ΔT pumping with radiant floor

ratio

Are there advantages to ΔT pumping on the circulator for a large (≈60' x 80') radiant floor in a church? Installed c. 1950, 2½" piping going into the floor, B&G 2½" PD pump with gate valves to manually balance the flow through a 2" bypass to control temp. Currently the slab pump is controlled via a Honeywell 8000 (set up as non-pregrammable) with a remote sensor drilled up into the slab as close to the surface as I felt comfortable. I don't know how the floor was controlled previously, although there is an abandoned aquastat in the return just prior to the pump. I'm having trouble with overshoot & keeping the supply temp comfortably low—I can't afford to have any issues with the old piping!

What I'm thinking of is a ΔT pump on the circ, deleting the bypass & moving to injection pumping (which should allow me to meter the actual BTUs going into the floor). IDK if there is a 2½ ΔT pump, but I can get a 3Φ motor for the B&G & run it off a VFD for speed control—all but the cheapest drives have the capacity for that on board.

The question remains: is this "Meh, lots of work for what?", "Why would anyone do that?", "Yah, sounds like a good idea!", or something else entirely? @Hatterasguy, @hot rod, I'm looking at you guys for lots to think about!

hot_rod

Trying to increase comfort? Or reduce fuel cost?

It's tough to control a high mass radiant in a space where the occupancy load changes so quickly. Outdoor and indoor reset is one way to control them accurately.

I'm not sure how a stat with weather prediction, like Ecobee would work with that control logic?
It has always been temperature overshoot that is hard to get a handle on, when the space goes from empty to capacity with a bunch of 400 btu/hr. radiators sitting in the pews

ratio

Boiler is a Peerless 211A, 5 sections. SWT control is manual gate valves & a thermowell with a now-operating thermometer inserted. Heatloss is not yet calculated. It's going to be a bear, 60' 14/12 ceilings & odd shapes all over the place. There is also a circuit of fintube on a stat and a circuit of hydro coils that were previously controlled to DAT via series 90 controls but now are just on/off via stat.

The desired outcomes are thus:

1. protect the system, i.e. old radiant piping buried in 60 y.o. concrete & relatively–new boiler
2. avoid complaints
3. allow the system to be reset via OAT, sort of
4. coordinate radiant operation with air handler operation

The last two need some clarification I believe. I think that if I increase the gpm injected proportionally to the OAT (or actually a slowly evolving mean temperature), that will simulate outdoor reset. Or at least, as the temp is dropping the slab will be getting warmer and vice versa. And by coordination I mean that the discharge air control will eventually be repaired & tuned to make the airflow comfortable, not cool or drafty feeling, during occupied times.

hot_rod

It sounds like a challenging, but interesting project.

60 years old steel pipe in a slab would make me want to include something in the contract about expected life span? I'd guess little, if any under slab insulation.

The main thing with iron pipe radiant is assuring the pipe is dry. Once a leak, or ground water gets in contact with the tube, they can corrode outside/ in. Site drainage around the building is one thing to watch, make sure all rain runs away, not under the building.
If the installers were careful and all the pipe is encased and surrounded by concrete, I suppose it could last another 60 years.

It is a tough call on how much flushing and cleaning you want to do with the old steel pipe. I would certainly add a good magnetic dirt separator.

Avoid complaints? What are the complaints, from whom? I doubt you will ever make the entire congregation comfortable

It sounds like you need a multiple temperature system, ideally.

Many of the new boilers, mod cons especially have the ability to run multiple ODR functions right from the boiler control. You should be able to have both high and low temperature distribution reset. That high mass radiant flywheel may have a mind of its own.

If you are going with a cast iron boiler, injection mixing is a very workable logic, ODR and return temperature protection comes along with that control arrangement. A mini tube injection is what I used on a few large commercial radiant slab. A couple small, high temperature lines injecting into that 2-1/2 main radiant loop.

I don't envy the load calc math, but it may be the key to reducing energy cost.

Zman

I would think that the best strategy would be to control the floor with outdoor reset and indoor feedback. You could set it up to error on the low side to avoid overshoot.

You could then use the air system to bump the temp if needed and provide some air exchange to offset the btu's of the occupants.

The worse condition you can have with a system like this is a slab that has been heated on a cool Saturday night with a fixed temp system. If Sunday morning is warm and 240,000 btu's of church members show up, you are garanteed to have complaints.

Zman

Hatterasguy said:

Zman said:

The worse condition you can have with a system like this is a slab that has been heated on a cool Saturday night with a fixed temp system. If Sunday morning is warm and 240,000 btu's of church members show up, you are garanteed to have complaints.

I see a boiler with 672K output.

I assume that it is not 2X the size necessary (because I don't have a heatloss).

So, let's assume the heatloss is 400K at design.

If you delivered 240K to this space, my guess is that it would take at least one hour to climb two degrees at design. Of course, at warmer ambients, the climb rate would be a bit faster but the presence of 240K is probably not onerous for a space that size if the congregation is only present for two hours.

The problem with overheating is due to the floor temperature, not the congregation, IMHO.

I suspect the problem is caused by both the slabs and the occupants. If the slab is running a fixed temp, it is the greater contributer for certain.

SWEI

Is there mixing now?

ratio

Ok. The overshoot I mentioned is the slab itself, it'll overshoot it's setpoint by two or three degrees. Control is via a Honeywell 8110 set to non-programmable operation. Mixing is a manual process of balancing gate valves, & it seems that I have to really crank the return valve nearly closed to force most of the water through the bypass. I'm concerned that the 2½" gate valve might not be up to the task of throttling the flow from a ¾ HP pump.

Oh, and this is my parish. No contract, just responsibility.

I just know that I can get this system working better than it is & it bugs the @?&?#/&@ outta me that it's so crummy now, both from not meeting current best practices & lack of maintenance.

bob_46

You don't mention if the air handlers have outdoor air or not. What were the series 90 controls controlling and how. Water side air side ? It sounds like this may have been an engineered system that has been knukleheaded over the years.

ratio

Air handlers are capable of 100% outside air, I added a Honeywell economizer control with demand ventilation last summer, so at least I have free cooling in the heating season. Don't yet have. The series 90 maintained discharge air temp via a 3 way valve. One unit had the valve replaced with straight pipe, so now I just switch the pump via a thermostat. I'm planning on getting the valve replaced & getting back to discharge air control, but but this season.

I keep fiddling with the valves on the radiant loop from time to time, but with so many variables to account for I'm just guessing. Since this isn't a paying job, I really don't have the time to spend a day or three watching what is happening.

hot_rod

I don't think you will ever get a handle on comfort, efficiency or control-ability with just a manual balance or two.

It sounds like it needs a complete $$ update, starting with the gate/ balance valve

It is nice of you to offer your services. At some point to correct all this, or at least formulate a plan should be a paying gig.

ratio

Hence the plan to switch to injection pumping. It does need a good repipe, starting with moving the pumps to pumping away. I've got an idea about a homemade hydraulic separator made from 8-12" grooved pipe & weld-o-lets, but I don't know how that will compare to say a Sep4™

We're a pretty poor parish. Whatever I can't afford the church will buy, but I try to give them what I can.

hot_rod

Not much to building a basic separator. About a 3-1 ration pipe size to barrel size.
Ideally you want it to do multiple tasks. Adding a media turns it into air, dirt and hydraulic separation.

We are adding magnets to all the sizes little by little. Weld a port or two and I can get a brass dry well for you too load magnets in.

ratio

I was thinking about something like this as a coalescing media, although maybe PTFE is out of my price range.

hot_rod

On the larger welded steel versions we use an expanded stainless mesh. Tough to weld a vessel with plastic inside.

Depending on your flow rates, the media in those higher flow vessels needs to be pretty robust. Early testing showed a simple roll could collapse or distort at high flow rates.

SWEI

I think I'd probably start by installing a 4-way mixing valve on the slab circuit along with some kind of dirt separator, hopefully magnetic. Pick the right valve and you can motorize/automate it later for ODR control.

Gordy

I see injection mixing with ODR, and indoor feed back for the RF.

Are you sure the floor piping is iron, or copper?

ratio

No visible copper, even the water service is threaded pipe, cornerstone is dated 1954. I don't know the layout under the concrete either. I believe it to be water tight too.