Why does a 1" zone valve have a tiny 3/8 port?

[Belmondo]

My indirect was originally installed with 3/4 rather than 1" (long story, salvaged unit tacked on existing zone manifold). I've gotten everything needed to redo the zone to 1", and when I looked at the valve, I was startled to see it has a 3/8 port! Isn't it stupid then to do all this work to speed the recovery time of the tank if the port is only 3/8"? Is this the normal ratio of port to pipe? Perhaps this is one of those pipe turbulence friction things of why a big pipe is needed but small valve port. I guess I've never looked at a 3/4 valve's port, is it 9/32?

I thought I was doing this to increase the GPM flow to the tank. Given the short run from boiler to tank, wouldn't I get as much bang by just upsizing the valve rather than tearing down the whole manifold to upsize the entire tank zone? A 3/8 port to a 3/4 pipe shouldn't be excessive flow? Or am I just not getting it?

Edit:
Researching valves. I see I can get a 3/4 White Rodgers 1361-102 that has a 23.5 Cv instead of the standard Honeywell 3/4 that has a 3.5 Cv. Is this the right move, get 6.7 times the flow, and bag upsizing the whole zone?

[Dana]

The Cv of the valve is but a tiny fraction of the total pumping head of the loop. Swapping it out will do very little for your loop gpm number, since the bigger pressure drops will across other components in the loop, though undersizing the plumbing & valve for the actual (or intended) flow isn't great design. Unless the valve is presenting a large fraction of the head at your needed flow (extremely unlikely), upsizing the valve diameter doesn't affect the flow in the loop.

The diameter of the plumbing is usually as big or bigger factor than the valve, if the loop long and undersized. But in a very short loop even undersizing the pipe doesn't matter much. The lions share of the pumping head in an indirect loop is the sum of the head of the boiler's heat exchanger and that of the indirect's heat exchanger, which typically runs on the order of ~10' @ 6-8gpm (but do the math on YOURS, at the flow YOU need.)

A valve with a Cv of 3.5 inserts:

...1 psi of backpressure (=0.43' of head) @ 3.5gpm

...2psi (0.87' feet of head) @ ~5gpm

...4psi (=1.7' of head) at 7gpm

...8psi (3.4' of head) at 10gpm.

A short loop of 3/4" plumbing might only add ~ 1' of head @ 6-8gpm compared to replacing it all with 1" plumbing.

So if your loop is only getting 4-5gpm rate looking to move up to 8 gpm note, even @ 5gpm the undersized valve is on the order of ~10% pf the total head, and upsizing the valve to one with a Cv=23.5 reduces the total head @ 8 gpm to 0.05', but that still only reduces the total head in the loop by about ~10%. That doesn't increase the loop flow by 6.7x, it only increases it by about 1.1x.

If you swap out the plumbing with 1" AND replace the valve with a 20+ Cv version you might improve the flow by ~15%, but not a whole lot more. If your indirect's performance is truly limited by low flow rather that other factors, bumping up the plumbing diameter & valve Cv won't solve a flow issue (even if it's moving in the right direction.)

Much depends on the head specs for actual boiler & indirect and how it's plumbed. A bigger pump might get you there, but whether that's the "right" solution depends on the rest of the hydronic design. The indirect loop may need/want a much bigger pump than the rest of the system, in which case a separate pump rather than a zone-valve approach may be called for.

[Belmondo]

Thanks Dana,

Not surprisingly, I'm a little confused. The heat consultant at my local General Plumbing Supply was so concerned about the 3/4 feeding the indirect that he sold me a priority zone control for it to speed it's recovery. But I found that it was taking so long to recover the rooms were getting cold. I turned off the priority and set the tank temp to 150, which seems workable in terms of supply. Now you say upsizing may not be that beneficial. What specs would help to determine this? It's a Triangle Tube HL-36 on a W-M CGi-4 with a Taco 007-5 circulator. I haven't the vaguest idea what the actual GPM is, the tank spec calls for a min 7 gpm, the Taco can give up to 20. The "head loss boiler side" of the tank is only 1.

I still don't understand "in my gut" how a hole a tiny fraction of the pipe cross section isn't the biggest source of added head.

[Runs with bison]

A valve with a Cv of 3.5 inserts:

...1 psi of backpressure (=0.43' of head) @ 3.5gpm

...2psi (0.87' feet of head) @ ~5gpm

...4psi (=1.7' of head) at 7gpm

...8psi (3.4' of head) at 10gpm.

Dana, unless I'm out of it tonight those conversions are reversed. 1 psi = 2.31 ft of H2O head.

[Dana]

I guess I was the one that was out of it, eh? Your right!

(mea culpa- I must need more coffee in the PM!)

psi= 0.434 x head

so,

head= psi/0.434 (and not psi x 0.434, as calculated.)

So at a nominal 7gpm it's more like 9 feet of head, which is comparable in magnitude to the head of the heat exchangers in the loop... (OK,just shoot me now! :-) )

Swapping it out for a valve with a Cv of 23.5 would deliver ~0.09psi @ 7.5gpm, or (0.09/0.434=) 0.2 feet of head, a big difference.

(What was I thinkin' ?)

[Belmondo]

So at a nominal 7gpm it's more like 9 feet of head, which is comparable in magnitude to the head of the heat exchangers in the loop... (OK,just shoot me now! :-) )

Swapping it out for a valve with a Cv of 23.5 would deliver ~0.09psi @ 7.5gpm, or (0.09/0.434=) 0.2 feet of head, a big difference.

(What was I thinkin' ?)

So, if I understand correctly, my gut is right, that swapping the zone valve will achieve far more than upsizing the zone to 1"? That the short run of 3/4 adds relatively little head? Or should I do both anyway?

[Dana]

That's right- the valve is a much bigger elephant than 3/4" plumbing on a short & simple loop. (If there are 27 ells in the loop it might be worth calculating though.)

[Belmondo]

That's right- the valve is a much bigger elephant than 3/4" plumbing on a short & simple loop. (If there are 27 ells in the loop it might be worth calculating though.)

I count 10 ells. I can definitely eliminate 2, maybe 3 if it made a difference. I guess I'll get a valve. It odd how the vendors don't list or sort the valves by their Cv, you have to look at the specs for each one. I had no idea it varied so tremendously. I think there's one more question: for $5 more I can get the 1" with 33 Cv. Why wouldn't I do this and mount it with reducers? That way I've optimized the valve totally and have my options open if I ever do want to upsize the whole loop.

I believe this is the right item, a 2 wire rather than 3 wire unit. I thought I was dense reading the instructions till I found a thread here saying the White-Rodgers manuals are the worst!
http://www.pexsupply.com/White-Rodge...ire-14292000-p

[Dana]

Each ell adds about an effective ~2.5 feet of length (not head) to the run, so 10 ells is like adding 25 feet of pipe from a pumping point of view, reducing it to 8 would be more like 20' of effective length, so if the rest of it adds up to about 10-15 feet of actual length it's an effective 30-40'.

At 7gpm 40' of 3/4"pipe has a head of ~7' of head, and at 5gpm it represents ~5' of head.

With 40' of 1" pipe @7gpm would run 2.5' of head, at 5gpm it's less than 2' of head.

So it'll still make a some difference to boost to 1" but not as much of a difference as the valve with the higher Cv.

Most heating zones are designed with flow rates between 1-3 gpm and have more than 100' of pipe, so a crummy zone valve with a Cv of 3.5 still isn't as much as big a fraction of the total head as with high-flow applications like indirect HW heaters. At 2gpm 150' of 3/4" PEX has a head of ~3', but the valve with a Cv of 3.5 has a head of only ~0.7 feet.

The CGi-4 has ~74KBTU/hr of output, only ~ 2x that of a typical 40 gallon standalone tank heater, so it will always have a recovery period no matter what flow rate you run through the HX. At 5gpm flow on the loop (2500lbs/ hr) the full 74K would have a delta-T of ~30F which is tolerable- the boiler won't cycle. At 7gpm (3500lbs/hr) that drops to about 21F, which is also fine. I doubt you're flow and heat-exchanger limited so much as much as you are burner-output limited in terms of the recovery rate. If you're not running it in priority mode the recovery rate will depend on whether your other zones are calling for heat.

[Belmondo]

Thanks so much Dana. I guess what I'm hoping to achieve is running it in priority without it taking so long to recover as to get noticeably cooler anywhere. If I can lower the tank temp too, I'll cut the standby losses, an issue in summer. I've replaced the tank thermostat with a settable gap digital temp controller, I've got it set max 150, min 135. The damn OEM one was cycling the boiler all day in summer with no usage, and I'd already blanketed it, heat trapped it and insulated the pipes. It still loses more than 1 deg/hr, well over spec of 1/2 deg, but now it can go from cycling after morning showers to dinner washup without cycling and the CGi can go cold all day. A neighbor has a similar CGi fired system with the same all day cycling, I've ordered a temp controller for him too. It was driving them crazy, their utility room is a sauna!

[jadnashua]

If you can add some insulation in the house, running that tank as a priority zone wouldn't make as much of a difference. Also, search for air leaks. Some of the utilities will do an energy use assessment for free or a very low cost and then subsidise fixes. These can often pay for themselves in the first year.

[Belmondo]

Insulation is a sore subject. I live in a >100 year old brick rowhouse that started life as a cold water flat tenement. The walls are horsehair plaster & wood lathe on 3/4" furring strips nailed to the brick. There's really nowhere to insulate! I've replaced the windows, but I'm not going to gut down to the brick and stud them out to be able to insulate, the rooms are small enough as is. On the other hand we live small, and are attached on one side, so it's not as bad as like a big ancient farmhouse. I suspect the heat loss isn't as bad as it could be. When I was running radiant under my kitchen floor the radiant gurus said with no insulation it would never work and I'd have to supplement it. It's worked like a charm, and no overheated floor either.

[Dana]

With a foot of brick, 3/4" of air and plaster/lath you're currently looking at about R2.5, best case.

In a NJ climate the 3/4" cavity between the brick & lath can usually be insulated with non-expanding injection foam without causing other issues. With a 3/4" cavity it would add ~R3 to the average wall R by filling those cavities with non-expanding injection foam, more than doubling the effective R value of the wall, cutting the wall losses in half. (See: TriPolymer, Corefill500, there are others, google "non-expanding injection foam".) This is enough to make both a fuel-use difference and a comfort difference at the temperature extremes. It's not cheap, but it'll be worth it on comfort grounds alone, and still cost effective on reduction of heating & cooling bills.

Do NOT substitute expanding polyurethane foam pours for this appication- the risk of bowing or even blowing out the horse hair plaster is high, especially with the rusting 100+ year old nails holding the furring to the brick. Injection foam is fairly low pressure during installation, much lower than the typical pressures reached by 2-lb polyurethane pours.

Non-expanding injection foam will also reduce the air infiltration. Once you're at ~R5 walls and have double-pane windows (or tight fitting storms) air-infiltration begins to loom large in the heat load numbers. Not all air leaks are the same- concentrate on sealing any penetrations into the attic first, then any leaks into the basement to reduce stack effect drives. When you've run out of places to seal, it may be worth hiring an insulation company specializing in air sealing to blower-door test & rectify the other 3 square feet of leak that you didn't find.

If you have reasonably tight single pane windows, adding low-E storm windows can be both cheaper & higher efficiency than low-end replacement windows.

Before you modify the hot water loop for higher flow, set the boiler's high limit pretty high (say 210F-220F), switch the zone controller to run the indirect on priority, and set the indirect to something low, like 115F-120F. Then back-to-back showers or something to deplete tank temperature to where it fires (or run the shower until the output is tepid,) then crank the aquastat back up to 150F and observe the boiler's operation during the recovery ramp.

If the boiler temp runs up to the high-limit kicking off the burner before the indirect reaches it's setpoint, then re-fires a few minutes later, improving the flow will improve the recovery rate and overall hot water performance.

But if the tank always reaches the setpoint without the boiler reaching the high limit, improving the flow won't reduce the recovery period at all.

Improving the flow may still be worth doing, since it reduces the delta-T on the boiler: A 20-30F delta-T doesn't put much mechanical stress the boiler but over 50F is probably beyond the operating spec for that boiler (I'd have to look it up to be sure but most cast iron boilers have a 50F limit on delta-T). Running them beyond spec reduces the service life of the boiler by warping the heat exchanger plates and/or stretching the through-bolts, causing it to leak when cold (or after awhile, even when hot.)

If you have an infra-red thermometer you can measure the delta-T directly by putting a spot of high-temp spray paint at both the boiler output and the return input (to equalize the IR-emissivities on those points), then measuring & subtracting the temperatures for the delta.

Lower pumping head also reduces the power used by the pump, but that's probably the least of your issues.

[Belmondo]

Dana, I don't need to experiment, as my workdesk is a few feet from the boiler. I've already observed it kick off while the tank is still calling, and kick back on some minutes later.

As for insulation, I already have double paned replacement windows. Certainteed were top rated by Consumer Reports, I'm not terribly impressed however. Blowing in foam would require a gazillion holes in the walls! In retrospect, I could have done it before renovation, but not now. And I have one 13' exposed brick wall. I can't imagine the recovering the cost of both the foam and the cosmetic repairs in any timely way, like under a decade!

[Dana]

You'll be happier camper after fixing the flow issues then, unless you had the high-limit on the boiler set pretty low, like 180F, which wouldn't give you enough headroom for a 30F delta-T when the tank set to 150F (the return water from the heat exchanger would be well over 150F when the tank is approaching it's setpoint.)

A decent foam-injection company can do it with 1/2" holes one (or at most two) per bay, which is pretty easy to patch. Sometimes it can be injected behind chair-rail or crown moldings, reducing the amount of cosmetic work. They can also drill in from the outside through the mortar, but that's more labor/money.

It would almost certainly result in fuel savings well into double-digit percentages, and could be as high as 35%. But again, the "payback" is primarily in comfort, not necessarily a net-present-value-on-fuel savings sort of thing, even if a decade long NPV analysis it might still work. The financial analysis depends on what natural gas prices ACTUALLY do, but it's clear they simply can't drop much below their current historical low. But the "how cold does it feel" analysis has pretty immediate payback: Even in a 70F room a cool 50F wall surface yields an average radiant-temperature in the room low enough that to exposed skin (like your face) it never feels very warm. Raising the coldest surface to 60F makes an immediate noticeable difference in coziness at any given room temperature.

Unless you're in love with the bare brick look, you can bring that exposed 13' wall up to R10 with 1.5" of rigid foil-faced polyiso and half-inch blueboard or metal-lath, over which you could do a textured plaster if you're shooting to match the look of the horsehair, giving up only 2- 2.5" of interior space. (R10 is the whole-wall R value of a 2x4 studwall 16" o.c. with R13 batts or open-cell foam, comfort level= pretty good)