Can't get heat to last 3 radiators on loop. Circulator pump?

[aarpcard]

Hello All,

I'm not a trained plumber so forgive any mistakes on my part. I have (I think) a fairly complex hydronic heating system in my house. The majority of the system is 112 years old comprising of large cast iron radiators and black pipe. I believe this system was originally 2-pipe steam but was converted to water at some point.

The boiler is 1991 vintage.

At some point the original radiators in the kitchen and 3rd floor were removed from the original loop and replaced with baseboards in the kitchen, and a smaller radiator in the 3rd floor bathroom on a new, all copper 2nd loop with it's own circulating pump. (electric baseboards on the rest of the 3rd floor). I estimate this was done around 2006.

The system operates at 18PSI. All of the radiators have been bled and balanced (and the majority have new shutoff valves).

The issue I'm having is the radiators circled on the right of the below schematic at the end of the primary loop, take an extremely long time to heat up. The boiler has to be cranked for 12+ hours before they get any heat.

My guess is it's a circulation issue because they are at the end of the loop (and the last two are on the second floor). Checking the pressure in these radiators from the bleed valves indicated full pressure (18PSI) so I'm pretty sure it is not a blockage.

My uneducated guess is I need a second circulating pump somewhere on the primary loop, but I don't know where nor if that is the right solution. Any ideas/advise would be appreciated. See below roughly drawn schematic.

 

[WorthFlorida]

Was this always a problem since you owned the home? The rads in question, did you open the bleed valves while the circulator was on? Did they squirt out about the same amount of water as the others? Just trying to prove if you have good circulation or just running out of heated water. Have you tried closing off other rads to to reduce the heat load and checked if these cold rads got warmer faster?

 

[Sponsor]

 

[aarpcard]

Was this always a problem since you owned the home? The rads in question, did you open the bleed valves while the circulator was on? Did they squirt out about the same amount of water as the others? Just trying to prove if you have good circulation or just running out of heated water. Have you tried closing off other rads to to reduce the heat load and checked if these cold rads got warmer faster?

Yep, this has always been a problem. All three do squirt about the same water as the others, however the 2nd floor bath radiator is a bit weaker (however there are weaker radiators elsewhere in the loop that don't have the heating issue).

I have tried closing off the majority of the other radiators to see what happens to these three. The 1st floor pantry will heat up decently fairly quickly, but the other two on the second floor still take hours to heat up with the boiler fully cranked and they are always significantly cooler than the other radiators in the system.

 

[TEch99]

Just curious if that's a Monoflo setup with a loop in the basement. If so ensure the arrows are pointing in the right direction on the Monoflow valves.

You can see this arrow faintly on the valve below. If it is indeed a Monoflo, I would investigate around these valves.

https://www.supplyhouse.com/Legend-...MI0I3PhfK_7AIVpf7jBx1nMwD9EAQYASABEgKPdfD_BwE

 

[Nebojsa]

Cant say if this is the problem you have. I could not heat the top units (2nd floor) but the bottom units were fine. After troubleshooting,changed the boiler pump. The impellers were worn down and could not move the water.

nebojsa

 

[Dana]

I think TEch99 may be on the right path. Are you confident that it's sketched correctly? (This one might be hard to nail down in a web-forum analysis.)

If the system is operating at 18psi at a boiler in the basement there is no way it should be reading 18psi at the second floor radiators. The pressure drops 0.433 psi per foot of elevation, so ~25' above the gauge on the radiator that reads 18psi it should be reading ~7 psi (when no pumps are operating), not 18psi.

 

[WorthFlorida]

I think TEch99 may be on the right path. Are you confident that it's sketched correctly? (This one might be hard to nail down in a web-forum analysis.)

If the system is operating at 18psi at a boiler in the basement there is no way it should be reading 18psi at the second floor radiators. The pressure drops 0.433 psi per foot of elevation, so ~25' above the gauge on the radiator that reads 18psi it should be reading ~7 psi (when no pumps are operating), not 18psi.

Dana, I learn something new every time on your post. Usually near the boiler there is an expansion tank. As a kid every summer my dad would drain it and I be carrying about 20 pails of water up a fight of stairs to dump the water outside. With an old fashion non bladder type tank, could a fully saturated tank cause this pressure reading?

 

[Dana]

Dana, I learn something new every time on your post. Usually near the boiler there is an expansion tank. As a kid every summer my dad would drain it and I be carrying about 20 pails of water up a fight of stairs to dump the water outside. With an old fashion non bladder type tank, could a fully saturated tank cause this pressure reading?

A saturated expansion tank could raise the pressure on the system everywhere when the system was hot, but the pressure difference in pressure between a basement boiler and a second floor radiator would still run about 10-12psi when no pumps are running. Nothing will cancel the effects of gravity, even though the density of 180F water (peak system temperature) is measurably less than the density of 100F water (after a long idling period between burns.)

 

[Fitter30]

Does the system have any automatic air vents? Does the steel expansion tank piped into the bottom or end? Where is the make up water piped in relationship to the expansion tank and any horizontal piping angled up towards the tank? Brand and model of pump on problem radiators and other pump? 1st floor kitchen heat better than 3 rd floor?

 

[aarpcard]

Thanks Everyone for the replies.

Just curious if that's a Monoflo setup with a loop in the basement. If so ensure the arrows are pointing in the right direction on the Monoflow valves.

You can see this arrow faintly on the valve below. If it is indeed a Monoflo, I would investigate around these valves.

https://www.supplyhouse.com/Legend-...MI0I3PhfK_7AIVpf7jBx1nMwD9EAQYASABEgKPdfD_BwE

I didn't find any monoflow valves in the system. The 1st floor kitchen/3rd floor bath radiator circuit is plumbed with 3/4" copper pipe. The rest of the circuit is a two pipe system with 4" black pipe running a loop around the basement. 1.25" black pipe or 1" black pipe is tee'd off from the 4" feed/return circuit to each of radiators.

The inlet and outlet from the boiler is 1.25" copper pipe which is tee'd into the 4" black pipe circuit.

Cant say if this is the problem you have. I could not heat the top units (2nd floor) but the bottom units were fine. After troubleshooting,changed the boiler pump. The impellers were worn down and could not move the water.

nebojsa

It's possible. The pump for the original black pipe circuit seems to be a few decades old. Maybe it is undersized as well? It has to move a lot of water around. No idea, but I'd estimate well in excess of 200 gallons is the capacity of the black pipe circuit.

I think TEch99 may be on the right path. Are you confident that it's sketched correctly? (This one might be hard to nail down in a web-forum analysis.)

If the system is operating at 18psi at a boiler in the basement there is no way it should be reading 18psi at the second floor radiators. The pressure drops 0.433 psi per foot of elevation, so ~25' above the gauge on the radiator that reads 18psi it should be reading ~7 psi (when no pumps are operating), not 18psi.

I remeasured it at the radiators with the pumps off and yes, there is about an 6-8 PSI drop per floor. The affected radiators aren't outliers when compared to the pressures read on the other radiators on the same floor.

Does the system have any automatic air vents? Does the steel expansion tank piped into the bottom or end? Where is the make up water piped in relationship to the expansion tank and any horizontal piping angled up towards the tank? Brand and model of pump on problem radiators and other pump? 1st floor kitchen heat better than 3 rd floor?

The system does not have an automatic air vent. The steel expansion tank is piped into the bottom. The layout of the piping to the expansion tank is close to how I drew it (although it does go into the bottom and not the side like I drew it.) The makeup water input is maybe 8' total of pipe away from the expansion tank. It consists of a horizontal section about 2 feet long, a vertical section about 4 feet long, a horizontal section about 1.5 feet long (where the shutoff valve to the tank is) and another 6" section up to the bottom of the tank.

Brand and model of the pumps I'll post in a bit.

1st floor kitchen heat is superb (perhaps too much). 3rd floor radiator on the same loop is acceptable.

This may be complete nonsense, but is there a chance that the kitchen/3rd floor (which is not original) is "stealing" hot water from the boiler from the other circuit? Compared to the original black pipe circuit there is a small fraction of water volume in the kitchen/3rd floor circuit and significantly less piping - so water would follow the path of least resistance and prefer to go that route right? (Electricity is my forte - not plumbing.)

 

[aarpcard]

Below is the layout of the system at the main boiler. The two circuits of 4" black pipe run the perimeter of the basement - at least 120 feet each.

Below is the singular circulator pump for the original/black pipe circuit. Maybe it is undersized or degraded?

Below is the pump for the kitchen/3rd floor circuit.

 

[Fitter30]

Pump curve for grundfos page 43
https://s3.amazonaws.com/s3.supplyhouse.com/product_files/59896155-brochure.pdf
16 radiators pump moving 8-12 gpm. Grundfos pump doesn't produce enought head or gpm to take care of the problem 3 zones. Are problem zones piped into the basement? Take the grundfos pump for problem zones and another pump for the 13 radiators. If the system was originally steam the 4" mains would be pitched.

 

[Dana]

The gpm of a Taco-007 & Grundfos UP 15-42 may indeed be a bit undersized for many (but not all) systems with high volume radiation & distribution plumbing. Your system has a comparatively a low pumping head and higher gpm type of radiation & distribution than the 3/4" plumbing & fin-tube convector type systems those pumps are designed for, but the -007 puts out substantially higher flow at The radiators are all in parallel, have fairly low pumping head, and are splitting the flow between them, and with 3"-5" distribution plumbing there isn't a whole lot of pumping head to help even up the flows (without resorting to flow balance valves to tweak, or TRVs to throttle flow once a radiator reaches temperature.)

That said, the -007 puts out a bit more flow at 2' of head than the -15-42. Are the problem rads on the loop with the Grundfos?

This system may need some beastie-pump like a Taco -010 to even it all out, but I'd personally want to calculate the system flows & head before making that change rather than spending several hundred on a pump and still fall short. This isn't very well suited to "design by web forum".

But pumping is not the entire problem. The total surface area of un-insulated distribution plumbing in that basement probably has greater output than a couple of radiators (!) . Retrofitting them with 1" wall (minimum) fiberglass pipe insulation (about R4 or higher) would be worthwhile sooner than later, from both a system performance and energy efficiency point of view, given that the foundation walls appear to be completely uninsulated(?).

Is the basement the warmest room in the house on the coldest day of the year? It could be, or at least WOULD be if the foundation walls were air sealed insulated all the way from the slab to the subfloor to the current IRC code min R10 for IECC zone 4, or R15 for zone-5 :

Even if the rest of the house has at least 2x4/R11 type construction with R19 in the attic, and at least storm windows over wood sash single panes it's likely that the uninsulated foundation wall losses account for more than 15%, possibly more than 25% of the annual gas use, even if not actively intentionally heating the basement. Insulating the walls to the IRC code min will raise average temp of the basement and lower the heat load of the first floor, making first floor room temperatures more even room to room.

Air sealing the basement to the outdoors is even more important on a house with a third floor. The distance between the lowest air leak in the house and the highest leak defines the stack effect pressure. With a basement + 3 story house the negative pressure at the basement that higher stack effect draws in a lot more outdoor air into the basement than a 1 story house with the same sized leaks. Sealing both the top floor (ceiling & walls) and basement (windows & doors, foundation sill, band joist, any cracks, etc) can make a HUGE difference in comfort on the intervening floors.

Nate Adams (out of Cleveland) runs a home-comfort & efficiency consulting biz, and recently published a book on the topic. There are several free download chapters and short but informative videos covering the general topic on his Nate The House Whisperer website. It's worth checking out the Home Comfort 101 materials before making any big changes.

 

[aarpcard]

@Dana , @fitter30

Thanks for the replies. I definitely have some reading and research to do. To answer your questions:

-The problem zones are all piped into the basement. They are the farthest 3 radiators on the parallel 4" distribution mains. The mains are indeed pitched, so these three are also at the high end of the pitch (two are on the 2nd floor, one on the first floor).

-The problem rads are on the loop with the Grundfos. The 4" mains and all 16 original cast radiators are on the Grundfos loop. The Taco runs the baseboards in the kitchen and a single cast iron rad on the 3rd floor.

-The basement is warm in the winter - likely due to the lack of insulation on the 4" mains. The mains heat the floor on the first level so much that you can feel the temperature difference with your feet and see their path through the floor with a thermal imager.

 

[Fitter30]

The system looks like the is piped (at least the mains) reversed return . Thermostatic hot water valves like what danfoss makes for radiators and fin tube should help with a larger pump. Would want to find out what max water flow through boiler. If to much primary/ secondary loops.

 

[Sylvan]

I would suggest nonelectric valves on each radiator to balance out the system and install a B@G series 100 circulator on the return

Also if you can try to get a measurement of the highest radiator /baseboard above the boiler and that should be a good starting point of the actual pressure required OR on a cold fill note the hydrostatic pressure with circulator off

If your height is 20 feet for example it would be 20 X .433 = 8.66 PSI

To get the water to overcome friction loss etc you take 8.66 X 2.31 = 20 PSI

Then increase it by 4 PSI as a fudge factor to overcome friction loss and make sure there is enough positive pressure to allow for bleeding out air

So you need 24 PSI and still will be below the 30 PSI relief valve

 

[Dana]

Then increase it by 4 PSI as a fudge factor to overcome friction loss and make sure there is enough positive pressure to allow for bleeding out air

The friction losses won't be very high until & unless a beefier higher gpm pump is installed. Even with higher head 3/4" distribution plumbing and a reasonably sized pump a 3 psi fudge factor is good enough, not that it matters much. With parallel rads and 4" plumbing there's next to zero friction loss until it's pumping great gobs of water with a gia-normous pump.

If (reasonably) possible I'd personally be inclined to go with fitter30's recommended thermostatic radiator valves (TRV) for either for all 16 radiators on the loop with the problem rads, or at least the handful of rooms on that loop that are running warmer than the average, even before upgrading the pump. While it's a lot less labor to just swap in a monster pump, the TRVs offer a huge flexibility for balancing room temperatures and higher overall comfort. A 1/12 hp or bigger B & G 100 series pump runs north of $300, A Taco 010 about $250, a 1" TRV runs $50-60, so the out of pocket material cost isn't going to be huge difference even if doing all 16 rads on the loop.

There is clearly enough gpm available from a UP 15-42 to deliver enough heat from the boiler into the system, as well as enough radiation to get that heat into the rooms, but insufficient flow balancing control. TRVs would throttle back flow for those warmer-room rads, increasing the flow to the remainder of the rads on that loop. Since it's mostly-working, increasing the flow to the problem rads with valve tweaks on some of the others could get it to clear the hurdle on the weaker-flow rads, and deliver excellent room to room temperature balancing control.

 

[Fitter30]

What the brand and model of boiler?

 

[Dana]

What the brand and model of boiler?

Looks like a cast iron Dunkirk, but I need better goggles to read the model number on that label: