UFH thermostats and hysteresis settings
I have a GSHP (Geothermal) supplying warmed water to a hydronic underfloor heating system which is controlled in six zones by six Warmup 4ie thermostats which do not have a hysteresis setting. The problem I have discovered is that the thermostats activate whenever the ambient temperature drops to within 2 deg C of the target temperature and then deactivate 30 secs later (which seems to be more appropriate for an electric UFH system). This has the effect of activating the heat pump and sending it into a heat production cycle but this is then cancelled when the thermostat drops its demand and before the hydronic valve actuators can open. There are two adverse effects of this: 1) the heat pump can sometimes shut down due to over pressure but this has been solved by leaving one of the hydronic circuits permanently open and 2) the inverter which drives the compressor receives too many activation cycles and fails.
My question is: what in room wall mounted thermostats are more suited to a hydronic UFH system and which include a hysteresis setting allowing the ambient temperature to drop below the target temperature and thus create a longer cycle time which (in theory) should limit the compressor starts and prolong the life of the inverter?
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Does seem like they are designed for electric heat[120/240 volt]
Forgive a stupid question, but is there no buffer tank in your system?
Seems like the thermostat should not be turning on the heat pump, but turning on a circulator pump and the system water temp would drive the heat pump.
Not a stupid question at all! The water circuit has a volumiser tank but it is not configured as a buffer tank ie the "flow" is shared between the volumiser tank and the heating circuits and the return is passed through the volumiser tank. There is therefore no thermostatic control of the heat pump by this tank. I attach a schematic.
The volumiser looks exactly like a 3-pipe buffer tank.
How many gallons (or liters) is it?
And what is the rating of the heat pump in BTU/hr or kW?
Does it have min/max ratings or just one rating?
It's a Clausius Classic 5-25kW
it's a 200 litre tank
Have you posted your question on the GeoExcange foum?
https://forum.geoexchange.org/
Walt
Not yet, Walt as I am new to this group and seem to be getting good feedback and ideas so far but I might check it out later.
The simplest is to get the correct thermostat. That will add up for that many zones though.
The bandaid you can do is install an off delay module something like this:
https://ca.rs-online.com/product/rs-pro/2257036/73028922/
You can get cheaper ones from your favorite on-line retailer.
Install this on the output of the thermostat before the zone valves. You can set the off delay to say 15min, this way the system will continue to run after these short calls which avoids the short cycling.
This extra bit of delay should not effect the temperature enough to be noticable.
Thanks for this. I was trying to work out if it would be possible to get away with just one such device on the output of the control board but it would make more sense to install one on each of the six thermostats' outputs as you suggest. This would indeed be a bandaid solution and maybe fit a true buffer tank as has been suggested elsewhere in the longer term.
You don't want the room thermostats controlling the heat pump directly, because the demand for heat doesn't match up well with the heat pump's ability to provide it. The proposed solution of increasing the hysteresis just sacrifices comfort -- and the whole point of a complicated system like this is comfort.
Here's what would work better: use a buffer tank, and put a thermostat on it that controls the heat pump. You can put wide band hysteresis on the buffer tank -- 5C is common -- so the heat pump can have nice long runtimes even when the demand for heat is low. There is a separate circulator that pulls from the buffer tank and returns to the buffer tank whenever any of the zone thermostats calls for heat and opens the zone valve.
I have an air-to-water heat pump configured this way and the hysteresis on the zone thermostats is 0.1C. It keeps the temperature rock steady all winter.
The buffer tank has a 3-pipe configuration as shown in this article:
https://www.pmmag.com/articles/103227-john-siegenthaler-combining-heat-pumps-with-buffer-tanks
The question is where to put the temperature sensor to get reliable readings. Depending on how many zones are calling the flow through the buffer tank changes, which affects how the tank temperature reflects what's going on with the system. My heat pump manufacturer (Chiltrix) solves this by putting the sensor in the return line to the heat pump and circulating water continuously even when there is no call for heat.
Question: is your heat pump capable of modulating, or is the output fixed and it's just on or off?
Thanks for this response. I'm tending to agree that this is the longer term solution and I wonder why my installer didn't configure it to include a "true" buffer tank but since he has given up heat pump installations, I figure he didn't know that much about the complex interactions between the various elements. The problem is, it is the most expensive and disruptive solution so probably deferred until next year.
The heat pump modulates the output based upon the weather curve but not sure it does so based upon the buffer tank temperature although there is a buffer tank sensor option (if configured with a true buffer tank).
Modulation and varying water temperature are separate but related functions. Modulating is slowing or speeding up the compressor in the heat pump to reduce or increase the output and try and match it to the load. You can modulate and have constant output water temperature, and you can vary the setpoint using a single-speed compressor and simple on-off control.
The reason I ask is that if you have a modulating heat pump, it has to have some way of sensing what the actual heating load is. The usual way is to monitor the send and return temperature and the flow rate. The heat pump will have a minimum ability to modulate, if the actual heating load falls below it will have to cycle on and off, the buffer tank provides a soak to absorb heat and keep it from short-cycling.
This got e thinking again....the compressor is controlled by the inverter which is the part that has failed twice - cause undetermined. The reason for this, so I am told, is to provide variable compressor rotation speeds. It is certainly possible to adjust max and nighttime rotation speeds so I'm fairly sure it does what you are suggesting and modulated the heat output ie it isn't just an on/off compressor. It definitely can sense the flow and return temperatures because I see these on the app.
"The heat pump modulates the output based upon the weather curve."
This statement gets me thinking.
What does the heat pump do if you close all the zone valves yet leave it enabled?
Because that's the way my heat pump is configured. It's enabled all the time, there is no thermostatic connection to the inside of the house. There is a set point on the water temperature, when that temperature is exceeded (by 2C) it shuts off, when it drops below by 2C it turns back on again. The setpoint automatically varies with the outdoor temperature.
If your heat pump is similar, all you need to do is cut the link between the indoor thermostats and the heat pump. Just let the heat pump run all the time. If there's no demand it will fill the volumiser and shut off.
This could be a really simple fix.
Firstly, to answer your question: if the heat pump is activated by the thermostats but the actuators haven't yet opened then the system occasionally shuts down on "over pressure." I actually solved this by simply removing one of the actuators which leaves the zone valve open.
On another chat group, it was suggested that I use the weather curve to adjust the flow temperature and set all the thermostats high. I think this would have a similar effect to what you are suggesting so if the winter setting (eg outdoor temp at , say, 5 degC) has a target temp of 28 deg instead of 38 deg, and runs continuously, the house should remain comfortably warm and if it is too warm, tweak the output flow temperature. Is this what you are suggesting also? It's certainly worth trying before I replace all the thermostats at a cost of £120+ each!
I'm going to put forth the hypothesis that your installer didn't know what he was doing and misconfigure the system.
So far, everything I've read sounds like you system works like my air-to-water heat pump. Here's what the manufacturer's manual says about the on/off input on mine:
NOTE: THIS SHOULD NEVER BE USED FOR A HEATING OR COOLING “CALL”. IT IS ONLY FOR MODE
SELECTION. Heating or cooling calls are not needed or allowed with CX34, as it monitors the returning water temperature and always knows what it needs to do.
The problems you're having -- poor temperature control and frequent burnout of the inverter -- would be consistent with short-cycling.
My system is 33,000 BTU/hr nominal size, so about 10 kW. The minimum buffer tank size is 19 gallons or 72 liters. Your system is 25kW with a 200 liter tank. So the tank capacity to the heat pump output is almost exactly the same. My heat pump is capable of modulating down to 25% of its rated output. When the heating load is below the minimum modulation it cycles on and off, with a hysteresis of 4C. So the tank represents a minimum run time of 12 minutes.
The one statement that has me puzzled is this:
" if the heat pump is activated by the thermostats but the actuators haven't yet opened then the system occasionally shuts down on "over pressure." I actually solved this by simply removing one of the actuators which leaves the zone valve open."
Is there a zone valve between the heat pump and the buffer tank (which is what I'm going to call the volumiser because that's what it looks like to me.) The heat pump should be able to freely flow into the buffer tank at all times.
I think your system can be fixed just with some rewiring.
Looking at the diagram I don't see an expansion tank. Maybe it's there but not shown. I associate over pressure with an insufficient or malfunctioning expansion tank. Although I don't understand why having more zones open would mitigate it.
There are no zone valves between the heat pump and the "volumiser." The zone valves simply turn on and off the flow to the various rooms in the main house which is about 5 metres away from the plant room where the heat pump resides. A thermostat call for heat begins the actuator process which takes a few minutes to open the valve associated with that zone. It also activates the "boiler on" and "pump on" signals. These are wired into the heat pump although I have not yet traced exactly where. To answer your next question (below): initially, there were no expansion tanks fitted but the service agent subsequently fitted two, one on the brine circuit and one on the heating circuit. These are not yet shown on the schematic (sorry!) as they were both subsequent additions.
When the service agent left, he inadvertently left the manual isolation valve between the volumiser return pipe and the heat pump closed. This led to immediate shut downs on "over pressure." Upon opening this valve the failures diminished but did not stop. Upon opening one of the zone valves, the failures ceased. Note that there is no bypass channel at the manifold position so my assumption was that the "pump on" signal started the heat pump's "flow" pump but before the zone (s) could open resulting in a "over pressure" failure. This may be poor logic or a misunderstanding of what "over pressure" actually means. The manual is written in Spanish and translated into English so can be a bit weirdly worded at times!
Again, I really appreciate all the insights you and others are giving me. I have several ideas to try out and I hope to be ready to implement one or all solutions when the weather turns cold again.
As to my supplier/installer....don't get me started! Suffice it to say, I am pursuing the company through the courts but since they lost their only staff member with any knowledge, they have stopped doing heat pump installations. They are MCS registered and I now find that that means absolutely nothing! The installed system is, as you rightly say, quite complex and the assumption has been that it will all work just like a boiler/furnace arrangement but this is simply NOT true when you understand how a heat pump works. I'm not a heating engineer but just a user with an engineering background and I'm having to teach myself about heat pump system design if I'm ever going to get a reliable system. This, and one other FB group, has been immensely helpful in that respect.
A couple of issues here. DC has touched on some. I think the good news is that you have most of the bits, it is just not wired/configured properly.
First step is to set the heat pump to monitor buffer tank temperature. Assuming it has its own internal pump, it should then cycle on off to keep that buffer tank at the setpoint. Your thermostats should not be connected at all to the heat pump. This means adding a temp sensor to the tank or configuring the heat pump to use the return water temp for controls. If it using the return water temp, the pump on the heat pump will need to be configured to run at some minimal flow at all times (usually in the settings menu somewhere).
While at this, you also want to set up outdoor reset. This is an extra temperature sensor (sometimes built into the outdoor unit) that adjust the buffer tank temperature based on outside air temp. If you dial this in perfectly, the zones should run pretty much all the time. This will reduce the cycling and most importantly bump up your seasonal COP significantly.
It also sounds like your zoning setup is miss wired. The thermostats should not be driving the circulator. What you want is the thermostat to drive the zone valve, when that zone valves opens all the way, it will trigger its end of travel sensor. It is this sensor signal that should drive your pump.
Post #6 suggests to me that outdoor reset is already enabled.
I believe it is and I did previously adjust the "weather curve" as it was set too high for the system at 48 deg C flow temperature with -5 deg C outdoor temperature.
"the assumption has been that it will all work just like a boiler/furnace arrangement but this is simply NOT true when you understand how a heat pump works."
Yep. I've run into guys who feel that hydronics is hydronics, a heat pump is just like a boiler only noisier. Although with geothermal it's probably quieter.
It sounds like you're facing a significant issue with your current thermostats not being suited for your hydronic underfloor heating system. For a hydronic UFH system, you'll want thermostats that are specifically designed to handle the longer cycle times and include a hysteresis setting to avoid frequent short cycles.
Thermostats such as the Honeywell T6 Pro Hydronic Thermostat or the Nest Learning Thermostat can be more suitable for your needs. These models allow you to set a hysteresis range, ensuring the ambient temperature can drop slightly below the target before reactivating the heating system, thereby reducing the number of cycles your heat pump undergoes.
By adjusting the hysteresis settings, you can create a more stable environment and prolong the life of your inverter. Additionally, these thermostats are designed to be more compatible with hydronic heating systems, providing better control and efficiency.
For those looking to manage their electricity bills online, especially if you're a GEPCO customer, you can easily access and pay your bills through our platform. Visit GEPCO Bill https://gepcoonlinebil.pk to get started!
Just for reference, here is how my air-to-water heat pump operates (Chiltrix CX34).
In the configuration a setpoint is entered. This can either be a constant temperature, or it can vary by outdoor temperature. The control software within the heat pump monitors the outgoing temperature, the incoming temperature and the flow. The heat pump has a variable-speed compressor and a variable-speed circulator. At all times it tries to match the output of the compressor to the heating load, and the speed of the circulator to maintain a 10F difference between the water being delivered and the water returning, while also maintaining the setpoint temperature.
The minimum modulation of the compressor is about 25% of rated output. If the heating load is less than that it won't be able to maintain the setpoint and the water temperature will rise. When the water gets to 2C/3.6F above the setpoint, the compressor shuts off and the circulator keeps running. The water will cool, when it gets to 2C/3.6F below the setpoint the compressor will come back on.
The variable-speed circulator in the heat pump has a minimum flow of about 6 liters/1.5 gallons per minute. It has a maximum rate of about 20 liters/5 gallons per minute.
My plumbing is pretty much identical to the diagram that was posted. I have a thermostat in each zone, and a zone valve on each zone. The end switches are connected in parallel to a relay that controls an external circulator, if any of the zone valves are open the circulator runs. There is no control connection between the heat pump and any of the internal components.
If the flow falls below 1.5 liters per minute the control software throws an error and the heat pump shuts off. There is a filter on the return line, typically this happens when that filter needs cleaning. I don't know where the debris is coming from, but I've learned to clean that filter twice a year when I switch between heating and cooling for the season.
DC, question about your setup:
"I have a thermostat in each zone, and a zone valve on each zone. The end switches are connected in parallel to a relay that controls an external circulator, if any of the zone valves are open the circulator runs."
How does this system work if no radiant panel zone is calling for heat, but yet the heat pump unit detects a drop in entering water temperature? What keeps it from running and heating up a very small loop of water (and/or buffer tank if you have one)? Your zone valves being shut stop the circulator from running, but what stops the compressor?
That's exactly what happens, and that's what the buffer tank is for.
When the compressor is off the circulator runs at about 1.5 GPM. When the water cools to below the setpoint the compressor kicks on and the circulator kicks up to match the compressor output. If no zones are calling the output goes into the buffer tank and after about 12 minutes the upper limit is reached and the compressor turns off and the cycle repeats.
While this sounds wasteful, keep in mind that no zones calling is only going to happen when the outdoor temperature is warm. The heat pump has outdoor reset, so when it's warm outside the water temperature drops, this is typically happening when the water temperature is only a few degrees above room temperature. At warm outside temperatures the COP of the heat pump is very high. Also, the tank, the piping and the heat pump are all insulated so the standby loss is low. Most of the heat that is lost goes into the building envelope anyway, and since it's heating season that's useful heat.
This is how most traditional boilers work as well -- an aquastat controls whether the burner fires, the thermostats control the zone valves which control the circulator.
I thought the circulator would not run unless one of the zone valves calls for heat? Or is this a secondary circulator, different from the one included with the heat pump unit?
There are two circulators. One, internal to the heat pump, runs continuously and variably. The other, external, runs when a zone calls for heat. If the compressor isn't running it's pulling heat out of the buffer tank.
Do you run the radiant panel zones in the cooling season as well, or do you rely just on the FCU zones during those seasons? I would think a motorized mixing valve would be necessary to keep radiant panels above dew point temperatures.
Right now I just run FCU's in the cooling season. My plan is to eventually run the radiant panels as well but I haven't gotten there.
To meet my heating load I needed two heat pumps, so I plumbed one to the FCU's and one to the panels. So I could run different temperatures in the two systems.
I'm not a fan of mixing water to temper it after it comes off the heat pump in either heating or cooling. Your efficiency and capacity is determined entirely by the temperature delta, if you heat water up just to cool it, or cool it off just to heat it, you're throwing away efficiency.
This is an example of the difference between traditional hydronics and heat pumps. When you're burning fuel a BTU is a BTU* and there's no penalty to heat water up and then mix it and cool it off to get just the right temperature, it's done all the time. It's just not a good idea with a heat pump.
*(I realize with condensing boilers this isn't quite true, there is an inflection point between condensing and non-condensing. But on either side of that inflection point it's true.)
re: water temperature. When you mix hot water you are mixing it with already heated water so there is no loss, but it would seem to me heating water hotter than you need it is always a waste of energy.
I think that mostly when you mix water in radiant heat it is to get the correct water temp for your application, I had three different mixing loops in my last house. In that application it had zero to do with efficiency.
I would think outdoor reset would save energy in a heatpump situation since it takes energy to heat up 1xx degree water that you never need, and I thought that the spread between input and output temps was related to efficiency.
It also occurs to me that since it probably isn't the thermostats that are the issue, and that if he has a decent ODR setup, the lack of hysteresis in the thermostat might not be the biggest deal, since water that is just as hot as it needs to be ought to result in less overshoot
Are you sure that when it shuts down it is due to over pressure and not low flow? Could the error be mistranslated?
That is indeed possible!
Just thought I would update everyone and thank you for all the advice.
I removed one of the actuators that controlled the UFH zone in our entrance hall. This stopped the “over-pressure” alarms and associated shut-downs. I found the buffer tank thermostat coiled up inside the heat pump unit (!) with one (of two) wires connected. I reconnected it fully and extended it to the buffer tank. I changed the settings to “buffer tank” mode and adjusted the weather curve so that the maximum temp is 35 deg C when outdoor temperature is -5 deg C (rare in Cornwall, England). I adjusted the hysteresis setting on the buffer tank control to 5 deg C. I also installed a separate remotely accessible thermometer probe to both the DHW and buffer tanks. On the indoor thermostats, I set the PID period to 60 mins from 10 mins. This seems to have stopped the frequent activations.
With one hand firmly touching wood… the house is now very comfortable and sitting at 20-21 deg C fairly consistently while the temperature of the buffer tank seems to oscillate between 27 and 33 deg C.
I might experiment a bit more with the buffer tank hysteresis settings to reduce the compressor start even more but for now, the system appears to be working much better.
Thanks again to all those who posted advice, most of which I have followed.
Congrats
Now you know way more about your heating system than you wanted to, and apparently more than the installer.