Have we forgotten to make heat traps? (2012)
esbe.eu
esbe.eu
Overall, the practice seems to be very heuristic based, with every installation being a bit different and rules of thumb being the norm. Our system seems to work fairly well in practice, but I've been curious to get a professional's opinion on it (the only pro that's seen it was the gas guy who did the final hookup, who seemed impressed but didn't really offer any specific criticisms).
Repo at https://github.com/mtrudel/boiler
I've since added an elixir based graphing system to it, at
Our initial experience with the system was even worse than it might have been. The people who installed it including the thermostats were plumbers and did not realize there was a switch inside the thermostat to choose among three types of heating systems. They left all the thermostats at the default forced air heating setting. We suffered for two years with terrible overshooting and undershooting of temperatures before my own research revealed that the hidden setting existed and that it was set incorrectly. This helped but it was still not great.
We were glad to get back to a forced air heated house. I do believe however that hydronic heating with radiators is somewhat better than underfloor hydronic.
I know the nest thermostat does a lot of systemic learning and understands the outside temperature. (I'm not recommending that system, just that a little modern control theory might make things workable)
There is a setting on the thermostat (True Radiant) that takes into account the long heat-up times of radiant systems. With this setting off, overshoot is real. On cold days, True Radiant does start the system earlier. For example, it take about 2-3 hours in the deep of winter vs 30-60 minutes now.
However, it seems to be based to on a look-up table rather than an algorithm. It doesn’t take into account forecasted conditions, such as a temperature rise overnight, strong winds or decreasing/increasing radiance of the sun over the season. Those latter factors do make a difference in whether the house reaches the target temperature on time or whether overshoot occurs, and whether I am slightly uncomfortable.
As far as I can tell, there is an automatic set-back for heat pump systems. I wish this was given as an option for manual scheduling in True Radiant mode. To this day, I can’t tell whether there is an effect on efficiency (based on run-times which the thermostat dutifully displays).
I also wish the thermostats (in a dual-zone system with a single boiler) be “genlocked” or run on common firing times with demand prediction for temperatures to make use of boiler run-time, rather than running the boiler just when there is a request for heat in one zone.
My underfloor water heating is powered by a wood pellet burner. The house is very well insulated so the burner is off May till late October. The burner has an outside temperature sensor and since it was initially set I haven't really had to mess with the settings beyond testing how to change it. The temperature in the house is pretty much constant 21C or 69F during the heating season. If I want to change it the response time is around 1-2 hours so although I don't use those settings one can imagine programming it for 5 degrees lower when you're at work and for it to bring up the temperature to the normal setting before you arrive.
As for the cost, I have a rather small 3 bedroom detached house (110 sq.meters or 1185 sq.feet), I live in a climate where we have -20deg.C winters (eastern Europe) on top of a rather windy hill, and I pay around 460 Eur per year for the fuel for the burner. I usually burn 2.5 tons of it. This is only to heat the house. Water heating for shower use etc uses electric on demand heating. Compare this to when I was living in a city in the north of UK, so much milder climate, in a 2 bedroom semi-detached house smaller than my current house with typical central heating (or what is called hydronic heating here) using gas fired burner. I would pay around 700-800GBP so almost 900EUR per year of heating so almost exactly double. I realise I'm comparing here two houses built 50 years from each other, but the UK climate is much milder and the older house is only semi detached. This should work in its favour. Still I spend half of what I used to.
1) Steam and other radiant systems don't deal with temperature setbacks well. They take FOREVER to warm up a house after a set back, so don't bother with a deep setback overnight, and if you are gonna turn down the heat because you are going to be gone for a while, expect your house to be cold for a couple of hours while it warms back up, or expect your heat to overshoot by 5-10 degrees.
2) Radiant heat seems to work well in the temperature range it is designed for, but outside of that range you get strong temperature gradients in rooms. Unless you have a fan in each room, expect the ceiling to be substantially warmer than the floor. I have 12 foot ceilings, and the ceiling can be 15 - 20 degrees warmer.
3) I personally prefer forced air with my allergies than radiant heat, because you can upgrade a forced-air system to use better filters, which basically gives you a filter for your entire house. Radiators are a pain in the ass to keep clean, and the air often seems to go stale over time. I have fairly bad indoor allergies and it just feels like over winter the house just fills with allergens.
4) On the same note, you can't get whole-house humidifiers with steam heat, nor can you easily add air conditioning.
5) Steam isn't THAT quiet. An improperly tuned system will bang (which shouldn't count against steam, because that's the fault of whoever is maintaining the system), one-pipe systems (which are common in Chicago) hiss when they are heating up, the valves sputter as they die, and when the system cools back down there is a loud inrush of air. Two pipe systems make noise as they heat up and joints flex, and the valves also make little clicking sounds.
Hot water can get bubbles in it and make wooshing sounds, but that's generally a sign of bad maintenance.
The sound of air blowing bothers me less in general than one off sounds that occur with steam and hot water though, and unless you don't mind the aforementioned temperature gradients you'll need to have a fan running anyways.
6) It's much much easier to find people who know how to work on forced air in the US. Replacement parts are easier to find, and forced air is usually more efficient.
One big benefit to steam is that old systems were massively oversized for the houses they are in. My house had absolutely NO problem getting up to temp during the week we had below zero temps. It was running a lot though, and the leaking shutoff valves and blurbling air vents were starting to drive me nuts, so I shut off the heat for a couple of hours on the coldest day (which got down to -40 or so) to do repairs. It got down to about 40 degrees inside before I turned the heat back on, and once it was back on it got up to 70 or so no problem. I know a lot of people who had more modern systems that struggled to maintain 40 or 50 degrees.
Side note: If anyone has a steam heat system in their home, I would highly recommend picking up The Lost Art of Steam Heating by Dan Holohan. As I mentioned, it's hard to find a real steam heating expert nowadays, most of the time you'll just get plumbers who are moonlighting. While steam heat involves plugging pipes together, there are a lot more things to take into account other than whether pipe A and pipe B are connected.
This is incredibly true in intermittently occupied areas. Just heating the air is one thing. Heating everything around it just to get warm is another.
If you're in a private/semi-private office 9-5, having your own heater or window A/C gives you a lot of control and is incredibly efficient since you can turn it on/off with your work hours.
It will cycle a lot because the envelope doesn't change temperature much (eventually heating/cooling the air), but you reduce consumption a lot by not getting to the point of majorly heating/cooling the envelope, just the air (which is what matters unless a mattress is involved).
With radiant heat, if I want it to be 55 while I'm sleeping, and 70 when I leave my bed, my heat will come on 2 hours before I'm going to get up, and the temperature will slowly increase over that time, leading to it being far too warm about an hour before I intend to get up. The heater turns off as the house gets up to temp, but the radiators thermal mass is such that they are still quite hot when I leave for the day, in effect heating the house long after I've left.
Furthermore, it's basically impossible to have zone heat with one-pipe systems, and it's much more difficult with two pipe systems. (This is less of a problem with hot water though).
With forced air, you can install active vents fairly easily, and while you still can't cut off TOO many of the vents at once, lest you cause too much back-pressure in the system, it is fairly easy to heat one side of your house or one floor of your house.
On top of all that, steam doesn't handle short-cycling well, so you have to accept a wider swing in temp than you do with forced air.
Ex1: https://www.heat-timer.com/steam-outdoor-reset-2/ Ex2: https://nest.com/support/article/What-is-True-Radiant#how-it...
Steam and other radiant systems don't deal with temperature setbacks well. In the UK and Ireland people perform a deep setback (on their hot water system) overnight then usually just accept that the house will be cold in the mornings.
This is a reason I really like our forced air system (New England house from the 50s): it can circulate and filter the air even without heating or cooling it. I have it run hourly for part of the day and it does wonders to keep the air fresh (i.e. we really notice when the system is off for some reason).
If the fuel is natural gas or propane, and,
If the thermocouple and gas valve are millivolt,
Then you can have fully functioning heat even when the power is out, because the steam does not need either a blower or a circulator pump to circulate, and the thermocouple provides the electricity to operate the thermostat and the gas valve.
Even without a millivolt system, a small UPS can run the little low voltage transformer a long time, while it would take an impractical amount of battery (which will crap out in only 3-4 years) to run a blower or circulator for any length of time.
Your points are all valid, and frankly outweigh this one except maybe at a vacation/camp house or something way out where the electric is bad, but this is something nothing else can claim.
It is pretty nice that when the ice storm pulls the power lines down, and, because it's an ice storm and it affected entire states all at the same time, your power may stay down for days, and does so coincidentally in the winter...your heat just keeps working even if nothing else does, and indefinitely not just until a 45 minute ups runs out.
Also... everyone with steam, replace your pressurtrol with a vaporstat and run your system at much lower pressure than it's probably set at. 1/2 psi per floor not counting basement, at most.
You will never find one of those houses still using it. Doesn't matter who built the house, where, when, what make of components or which fuel. Usually electric, but there were quite a few gas. Effectively all have been ripped out now. You might find evidence of ducting when rewiring.
This was quickly condemned when we got it inspected - it had a crack in the heat exchanger and - these things were apparently only designed to last 25 years. Lasting 50 was pretty impressive/dangerous.
You can get replacement heaters for them, but it's pretty niche, needs to be imported from Europe, and because of the age normally involves dealing with asbestos insulation. It's also apparently virtually impossible to get the venting cleaned, as it's embedded in the flooring (under concrete for the group floor). (Edit: And this caused real problems for Asthma sufferers)
While it worked though it was pretty good, it heated rooms extremely quickly.
From various encounters over the decades they all seemed to come with feint smell of burnt dust or just dust. By the eighties it was usual to see a radiator in front of or above the old vents, or the vents simply boarded or painted shut.
No idea why it was so comprehensively and quickly rejected despite that brief popularity. Maybe the impossibility of proper cleaning or expense? Though I wonder what they do for cleaning in the US. It's the only approach that hasn't managed to keep some spot in the market even after large declines - like storage heaters or, more surprisingly, even coal have.
Also, the heat in Santa Fe is much different than that in New Orleans.
>Newer water heaters have built-in heat traps.
So it's "no, they're just done by default and hidden". Seems like a reasonable thing to do.
With 3/4" PEX at a relatively slow 1.5 GPM flow rate, that adds about 1.4 seconds to 'hot water delivery time', though that's not entirely accurate, because the loop itself has some heated water trapped in it.
It's a bit like people from 200 years ago debating what to make the candle holder from so that it lasts 100 years.
The whole house is built (and maintained) with a different mindset.
(I think they removed the last knob-and-tube light socket a couple years ago)
My concern is not so much that a sample of 100 PEX heat traps installed today will all burst on the same day in 2059 or 2159. It's more that if the average lifetime of PEX under those circumstances is only, say, 50 years, maybe there will be a few percent that burst in 5 or 10 years, depending on how variable the corrosion rate is. So if you can protect yourself from that with 45 centimeters of copper, which will cost a heck of a lot less than the water heater does, maybe you should.
- '-'---'---,---- '-'---'---,--- ...