More efficient control for slightly archaic multifamily hydronic system

keep the heat in!

Tighting the envelope has the highest return on investment.
properly sized and set up boiler is next.

Seems like with 20 units there will be demand nearly all of the time in at least one unit. Wouldn't the people that are there during the periods you propose to turn off the boiler be very unhappy during those periods? Beyond making sure the piping and building are insulated and air sealed I don't see where there is a lot you can do, the heating system is just producing what the building is losing.

Unfortunately, it does not seem feasible to wire all the zone valves to connect them to a boiler controller - too much distance, too many walls and floors.

Possible to do this wirelessly?

most of the heat loss will be in the units on the outside. units in the middle aren't a good indication of what the heating demand is.

a buffer tank could increase the cycle time of the boiler. constantly heating the system then letting it cool isn't going to change your standby losses much.

" the circulator (B&G e615S) runs continuously. I don't think this circulator would be happy getting turned on and off often - seems better to leave it running and stick to controlling the boiler. "

Seem like a waste to me having a circulator running with all the zone valves closed. May not be good for the circulator, although it has lasted this long. Does it run all Summer too ?

You certainly can just monitor the flow and control the Tekmar 256 Demand input.

Curious, does each unit have its own 24 VAC transformer for the thermostat and zone valves.

I'm assuming there is more than one circ here. Typically this would be done with primary secondary piping to so the flow rate in the primary loop will not change with zones cycling.

Usually local ordonnances dictates that you must provide heat up until a certain date, so shutting down the boiler is not really an option.

The usage is quite low, on average that works out to about 9000btu/unit. Not going to get much better than that. There is not much you can do improving the existing setup. Provided those pipes and boiler are in conditioned space, those standby losses go to heat the building anyways, so shutting down the boiler won't save much fuel.

If you want better, when the boiler comes up to replacement, you can look at a condensing unit which will allow you to run a much more aggressive outdoor reset curve so you won't need to circulate 140F when it is warm outside.

With the 5 Differential Pressure Bypass Valves (item "K" on the drawing) seeing flow changes in the boiler room may be difficult.

We have a 21-Unit multi-family building with a hydronic heating system built in 1997. We have high-mass PEX in Gypcrete in-floor heating so Unit temperatures are very stable. We installed a Taco SR-506 near the Boiler and hard-wired it to the thermostats in 6 Units. Selected the 6 highest Heat Demand Units due to orientation or residency. The Taco SR-506 provides the Heat Demand to the Mod-Con Boiler and one ECM Pump when any one Unit calls for heat. Do Not need all 21 Units providing the Heat Demand to the Boiler - that is redundant. The rest of the Units heating fall in sync with the weather and the 6 highest Heat Demand Units. Works like a charm - been doing it this way for 20 years - no problems. Very Simple - Meets everyone's needs - Big Cost Savings. Doug

A couple of thoughts:

First, what was considered well-insulated in 2000 would be considered marginal in new construction today.

Notwithstanding the foregoing, as the lawyers say, I would assume that a 2000 building has an insulated basement. So all the heat you're producing is staying within the insulated building envelope and it isn't really wasted, it's heating the building. Unless parts of the building are uncomfortably warm there's really nothing to fix.

I agree with @Kaos that the overall level of usage isn't that high. The reason it seems expensive is that your gas is expensive.

That main circ is probably about 1.5kW running 24/7. I think replacing that with an ECM unit is an easy cost save that is not intrusive.

The boiler is definitely short cycling, a buffer tank would help just not sure if it would save enough money to justify the work.

I really like @Doug_7 solution. Simple and effective.

And, what model are the zone valves?

This

That main circ is probably about 1.5kW running 24/7. I think replacing that with an ECM unit is an easy cost save that is not intrusive.

You could put a flow switch on each heating loop branch after the differential bypass valve (marked K on the drawing). The flow switch would only now activate if there is a zone on that branch is calling for heat. This would get you a demand signal.

The other option is to use the return water temp as a demand signal. You can generate this by puting an aquastat and delay timer on the return water. The idea here is to set the aquastat a bit above the temperature you expect when there are zones calling for heat. If the zones shut off, the return temperature rises triggering the aquastat. This now starts a timer that keeps the boiler off for say 2h. This is pretty simple to install and set up and should see pretty quick return.

The nice part about this is that it will force the zones to synchronize as more zones will eventually be calling during the off period but since there is no heat, they will stay on together. This extra load at the same time will also reduce the short cycling during the shoulder season. Thinking about this, you might want to have a programmed off period even in the winter just to do this.

If your emitters are also as oversized as the boiler you can also adjust the reset curve. There is no reason to make 180F water when cold if 150F will heat the place.

Just to sanity-check:

I'm going to assume a townhouse-style unit is roughly 1500 square feet, so 20 of them is 30,000 square feet. Let's say a 2000-vintage building with a 40F design temperature has a design load of 10 BTU/hr/sf, that's 300,000 BTU/hr as the design load. At 40F you're 25F below 65F, so that's 12,000 BTU per degree-hour or 288,000 BTU per degree-day.

According to degreedays.net, San Francisco had 390 cooling degree-days in January, which would require 112 million BTU's. A therm is 100,000 BTU's, so that would be 1120 therms.

Actual observed usage is 1050 therms.

So just at first blush that doesn't seem that bad. Obviously all of the assumptions can be questioned and refined, but the energy usage isn't way outside of what would be expected.

I'll throw in that if you have heat pump water heaters, in the winter the heat for them is coming from the heating system. I would expect each unit to use about 20,000 Btu per day for hot water (50 gallons per unit) which comes out to 12 million BTU for the whole building per month.

Setting it up primary secondary and setting up odr on the building loop will likely solve a lot of your problems, it will reduce the standby losses substantially and it will mean the units get decent run times instead of very short infrequent cycles and it will stop the whole building form overshooting possibly just from the standby losses.

In this climate replacing the boiler with an air to water hp might be feasible in the long run.