dribble more than a few drops while venting. If yours does, you can let it go until all air is out and then close the top cap, or you can take it apart and get the debris out of the needle valve, or you can wait for more air to go through to clear it, or you can tap on the side of it a bit to wake it up, or you can replace it. You can also install a ball valve below it and just shut that off when the air is gone. The valve will also allow you to remove it later without draining any water and losing more conditioner.

Sounds like you are almost there. Good luck! Are you having fun yet????????

 

[CalG]

mx,

It sounds like you are fighting air.

Install the purge valve and hose bib I suggested and purge out each loop separately. This will purge the entire system easily and get you started. It will also for air out of the pump.

See,!!!!! "Call me stubborn, But I would close down, and deal with one circuit at a time."

Jeesh! How the mind complicates the simple facts. ;-)

 

[Raspy]

See,!!!!! "Call me stubborn, But I would close down, and deal with one circuit at a time."

Jeesh! How the mind complicates the simple facts. ;-)

Well why didn't you say so!!! Oh, wait......