What is state of the art for controlling zoned hydronic circulator pumps?
I’m looking at installing zoned hydronic radiant heat in my house, and so far I’m surprised at several aspects of how people seem to do this.
For starters, the most efficient circulator pumps are electronically controlled and variable speed. For example, the Grundfos Alpha2. These pumps mostly have push button controls for changing parameters manually. However, the only wires available for machine control are the two in the 120V power cable. The “zone controller” (Taco SR504 seems to be popular) seems to only control the pumps by turning on and off relays. Why not ditch the relays and communicate with the pumps digitally? My guess is many pumps are suboptimally configured with the buttons. I would rather let a smart controller learn the best settings.
The second practice that seems weird is that each zone seems to get its own thermostat controller. I would expect you could use one thermostat for a whole house and use temperature sensors in rooms for feedback.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Bad Mechanic,
Like you, I'm interested in information from GBA readers on this topic.
Back in the late 1970s, when I designed hydronic heating systems at a plumbing wholesaler, we advised our customers to create zones by installing a separate circulator for each zone, with a separate thermostat for each circulator. This approach costs more than the use of nonelectric zone valves, but is more robust and trouble-free than cheaper alternatives.
I'm curious to know whether new technology has made our robust approach obsolete.
There doesn't seem to be a pump whose settings you can control externally. The Alpha2 is the most advanced I am aware of. It has an "auto" mode that adjusts speed based on some learning algorithm. However, I have two of these, and despite tinkering a bit and reading all the documentation, I have no idea how this is supposed to work (i.e. what conditions is it monitoring and how?) It seems like voodoo to me. Having said that, I think there's not really any need for the pump settings to be continuously changing. Once I got my two circuits balanced and dropping the proper temperature throughout the line, I never felt the need to tweak anything.
As for the single thermostat with sensors in each room, I can't see how that's going to be superior in any significant way. If the sensor is able to turn on and off heat to that room, then it IS a thermostat, albeit one you can't set individually and without a display. I'm sure you can get a thermostat with remote sensors and channel outputs.
>"The second practice that seems weird is that each zone seems to get its own thermostat controller. I would expect you could use one thermostat for a whole house and use temperature sensors in rooms for feedback."
Nothing weird about that- most people prefer to sleep in rooms 5-10F cooler than where they eat or lounge around watching K-pop on the big screen.
There are lots of variables to optimizing pump speeds to the types of heat sources & radiation. Hydronic heating sytems are almost always custom designs, and the pumps are not manfactured by the same companies as boiler, chillers, heat pumps. etc, all of which have their own internal and often complex control schemes. Maybe a communication standard will evolve to control the pumps digitally in a more nuanced way by the system controller or heat source's internal controls, but I'm not holding my breath on that one.
My guess is that delta-T pumps are close enough to optimal that it hasn't been worth more expense.
I was previously unaware of the existence of delta-T pumps. That would probably have been a better choice for me than the Alpha2.
Delta-T pumps are optimal for some types of systems, constant pressure pumps more optimized for others.
But the expense of defining a communications standard and integrating those into the zone controllers boilers & heat pumps would be pretty daunting. There are still many boilers delivering reasonable efficiency using dumb aquastat controls not very different that what was available 100 years ago. Inventing a high flexible dynamic wheel to replace an existing pretty-good wheel doesn't guarantee that the system designer will make optimistic use of it's capabilities. The number of grossly oversized modulating heating equipment already installed out there that is too oversized to deliver any modulation benefit demonstrates that.
Delta T circs are the way to go . The universal hydronics formula states that GPM= BTUh divided by Delta T . We design for Delta T in order to know what a system should be doing at various points in time , various because , of course , we are not always at design . The pressure sensing circs will maintain a set constant pressure throughout their operation , most of the time this will result in a narrower Delta which effects SYSTEM efficiency .
There are systems (climate automation ) that allow use of sensors and lotsa programming that can controls speeds and flows and many other things . All just dazzle that are really not worth the scratch in all but the largest projects .
Too many times though manufacturers suggested maximums and other advice get in the way of OPTIMAL design . Like maximum recommended loop lengths which require more outlets , loops , actuators yet hardly ever offer enough RESISTANCE to keep even the newer ECM circs
in a place where pump affinity laws do not take over . You can't fool Mother Nature and you never will
>" You can't fool Mother Nature and you never will"
The laws of physics ARE self enforcing, after all...
The state of the art hydronic system is one that is simplest. Coming up with a simple hydronic system is way harder than something with lots of sensors and controls. My hat off to people that can pull it off (like Flatrock Passive by David Goodyear).
Heating is something that just needs to work and when it doesn't you can call plumber that can fix it.
It is easy to get cought up in optimizing a heating system, I've gone down that rabbit hole myself at home (mix of cast iron rads, floor heat, multiple indirect hot-water heaters), I hate to think what it would cost to troubleshoot by outside help.
As for high tech parts, my experince.
Delta T circulators are pretty decent, I have used both Taco bumblebee (noisy little pump) and their non ECM deltaT units (much better but still loud when they are throttled back). All need external sensors, which can come unplugged/fail. In case of the Bumblebee it causes the pump to shut off, which is the last thing you want your circulator to do. My takeaway from this is the fewer the external parts to the pump less chance of something failing. There is a good reason why most residential pumps have only two wires!! (you can guess which pump is no longer being used in delta T mode)
Grundfos ALPHA is by far the quietest pump I've used. Uses next to no power, for anything that doesn't require a lot of pressure, can't beat it for the performance and cost.
Akos ,
Bumble Bee has been discontinued and the sensor issue has been addressed on the new VT2218 . Which , by the way , is as quiet if not quieter than Alpha .
Your idea of simple systems is spot on also .
I found this article to be extremely helpful.
http://solarhomestead.com/off-grid-circulating-pumps/
Scott,
My house is off-grid, and my solar thermal system uses an El Sid circulator from Ivan Labs to circulate fluid between the heat exchanger in my hot water tank and the solar collectors. The El Sid circulator is wired to a small PV array on the roof. I've been very happy with the performance of the El Sid circulator -- no problems at all.
Martin, but do you use them for getting the hot water to radiators, or to the taps? The article suggests that the El Sid pumps are perfect for that.