How do I accomplish this radiator install idea?

[Mr. E. Bernstein]

I am a handy homeowner. I can do some heating/plumbing work, but in 2011 I hired a professional installer to replace my heating system which was forced hot air/furnace with separate water heater. The new system is a boiler and indirect DHW. Since the house never had radiators, I asked the installer to use the existing ductwork by connecting the boiler to an air handler. There is one zone valve. This design has worked without problem.

However, there is a modification I would like to make. When the boiler was installed, I asked the installer to put 2 extra taps (3/4") on the supply and return manifold. The third tap is for the air handler (1"). The DHW is not on the manifold. It has its own taps from the boiler pipe (I believe you call that pipe the header). So there are 3 circulators: the boiler, the entire manifold and the DHW. I want to use the extra manifold taps as follows:

While I didn't want to install radiators for the whole house due to cost and damage to floors, walls, etc., there are two areas in the house where I would like to put radiators for supplementary heat. I want to be able to control these radiators independently of any call for heat from the house thermostat. In other words, I want them to have their own thermostat wired to the boiler, not thermostatic valves on the radiators.

I'm pretty sure I can do the plumbing connections myself using pex. I have worked with this before, but I don't know how to make the electrical connections to the boiler. I assume I will need zone valves on the two 3/4" manifold taps. Can I carry out this idea using the existing air handler circulator? I prefer not to have the air handler come on when I want extra heat from the radiators unless there happens at that moment to be a call for heat from the house thermostat. In that case, it doesn't matter if the air handler also comes on.

Besides the wiring, am I missing anything that I need to know? I assume I will need ball valves and check valves in series with each radiator and a separate wall thermostat in each room where I want the two radiators. Another issue I wonder about is how to fill the radiators and tubing before connecting to the manifold. Are any additives needed in the water? The boiler is in an unheated basement but I doubt it ever drops to 32 degrees. Can anyone tell me what I need to know to do this or know where I can find any literature to explain?

It would also help me if someone could explain the exact sequence of events in the boiler when a house thermostat calls for heat from a boiler connected to an air handler. The boiler is WM Ultra 105 ser. 3. It occurred to me I might need a heat exchanger, but it would have to be fairly small to serve just two radiators or 1 for each radiator. I don't think they make them that small. I don't have an idea of the radiator BTUs, but they are wall radiators about 3 foot square. I already have them so I want to use what I have. Thanks

 

[Dana]

You're clearly in over your head here. Hydronic heating design is way more than a plumbing problem. But let's run though the high points:

The WM Ultra 105 is about 3x oversized for most homes in NYC/Westchester/L.I. and 2x oversized for most homes upstate. But it might need to be that big if the hydro-air handler is similarly oversized. Got a model number on the air handler? How about part numbers for the panel rads?

Because it IS such a large boiler you're likely to run into short-cycling issues on zone calls if adding a couple of separate micro-zones. The minimum-fire input to that beast is 21,000 BTU/hr, and at condensing water temps the output is about 20,000 BTU/hr. (That is more than half the heat load of my house at 0F outdoors.)

Unless there is enough radiation on each of those radiator zones to emit a large fraction of that 20K (at least 15,000 BTU/hr) at an average water temperature of 120F it's going to short cycle itself into an early grave. A typical euro style ~3' square panel rad delivers about 5000 BTU/hr at an average water temp of 140F, about 3500 BTU/hr at an AWT of 120F, which is what it needs to be to deliver 95% combustion efficiency. I doubt even together as one zone it's going to be enough radiation to keep the boiler from short cycling.

There isn't enough thermal mass in that boiler to help much, and the radiators would have to be ludicrously oversized for the zone loads to suppress short cycling on the boiler. But for the sake of arguement, run heat load calculations on the two rooms/zones in question using the most aggressive assumptions possible about air tightness and R-value with a freebie online tool such as LoadCalc. In the unlikely event that the loads of both radiator rooms together add up to more than 15,000 BTU/hr this might be made to work (as just a single zone, not two) without buffer tanks or crazy-oversized rads.

If the loads don't add up to that much, or must be separately controlled, setting up a buffer tank as a thermally-massive hydraulic separator, with the zone rads sipping from the tank can work. It starts adding up to quite a bit of money though, so you have to really run some math on it all to keep costs under control- don't just hack at it or you can spend a lot of money on something that has to be yarded out and replaced to make it work.

If I understand the configuration correctly, the boiler is plumbed primary/secondary (probably closely-spaced tees as the hydraulic separator?), with one pump serving the manifold loop, which is currently serving only the hydro-air, and another pump (maybe plumbed as another secondary, maybe not) serving the water heater? Some pictures of the pump & manifold configuration would be useful here.

A heat exchanger won't help any of this.

 

[Sponsor]

 

[Mr. E. Bernstein]

You're clearly in over your head here. Hydronic heating design is way more than a plumbing problem. But let's run though the high points:

The WM Ultra 105 is about 3x oversized for most homes in NYC/Westchester/L.I. and 2x oversized for most homes upstate. But it might need to be that big if the hydro-air handler is similarly oversized. Got a model number on the air handler? How about part numbers for the panel rads?

Because it IS such a large boiler you're likely to run into short-cycling issues on zone calls if adding a couple of separate micro-zones. The minimum-fire input to that beast is 21,000 BTU/hr, and at condensing water temps the output is about 20,000 BTU/hr. (That is more than half the heat load of my house at 0F outdoors.)

Unless there is enough radiation on each of those radiator zones to emit a large fraction of that 20K (at least 15,000 BTU/hr) at an average water temperature of 120F it's going to short cycle itself into an early grave. A typical euro style ~3' square panel rad delivers about 5000 BTU/hr at an average water temp of 140F, about 3500 BTU/hr at an AWT of 120F, which is what it needs to be to deliver 95% combustion efficiency. I doubt even together as one zone it's going to be enough radiation to keep the boiler from short cycling.

There isn't enough thermal mass in that boiler to help much, and the radiators would have to be ludicrously oversized for the zone loads to suppress short cycling on the boiler. But for the sake of arguement, run heat load calculations on the two rooms/zones in question using the most aggressive assumptions possible about air tightness and R-value with a freebie online tool such as LoadCalc. In the unlikely event that the loads of both radiator rooms together add up to more than 15,000 BTU/hr this might be made to work (as just a single zone, not two) without buffer tanks or crazy-oversized rads.

If the loads don't add up to that much, or must be separately controlled, setting up a buffer tank as a thermally-massive hydraulic separator, with the zone rads sipping from the tank can work. It starts adding up to quite a bit of money though, so you have to really run some math on it all to keep costs under control- don't just hack at it or you can spend a lot of money on something that has to be yarded out and replaced to make it work.

If I understand the configuration correctly, the boiler is plumbed primary/secondary (probably closely-spaced tees as the hydraulic separator?), with one pump serving the manifold loop, which is currently serving only the hydro-air, and another pump (maybe plumbed as another secondary, maybe not) serving the water heater? Some pictures of the pump & manifold configuration would be useful here.

A heat exchanger won't help any of this.

Thanks for your technical assistance. Photo attached. I can see what you mean about high output to a small load, but the rads would not be on for long, maybe a 5 degree rise in room temp. would be enough, max. 10 degree. What if the thermostat/controller was wired just to use the standby boiler temp. without firing the boiler, in other words to activate the manifold circulator but not the boiler. I discovered online another issue that would have to be dealt with. I have a supply of 1/2 inch pex so that is what I want to use to connect to the radiators. The air handler is connected to 1 inch tap on the manifold. If the zone valve to the air handler were to come on while I am trying to heat the radiators, it would just about cut off the flow to the other taps on the manifold (path of least resistance). That very problem appeared on a TOH episode where their master plumber had to install zone valves on each manifold tap and a controller to solve the problem of low flow to the smaller tap. I assume I would also need a controller. I may need professional help with that, unless the wiring to the controller, zone valves, boiler, etc. can be explained to me and I can understand it. I am not in a hurry, so I can take whatever time is required to try to solve the issues or not do it at all. P.S.: There are no specs on the rads. They are custom extruded aluminum panels with 1/2" ID copper serpentine pipe. That's why I can't give you specs. The length of copper is about 50 ft. with about 25 ft. of pex each way between just one rad and the boiler. I am just interested in trying 1 rad for now to one of the two available taps on the manifold. I don't see any specs on the label on the air handler. It is a First 36VHBXB-HW "two stage" with variable speed motor (I set it very slow speed so it's quiet).

 

[Dana]

What water temp are you feeding the 36VHBXB-HW?

At an entering water temp (EWT) of 140F, and the blower at it's lowest speed it delivers ~30,000 BTU/hr give or take 2000 BTU/hr, depending on the water pumping rate. At higher temps it puts out a lot more, at lower temps it puts out less, but we'd need to estimate it, since it's not spelled out at temps lower than 140F in the short form spec.

I don't see the picture attachment (perhaps too hi-res?)

When any zone is calling for heat the zone controller calls the boiler. The boiler temperature rises to some level above it's programmed setpoint and turns the burner off off, but the circulators keep running. When the system cools down to a few degrees below it's setpoint it will re-fire. But there is very little thermal mass in the boiler and the described panel rad, so if it's not emitting the lion's share of the minimum-fire output those overshoot/stop/refire/repeat cycles happen pretty fast. I suspect based on the radiator description, if serving just one panel rad the burn cycles on an Ultra 105 would be under a minute, with potentially dozen or more cycles per hour, a mode that will shorten the life of the boiler. With a thermally massive buffer tank serving as the hydraulic separator the burn cycles are much longer, with fewer burns per hour.