GPM in a hydronic slab system

Al Letellier

This is as critical a setting as having the right amount of tubing and proper temp, etc. They all work together to give you maximum performance and efficiency. The GPM setters/meter are used to set the flow to each loop as they are rarely the same.
the more I do radiant the more forgiving I find it, but you still need proper design perameters to get it right. Have someone (contractor or supplier) enter your information in any manufacturers design software to get proper flow settings. You need this information also to properly size your circulators, etc.
This kind of system is a big investment and well worth doing right the first time.

<A HREF="http://www.heatinghelp.com/getListed.cfm?id=248&Step=30">To Learn More About This Contractor, Click Here to Visit Their Ad in "Find A Contractor"</A>

johnnyd

GPM for a hydronic slab system

Newcomer here, but expect to be a regular.

Plumber by trade 25 years ago, now a DIY, HO, GC doing a 1200 sq ft addition to 480 sq ft cabin. Plan to heat with staple down PEX/plate/OSB sandwich, have framed accordingly for doors etc. Also have a nearly completed garage/workshop/guest bed-bath with a just-installed heated slab. Location is SE MN with -30F typically the coldest day, BUT it's 58 degrees F today.

Details of question have to do with the heated slab:

question about the GPM that I should be pumping at. This is a single zone, 3 - 250' wirsbo circuits stapled to the 2" rigid under the 24 X 29 foot 4" thick slab. 4" rigid perimeter insulation. Has a Rehau (?) manifold and a Grunfos "medium head" three speed pump controlled via slab sensor and SP 30 Goldline unit. Heat source is a 50 gallon A.O. Smith propane WH.

Initial setup has the water temp going in at ~100 degrees and coming out at ~ 80 degrees, running at 14 psi. Hardly needs to go on at all to maintain a slab temp of 55 degrees. Actually, the 10 -150 watt bulbs that provide overhead light warm the air temp up 4- 6 degrees in a 'couple of hours with outside temp at 20 degrees. R40 ceiling, R23 walls, no windows, two insulated garage doors, one man door. Purpose of the system is just to maintain a comfortable baseline temperature to work in, with a side benefit of helping to keep the well pressure tank, BIRM filter, etc, located in an adjacent room, from freezing.

The manifold has little red knobs on the return side that screw in and out, controlling little pins, which in turn affect the GPM readings on the little floating meters right above the valves. When wide open, the meters read almost 2 GPM. What should I run them at? Are they intended to provide the right flow so you can maximize the amount of heat passed to the slab, or to control the flow to individual circuits? Or both?

Should I run them wide open? Does GPM in a system like this really matter very much?

Mike T., Swampeast MO

2 gpm through a 250' loop is an extremely high rate of flow--even for ¾" tube.

The 20° temperature drop (100° in, 80° out) is what is called the delta t or change in temperature.

Each gallon per minute of flow at a temperature drop of 20° has liberated about 10,000 btu/hr. At your total flow of about 6 gpm, this would imply that your system is delivering about 60,000 btu/hr at a period of LOW need. No wonder it "Hardly needs to go on at all to maintain a slab temp of 55 degrees."

While you haven't provided all the variables, it sounds to me as if your circulating pump is greatly oversized. Likely you need no more than ¾ gpm or so through each loop--possibly much less.

I don't know if there is any terrible harm in operating it at a flow rate much greater than necessary, but it is certainly costing more than it should in electricity to circulate the water.

Could you reply with:

1) Your btu/hr/sq.ft. output at design (or heat loss and square footage).

2) The size of the tubing used in the loops.

3) The on-center spacing of the tubes in the loops.

4) The desired air temperature at outdoor design temp.

"Are they intended to provide the right flow so you can maximize the amount of heat passed to the slab, or to control the flow to individual circuits? Or both?"

"Or both" is probably the correct answer but such is more a "tweaking" than the radical adjustment in flow that you would seem to need. If you adjust them down--greatly--your circulator will still be oversized and working hard. You have just changed the reason for the head loss from "high velocity" to "high restriction".

johnnyd

Glad I asked. I'll investigate this a little more tonight.

Meanwhile:

Your btu/hr/sq.ft. output at design (or heat loss and square footage). 27 btu/hr/sq.ft. 20,000 btu/hr/total room

2) The size of the tubing used in the loops. 1/2"

3) The on-center spacing of the tubes in the loops. 16"

4) The desired air temperature at outdoor design temp. Typically 50 degrees F, but may ramp this up to 65 for long weekends.

Maybe that's why they supplied me with a three speed pump. It's running at the highest speed now...should I experiment with the lower speeds?

I don't quite understand the relationship of flow and btu transfer. Is there a direct relationship between higher flow and higher transfer?

Mike T., Swampeast MO

Is there a direct relationship between higher flow and higher transfer?

Yes, but it is not linear and only if such change in flow decreases the delta-t of the system. A decreased delta-t actually indicates higher output as the AVERAGE temperature has increased. However, you can only "play" with delta-t this way in quite a narrow velocity range for any given size of pipe.

I presume you mean 700 square feet in the room with a loss of 20,000 btu/hr and NOT 700 btu/hr/sq.ft. Such an output would be sufficient for cooking!

The 700 square feet at 20,000 btu/hr loss equates to about 29 btu/hr/sqft--a VERY reasonable number for the construction you described.

My handy-dandy Rad Pad shows you need right at 1 gpm in each of those 250' loops. The pressure drop through typical ½" PEX at 1 GPM over 250' is about 13'.

This means you need to select the speed on your pump that delivers closest to 3 GPM at about 13' of head loss. There will be a simple table supplied with the circulator that contains both of these variables at the various pump speeds.

If even the slowest speed will deliver significantly more flow at the head loss you really should get a smaller circulator. There should be a simple table supplied with the circulator.

Since your tube runs are identical (or at least nearly identical) in length, you need only use your balancing manifold to ensure that the flow through each is as close to identical as possible. Again, turning them all down by a similar amount only changes the location of the head loss, not the amount.

johnnyd

Thanks, I printed out your reply for reference tonite.

I find this whole concept fascinating. Don't have to wear wool sock anymore.

Mike T., Swampeast MO

You're welcome

I triple-checked the numbers I gave--based of course on the numbers you supplied.

Just wish that I could better explain the relationship between flow, delta-t and output other than to say, "stay within generally established parameters".

For your possible enlightenment however, from the Buderus Design Manual for Panel Radiators:

Heat loss (Q), water flow rate (GPM) and temperature drop (delta-t) through a hydronic heating system are related to each other as:

Q = 500 * GPM * delta-t

Said of this formula, "This equation is used extensively for accurate sizing of radiators."

Note that the equivalance in the equation is "heat loss", NOT "heat output".

Mike T., Swampeast MO

You're welcome

I triple-checked the numbers I gave--based of course on the numbers you supplied.

Just wish that I could better explain the relationship between flow, delta-t and output other than to say, "stay within generally established parameters".

For your possible enlightenment however, from the Buderus Design Manual for Panel Radiators:

Heat loss (Q), water flow rate (GPM) and temperature drop (delta-t) through a hydronic heating system are related to each other as:

Q = 500 * GPM * delta-t

Said of this formula, "This equation is used extensively for accurate sizing of radiators."

Note that the equivalence in the equation is "heat loss", NOT "heat output".

johnnyd

Thanks again. Last night I double checked the readings in the flow meters. I was in error reporting 2GPM, that IS the highest reading on the scale, but all three circuits are actually running a bit less than 1 GPM, which would be in line with your numbers.

Still can't get out of the habit of wearing wool socks, though..my feet were roasting, even though the slab feels cool to the touch. Amazing.

Einsiedler

quick cheat

total btu transfer divided by 10,000 = GPM

or

1 gpm per loop

EIN