I have a 6 year old house with a hydronic radiant floor. When we first built the house the radiant floor heated the home quite nicely. Now it cannot get the house hotter than 16-17 degrees C even when the radiant floor pumps run full time. We have made some plumbing changes over the 6 years which may be the cause. I have consulted with a number of local plumbers none of whom have been able to identify the cause. I'm looking for an engineer or architect who can help me. First to do a heat load calculation to establish what the system "should" be doing. And then to help me triage possible causes for why the house won't get warm. Any referrals or recommendations would be very appreciated. Thanks

This is more of a comment about radiant heating. I lived in an older home that used radiant heating and it ALWAYS needed constant upkeep and broke down every year.  I've been reading A LOT of threads about broke and problematic radiant heating. While technologies have advanced, it seems that this is a problematic heating method. Right behind geothermal, which always seems to be having problems.

Hopefully someone can help you out with your issue...

Where are you located and what kind of insulation do you have in the walls/ceiling?

what were the plumbing changes you made? If it worked fine at first and doesn't now, unless you changed your heat load it's unlikely that the cause is heat load related.

as for Lbear, there have been a lot of MFGs with a lot of materials and some runs of some items have had problems over the years. those people may talk about their experiences. but most people don't need yearly upkeep.

Location?

bluelagon, you will need to get someone knowledgeable about hydronic heating to come out and troubleshoot your problem. An architect would be my last choice. A mechanical engineer who is a member of American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE) would be my first choice. However, you will likely have to settle for a HVAC technician. If you are lucky enough to find a good one, you may be well served. Please feel free to use the heat loss analysis calculator on our website:

Borst Heat Loss Analysis Calculator

Normally one does a building heat loss analysis BEFORE designing and installing the hydronic heating system. You are likely suffering from having some clueless technician work on your system. Do you have supply/return temp gauges on manifolds and, if so, what are they reading when system is operating?

lbear, a well designed hydronic heating system is likely the most reliable heating system there is...except for passive solar. Yes, an old home that used steel or copper pipe for the circuit runs will have problems. Properly installed PEX will outlive several owners of the building. The real problem now-a-days is that there are many incompetent people in the HVAC business. As I wrote in another thread, one does not even need a HS education or a HVAC license to become HVAC technician in many states:

HVAC License Requirements by State

Hopefully this will change if enough people complain about the clowns that cause these problems. We advise applying due diligence when it comes to hiring someone to design and install a hydronic heating system. We don't recommend using a design unless a licensed professional engineer seals and signs their name on the design.

a licensed PE for MOST residential systems would be cost prohibitive and pretty unnecessary. And from my experience, most trades guys will out design most PE's out there for residential sized hydronics. the stuff out there with PE stamps on it outside of the commercial realm is almost always wildly oversized and overdesigned.

there are no shortcuts to avoid the need for references and a proven track record if surety is the goal.

I would say a professional engineer has both the knowledge and the legal responsibility to properly design a hydronic heating system. Few architects, builders, or technicians really understand the engineering principals and construction techniques required to achieve a successful result. Most technicians are incapable of determining the hydraulic friction of a hydronic heating system. However, this must first be done to ensure all the circuit fluid flow rates and associated velocities are within required limits to ensure air entrainment while also minimizing circuit noise/wear, and to then properly select a pump suitable to this head/flow rate performance curve to ensure maximum system efficiency and long pump life.

This may well be what the poster of this thread is dealing with…a bad design that has now resulted in premature pump failure such that there is now insufficient pump head to overcome the hydraulic friction to achieve the necessary flow rate to generate the required BTU heat gain.

We have seen exorbitant fees charged for so called hydronic heating designs and passive solar house plans that should have never see the light of day. We would expect any intelligent consumer to get at least three estimates and verify the credentials of anyone they are considering to hire. We post our engineering analysis /design fees and our credentials on our website. We also provide DIY calculators on our website free-of-charge so everyone who is willing to spend their own time can have a properly designed hydronic or passive solar heating systems. We did this NOT because professional engineers over-design heating systems or charge excessive fees. We did this because there are very few professional engineers compared to the large number of incompetent technicians available to properly serve consumers. We believe that everyone should have the opportunity to have an affordable, well designed, highly efficient, low maintenance, and long-lived hydronic heating and passive solar heating system.

Rob asked the right question,
We are all blind to what is installed but we have 2 clues
The system worked at one time. So someone got it right on some level.
There were changes to the plumbing. What could that have to do with an isolated hydronic loop?
My gut instinct tells me this is an open loop system, ( bad choice if so in my opinion)
If that's the case perhaps the plumbing changes mentioned are in effect here, return water diversion, another bath room shower or tub adding demand to the water heater, or simply a failed pump due to bad water.
More clues please?
Dan

sailawayrb, you can feel as you like. In my time I have seen countless stamped drawings I would not recommend to anyone for any reason. a PE stamp is not in any way, shape, or form a guarantee of success or even passable performance. there is no shortcut whatsoever for references and track records IF SURETY IS THE GOAL. Very few PE's specialize in residential grade hydronics which is a very different animal than commercial grade hydronics.

I've also worked with many, many DIY designs over the years. none have been optimal. Very, Very few have been good. Some have been ok, but spend enough more on energy or initial material that a good design would have paid for itself easily. quite a few have had significant failures or shortcomings that required rectification.

I also have worked with many DIY heat loads over the years. I programmed my own load calculator and have been doing heat load calculations for almost 15 years without failure yet (which means I am still oversizing, I suppose, even though we usually come in 30-50% lower than most other load calcs). And still, on a large percentage of projects I deal with there are significant factors that require experience and judgement calls, and there are many rookie errors that can occur in load calculation. identifying heavy framing areas, infiltration estimates and the effects on it, window U-factors and sizing, "mass enhanced" r-values, "r-value equivalents", I generally see at least one, usually multiple significant issues on every homeowner-driven load calc I see with very few exceptions.

to be fair, they aren't any worse than most heat load calculations I see out there from "professionals", but garbage in, garbage out, and in our industry even many professionals are doing poor quality work in that regard. at best the load calcs may be shrinking heat sources from 3x oversized to 2x oversized in most cases, and failing to deliver the room by room comfort balance that requires relative precision. It's probably a good reason why 70% of people... including radiant owners... in this country are unhappy with their heating systems.

Design makes a difference.

Showing off is just human nature as is casting aspersions...the latter being one of the lower examples of our nature.

I am happy with all my radiant floors...every one.

A charitable Merry Christmas to Engineers and Mechanics alike.

Sounds like we largely agree...design is very important and there are a lot clowns in the HVAC industry. This is why consumers should apply due diligence by always checking credentials/references and getting mutiple estimates. This is doubly true for those states (see above URL) that do NOT require licensing, bonding, or insurance such that any clown can claim they are a HVAC expert and the consumer has little recourse to seek legal remedy when there are problems.

Best wishes for a happy holiday season and new year to one and all!

So,
A lot of chatter, but who has a clue why a once working design is now not?
What could have changed? Mr. Blue Lagoon?
Just curious,
Best for the Holidays all,
Dan

Hopefully the OP was actually able to get someone out to resolve the problem and will eventually tell us the cause.

My bet is either premature pump failure from operating the pump(s) near or at max head for several years or the consequence of the plumbing changes. I am leaning toward the former since the OP indicated that the plumbng changes occured over the 6 year period that the system has been operating and the impact of each of these plumbing changes should have been immediate and likely noticed.

Hi,

Wow! So many responses to a brief request for help. Thank you so much.

Here are some details of my system and my history with it.

System
- Thermomax 80 gallon hot water tank (see http://www.solarthermal.com/wp-content/uploads/2010/12/sst80db.pdf)
- Hot water tank is indirectly heated, in the winter by a tankless hot water heater, in the summer by solar hot water with tankless hot water heater backup
- Radiant floor covers 3000 sq feet. The pipes are in 1.5" of concrete and I have bamboo floors (mostly) and some carpet on top of the concrete.
- The radiant floor is divided into two heating zones.
- The radiant floor pipes have glycol in them. The glycol runs through the floors and then into a closed loop that is inside the hot water tank and just keeps circulating from floors to tank and back.
- Originally the pumps for the radiant floor were Wilo Star pumps that could be set to 40W, 60W or 80W. They were replaced by Grundfos Alpha pumps which can be set to 15W, 33W or 45W. There are two pumps, one per zone.

Statistics
- With the pumps running at 15W each, there is about 1.5 - 2 gpm flow. With the pumps at 33W each, the flow rises to about 2 - 2.5 gpm. The temperature of the glycol coming out of the hot water tank does not seem to change whether the pumps are at 15W or 33W.
- The temperature of the glycol coming out of the hot water tank is 30 degrees C. The temperature of the return glycol is 21 degrees C.
- The hot water tank is set at 160 degrees F. The hot water tank does get up to temperature: I have three different readouts for temperature and recently the bottom one read 50C, the middle one read 65C and the top one read 68C. Just as I took the reading, the tankless hot water heater came on to heat the tank a further 6C before turning off.
- The radiant floor closed loop is at the top of the tank.

Note
One thing you should know is that we are on solar power. So every watt saved counts, especially in winter where all of our electricity is generator-driven by expensive fuel. So that was the drive to put in the Grundfos Alpha pumps. My goal is to run those pumps at 15W each. They run pretty much 24/7 (I think) so with the Wilo pumps running at 60-80W it was just far too much electricity.

I did not take a recording of how much flow in gpm the Wilo pumps put out versus the Grundfos Alphas but my impression is that the difference was very small.

History
When we first had the system going, it used the Wilo pumps. It could get the house incredibly hot - easily to 25 degrees C. We would actually be hot and have to turn the temperature down. Another interesting point is that when the house was first built, for the first three years, we had the hot water tank set at 120 degrees F - 40 degrees lower than it is set to now.

Nowadays, six years later, the system can barely get the house to 17 degrees C. It's very cold at home. I have found that if I run the pumps at 33W instead of 15W I can get the house up to 18 degrees C. In order to try to raise the house temperature, one plumber recommended we raise the temperature of the hot water tank to 160 degrees F. (We have a mixing valve between the hot water tank and the house hot water to prevent scalding.)

Possible Causes
I am pretty stumped at this point and so are several plumbers I have worked with. Here are a few ideas we have tested:

- There is a mixing valve right as the glycol comes out of the closed loop in the hot water tank. It is a Caleffi 521 valve. We tested if it works; it does.
- We tried replacing the Grundfos Alpha pumps with the old Wilo's. The house did get warmer but still not back to 25C. And, as mentioned, we simply cannot afford the Wilo pumps electricity-wise. I have to get the radiant floor working with the Alphas, preferably (strongly preferably) at 15W each.
- We tried flushing the glycol and replacing it.
- We do have heavy water. I have wondered if perhaps the loop inside the hot water tank is coated with minerals. The tank has not had any maintenance in 6 years. We are really not sure how to flush the minerals out of the tank if this is the case.


So....any thoughts?

Oh, also, I am located near Victoria, BC, Canada.

Bluelagoon, you indicated that your supply is 30 C (86 F) and your return is 21 C (69.8 F). The difference between the supply and return temps (allowed circuit temp drop) should normally be less than 8.3 C (15 F) for barefoot friendly residential hydronic heating. So you slightly exceed this recommendation...not a big problem...unless you have a spouse who complains they can feel the temp difference when they walk around...

Typically, the floor surface temp should be between 21 C (70 F) and 29.5 C (85 F) to maintain a room temp of 20 C (68 F). The actual required floor surface temp depends on the actual room/building heat loss.  If you have a IR handgun, perhaps you could determine your floor surface temps.  Anyhow, one would expect your return temp to be at least as high as this floor surface temp...and likely significantly higher. So your 21 C (69.8 F) return temp is too low and I would expect your floor surface temps to be too low as well.  So, for your new pump setup, your existing 30 C (86 F) supply temp is also too low.  However, if you were to pump at a higher flow rate (GPM), your existing supply temp may very well be entirely adequate.  So your problem may simply be that your new pump setup is perhaps not pumping at the higher flow rate (GPM) that your original pump setup did when these original pumps were new.  In this case, you will either need to increase your supply temp and/or increase your pumping flow rate (GPM) so as to increase your floor surface temps.  It is also possible that your supply temp was significantly higher 6 years ago and has since degraded because of scale from using hard water on the heat transfer pipes as you surmised.  A good installer will log all these original temps and flow rates (GPM), provide you a copy, and thereby make troubleshooting of these sort of problems much easier.

To maintain your desired room temp of 25 C (77 F), your return temp should be at least 25 C...and likely significantly higher.  Your supply temp should not be more than the 8.3 C (15 F) allowed circuit temp drop above your return temp.  Your pump setup will need to provide sufficient flow rate (GPM) to accomplish this.

Normally, for a given supply temp and allowed circuit temp drop, a good designer determines BOTH the required flow rate (GPM) to obtain the required floor surface temps (which depend on actual room/building heat loss) and the resulting PEX tubing system friction head (FT) at this operational flow rate (GPM).  Then the designer selects pumps that have a head/flow rate performance curve that are appropriate for this required flow rate (GPM) and this system friction head (FT).  This required flow rate (GPM) and system friction head (FT) point must be near the middle of pump performance curve for the selected pump in order to maximize pump efficiency and ensure long pump life. 

A good designer will provide the performance curve for the selected pumps showing the system operational point.  Sometimes a less-than-good installer will substitute a different pump (or just not be working from a good design) such that the installed pump gets operated on the high head side of it's performance curve.  A good installer will provide photos of the actual PEX tubing installation (useful if you ever develop a leak) and will log the actual lengths/diameters of all the PEX tubing circuits and leaders (originally specified by the designer and used to determine the PEX tubing system friction head).  If the both the designer and installer do their jobs well, the actual circuit flow rates (GPM) and actual differential pressure across the pumps (PSI) will be in very close agreement with the analysis/design.

Good luck and please let us know how you ultimately resolve the problem!

So, the radiant floors are on one tank loop. Is the other loop the solar panels? What is the make/model of the tankless water heater and how is it set up to circulate with the hot water tank?

Heavy water is technically deuterium oxide, should we understand that you have hard water, rather than heavy water in the DHW tank? From a well?

1. do you have any cold spots on the floor, like loops that aren't circulating perhaps? could have air problems in selected loops, lower your output.

2. is your water temperature to the radiant as high as it used to be? if it's dropped it could be a scale, or a mixing valve problem.

3. were the floor coverings a later addition? that would reduce your output radically.

4. note that the alpha "single line" setting has a max draw of eight watts. if you're happy with a higher loop delta-T (supply to return) you could try that setting to save energy. I design systems at 20 degree drop routinely without complaints and if you're an adventurous sort you could try even low flow/higher drops.. but lower flows require a higher supply temperature to compensate for the drop in average temperature across the loop.

NRT.Rob raises some very good points to consider/discuss.

I already discussed points 1) and 2) in detail in my previous responses so no need to repeat those again.

Point 3) could very well be your problem if your floor coverings and furniture were added later and were NOT included in the original analysis/design. This actually happens more frequently than it should.

Point 4) might make sense if you need to risk/trade floor barefoot friendliness for pumping watts. An allowed circuit temp drop of more than 15 F is often done for commercial buildings, but is not often done for residential buildings. If you elect to go with this approach with the resulting lower flow rate (and lower velocity) you should calculate your Reynolds number to ensure that it does not get less than 2300 to ensure your flow remains turbulent and does NOT become laminar. If your flow were to transition from turbulent to laminar flow, your heat output would decrease significantly. In fact, if your flow rate has decreased significantly from what it originally was, this could very well be your problem too. If you elect to go with this approach with the resulting lower flow rate (and lower velocity) you should also confirm your resulting velocity is between 2 and 4 FPS. As I previously discussed, if your velocity becomes less than 2 FPS, air entrainment is not ensured and you will be at increased risk of having air problems. Do you know how many circuits you have in each of your 2 zones and the diameter of PEX used?

Again, I would feel much better if you had all this analysis/design data sealed/signed by a PE and you also had the installation data/log documented/provided by a licensed contractor (or accomplished this yourself, DIY).  It can be difficult to reverse engineer and troubleshoot/remedy these sort of problems without having credible data.