HTP UFT Delta P Circulator w Constant Circulation

Brirob

Hi All - About to replace an older p/s piped watertube boiler (Burnham X-C series) with a direct piped UFT 80w and have a few questions. For a little background on my current system and why I'm moving to the UFT: I installed the current system about 5 years ago and it works ok but leaves a lot to be desired. Heat loss for the house is around ~57k btu - turn of century house in the lower Hudson Valley NY (Rockland Co) lots of windows, no insulation in the walls, 15* design day temp. Current boiler is 94K btu with a 5:1 turndown - modulates down to 20K btu - so it's quite oversized and short cycles more than I would like. It also requires a high head circ (taco 0013) for the primary loop which draws a lot of power and makes a lot of noise. Delta T for the primary loop is consistently ~5*, even at high fire, while my system loop delta t is almost a perfect 20* as designed. Manufacturer (Burnham) didn't publish any pressure drop figures for the heat exchanger. They just spec the max effective length for the primary loop and tell you to use the 0013 or the Grundfos equivalent if you're under that max. Design day supply temp is 160* so I should be in condensing range most of the heating season but the primary loop is killing efficiency. Image of current system is below:



Current emitters are fin tube (two zones, around 100' total) with a couple of panel rads (on monoflo tees) where I couldn't fit enough fin tube. New system will be 3 under floor radiant loops of around 270' each for the first floor - extruded plates, insulated, etc - and a few panel rads as second stage. Upper two floors will all be panel rads. There will be 12 panel rads total on 8 circuits. 6 rads will be home runs, one circuit will have 3 rads in series with "H-Valve" bypass valves, and one circuit will have 2 rads in series with H-Valves. All panel rads will have TRVs. Floor loops will have actuators on the manifold. Rads are sized for 150* design day supply temp with a 20* delta T. Wish I could have gotten that down to 140* but there are lots of windows that go down to about a foot above the floor so wall space is limited.

I'm planning to pipe the UFT direct, ODR, constant circulation with a Viridian 1816, 12 port Caleffi manifold. One design day supply temp (150 *) for both the floor loops and the rads. I have a t'stat with a floor temp sensor to make sure I'm not getting the floor surface over 85ish *. Trying to keep it fairly simple and not run a mixing valve, additional circ, manifold, etc for the floor loops. Ran the numbers and the 1816 will supply all 11 circuits.

My question is this: With constant circulation (tt shorted) on a direct pipe system how do I ensure the min flow through the boiler (1.3 GPM per HTP manual) is maintained if all TRVs close off most of the way and the floor loop actuators are closed? My thought, and this may defeat the purpose of the viridian circ in delta p mode, would be to add a press differential bypass valve set to ~ 2 GPM as a safety. Or perhaps the UFT control logic shuts down the burner if there is little/no demand? Also, being that two of the rad circuits are using bypass valves, there will be some flow through the system either way - not sure how to figure out how much. Is there any point in using a viridian circ in constant pressure mode in this scenario? This config (constant circ, direct pipe, TRVs/actuators) is all quite new to me so can't quite grasp conceptually how it should work. A (very) rough sketch of the proposed system is below for reference...need to work on my visio skills. Thanks in advance for any guidance.

ewang

Pipe as Primary/Secondary either using closely spaced tees or using a hydraulic separator. That will ensure the flow across the boiler is not influenced by activities in your system loop.

hot_rod

Good question on how the ∆P circ would behave with the pressure bypass.

I'm a fan of a small buffer tank, it handles the micro loads, gives you separation, and also assures cold RWT. In your case it also assures adequate boiler return.

Basically it increases the water content of the boiler while adding air, dirt, hydraulic separation and some buffering to eliminate short cycles.

Ironman

I would agree about using hydraulic separation with a buffer tank. Can your panel rad's operate at the same SWT as the floor and give sufficient output? If not, then you're gonna have to create 2 SWT zones, one higher one for the panel rad's, and one lower for the floor.

Brirob

Thanks for the replies. I really do like the idea of hydraulic separation using a small buffer tank for some extra mass. I've always piped my modcon installs p/s with close tees and that works well enough but I always end up with short cycling in the shoulder seasons and higher than desired RWT. The buffer tank should help with those issues. Wondering what size buffer would be best? I was thinking a 20 gallon (or perhaps smaller) electric tank would work here if I can sort out the tappings.

The panel rads are sized for 150* SWT. Wanted to get it lower but couldn't find the wall space. I'm planning to use one SWT for both the rads and floor. Single manifold/circ. Floors have a fairly high R value. Original 4/4 heart pine over 4/4 t&g douglas fir. Heat load for the first floor is 22K. I can get tubing/plates under about 600sf of the floor. I ran the numbers and I should conservatively be able to get about 25 btu/sf. That would leave a 12K btu deficit for that space. Rads will make up the difference.

hot_rod

I have used as small as 6 gallon electric HW tanks for buffers. Many of them have two side taps in addition to top connections, so you have plenty of access points. 12 or 20 gallon are them next two common sized, but tend to be $$.
You can guestimate or use a calculator to predict run and coast times. I think Lochinvar and HTP- Boiler Buddy have online buffer tank sizers.

With a modulating boiler the tank can be much smaller compared to a single speed boiler.

Brirob

Did the math and decided to go with a 6 gallon electric water heater for a buffer tank. That would give me 10+ min run time on the warmest day of the heating season/coolest SWT with only one TRV on the smallest rad partially open. Going with a two-pipe design per idronics 17.

Next question is does it make sense to add a sensor/control to a tank this small or just leave it sensorless extra mass? If I add a sensor, I assume it could be controlled by a Taco SR501 relay, Tekmar 256, or the like. Tank ODR would be set to the same ODR curve as the boiler. Using a Taco PC700-2 with an SR EXP to control the circs is another option but adds quite a bit to the cost and complexity. What's the best option with a small buffer tank?

mattmia2

You have to think about the cost of the extra complexity vs the potential savings. My guess is that the savings will be so small it doesn't make sense from an efficiency standpoint. In some cases it could make sense from a comfort standpoint. The boiler has an optional remote system temp sensor that you can use to control it with the built in boiler control.

hot_rod

Youngplumber said:

OK question @hotrod maybe you've thought of this or heard some interesting info on it. How does the sacrificial Rod inside of a water heater help or hurt a hydronic system? I'm inclined to think help, but I'm not sure full side effects of a magnesium  rod inside of a closed hydronic sytem vs closed freshwater system. Thanks

as long as there is no glycol in the system it should be ok

hot_rod

Brirob said:

Did the math and decided to go with a 6 gallon electric water heater for a buffer tank. That would give me 10+ min run time on the warmest day of the heating season/coolest SWT with only one TRV on the smallest rad partially open. Going with a two-pipe design per idronics 17.

Next question is does it make sense to add a sensor/control to a tank this small or just leave it sensorless extra mass? If I add a sensor, I assume it could be controlled by a Taco SR501 relay, Tekmar 256, or the like. Tank ODR would be set to the same ODR curve as the boiler. Using a Taco PC700-2 with an SR EXP to control the circs is another option but adds quite a bit to the cost and complexity. What's the best option with a small buffer tank?

Use the extra sensor that either comes with the boiler or is available, usually labeled system sensor on the control board.