Constant circulation, outdoor reset and TRVs

dm9321287

I've spent a fair bit of time over the last couple of heating seasons trying to figure out whether (i) constant circulation with ODR or (ii) setback is more efficient, and broadly, not finding an entirely satisfactory answer - IMO, the former is much more comfortable, but the latter is somewhat more efficient, particularly in properties with poorer insulation.

A trusted heating engineer recently told me not to use TRVs with ODR. He said it was less efficient, but didn't explain why. I got a second opinion from another trusted engineer, who agreed with my instinctive view that TRVs just hold the rooms steady, addressing potential overheat, so why would they make the system less efficient?

This week though, in another discussion on here, a fellow homeowner raised the same point, which got me thinking, is there something in this, and I think there might be.

When running a system using ODR, the goal is to put in just enough heat to maintain the property temperature in all conditions. This is, of course, almost impossible, especially where there are some system limitations. Mine, for example, runs a Viessmann 100 NG mod con, which has basic ODR. I can move the curve up and down, but not change the angle. The 200 has full ODR. I didn't know this when the boiler was installed (by a not so trusted engineer) 10 years ago.

Given system limitations, at certain temperatures, the house will overheat or underheat. This can be compensated by also running a room stat, but that results in the system shutting down, then cooling down and noisily coming back up to temp when the demand kicks back in - i.e. not constant circulation, which gives a much steadier heat.

TRVs are therefore beneficial in preventing shut down in any system with a heat authority other than 1.0 in all conditions. They also compensate for imbalances in the system, and are advocated for constant circulation. Without TRVs, some rooms would overheat if the heating is running constantly, unless the system balance is spot on. The comfort of combining TRVs with constant circulation even on basic ODR is exceptional. The property temperature never changes.

So why would a knowledgeable heating engineer says they impact the efficiency of the system? The answer I'm contemplating, is that the TRVs effectively resize the system continuously and, when the property is at temperature, they dramatically decrease the required heating load.

Now, my first instinct is that reducing the heating load must be a good thing, right, because it reduces the amount of fuel required to heat the property. However, what it also does, counterintuitively, is potentially increase the cycling of the boiler.

My Viessmann 100 doesn't have the best modulation ratio. Whilst the system can pump around water at any temperature, the boiler won't fire to heat up that water to less than about 35C/94F. Below that temperature, the boiler cycles on until the temp is exceeded, and then off again, until the temp falls back, and repeats. Not frequently, but this is basically what it's doing. So right now, for example, it's about 5C/40F outside and the water is circulating through the system. The boiler is not firing, and will not fire again until the circulating water temperature falls enough to need another heat injection. Then the cycle will repeat.

All very interesting, you are probably (not) thinking, but what has this got to do with the TRVs? Well, my key question is this. Do the TRVs exacerbate this effect by shutting down radiators and therefore returning warmer water to the boiler, such that it cycles more (because even less heat is required to reach the target flow temperature)? I should mention we generally have a good 20C delta T.

Instinct tells me this might be the case, but surely it's marginal because if the system were finely balanced, the system heat demand would be practically the same anyway, wouldn't it? In other words, don't the TRVs effectively rebalance the system based on heat demand?

How does all this effect system efficiency? My understanding is that if a mod con boiler is cycling more, it's less efficient, so modulation is better than cycling. However, with the inherent limitation of my own boiler (relatively poor modulation ratio), if the ODR is designating a lower flow temperature, it's going to cycle anyway, isn't it, regardless of TRVs?

I'm pretty sure I've not fully formulated my thinking on this. This is definitely more of an inkling as to the possible reasoning behind this engineer's statement that TRVs make reset less efficient, which honestly, I can't entirely fathom, given TRVs are advocated for constant circulation, not least for the obvious reason that few people like sleeping in a 21C/70F bedroom.

Maybe TRVs do make ODR less efficient, but if fully opening them all up means less comfort, I know which I'll choose. Nonetheless, I'm intrigued and would be grateful for thoughts on this one.

hot_rod

I guess you need to define efficiency? Boiler efficiency, distribution efficiency, pumping efficiency, fuel utilization efficiency? Or how inefficiently he explained efficiency.

To me, a perfect system, properly designed, installed and balanced would constantly modulate to cover the load at any given time. My experience has shown TRVs with a ECM delta P circulator and a boiler sized properly and controlled by indoor and outdoor reset is as close to perfect as you get.

The beauty of TRVs is they are an indoor feedback proportional device. An electronic wall thermostat, even sophisticated ones, are mostly a bang/ bang switch. Whereas the TRV constantly modulates to cover the ever-changing load.

Two way to control heat output from an emitter, vary flow, vary temperature. Temperature control is very linear, that's why the ODR profile is a straight line. flow regulation not really, it a bell curve. So a proper TRV system with SWT modulation should be ideal.

The main drawback of ODR is homeowners that try deep setbacks. You would want some sort of boost function to overide the ODR curve to speed up a deep recovery. Some boilers have that ability to increase temperature if the boiler output load falls behind the load in a "time" period.

I guess at the end of the day, are YOU comfortable and happy with the systems performance and energy bills Does it really matter the engineers opinion of your comfort system?

dm9321287

Thanks for taking the time to share your thoughts @hot_rod. My system is certainly short of perfect, lacking a wide modulation ratio, adjustable reset curve, and direct indoor feedback, and is likely not ideally balanced. However, it's currently performing well with it's current set up so, albeit I'd like it to be more fuel efficient.

With that as my target (and the root of my engineer's point, I believe), the question would be, is there anything more I can do with the current set up? I don't actually think there is. I'm running constant circulation with the master zone (coldest room) at around 19.5C/67F when unoccupied (with the room stat set to 21C/70F as an occupied high point). Using the TRVs, the rest of the house varies from about 17C/62F (bedrooms) to 20C/68F (breakfast room). One bathroom has no TRV and is balanced down. There's also an ABV in the system (a UK legal requirement). It's almost static at those temperatures throughout and is incredibly quiet and comfortable. Most of the time the rads with the lower set TRVs are off.

In terms of fuel efficiency, it's comparable to setting it back overnight, which requires me to raise the reset curve to facilitate recovery, and which is also less comfortable, as everything else cools down so the house feels colder even when the air temperatures are comparable (so set back gains are lost through higher room stat temp and higher flow temp).

Back to the original point - are TRVs making it less efficient? I cannot see how they can be. If I opened them all up, some rooms would eventually overheat unless I were able to balance everywhere perfectly. That probably being beyond my capabilities, I'm going to let the TRVs do that for me in real time, constantly reacting to indoor conditions and, as you said, operating as indoor feedback to the system overall, i.e. they're not directly talking to the boiler (like an OpenTherm stat) but they are indirectly, by adjusting the return flow temperature, which affects how much fuel the boiler needs to burn to hit the target flow temp on the curve. Were I to open them all up, the boiler would receive less real time feedback and logically, be less responsive to indoor conditions (which I consider to be a measure of efficiency).

Which is all a roundabout way of acknowledging your thought provoking input, and hopefully finally recognising (after several years of living with a (limited) ODR system), that 'set it and forget it' really is the only way to operate these boilers. It's just our reluctance to run 24/7 that prevents them doing their job properly and most efficiently.

Thanks again.

Wellness

Isn't it possible that another factor to consider is a home's level of insulation and heat loss over time? For example, I definitely save energy blocking heat calls on my insulated indirect DHW tank for 7 hours every night. Over those 7 hours the water tempurature in my tank falls about 7 to 10 degrees if nobody runs hot water after 10 pm, a loss I can recoup in about 15 minutes when the DHW comes back on line at 5 am in the morning. When I left the tank on, without the 7 hour blackout period, it fired for a total of 28 minutes overnight to keep the water at 138 degrees. Space heating is a little different obviously because the firing rate fluctuates according to outdoor temps. But if a house is well insulated and the DHW heater comes on at the boiler's max firing rate in the morning at any rate, it could be that turning off heat calls at night could save energy. It's pretty clear from this thread lots of tesing is needed, especially when TRV are figured in the equation, so your mileage may vary.

tocker

Glad I found this discussion. I’m replacing the system in a 1955 Ranch with 1500 SF on the main floor and 1500 in an open basement. I’m considering an NTI FTVN 150 Mod/con, 130K output, or FTVN 110, 90k output (on ODR) and 10:1 TDR with home-runs. Two zones would each be on space thermostats with each having 3 Runtal radiators and would have 15-20,000 BTUs in each of those 2 zones. I’m having second thoughts on my plan to have a 3rd ‘group’ of 3 baths and 3 BRs on TRVs (while using ODR) based upon feedback here and elsewhere. I’ve been told multiple times how the TRVs will work against the ODR and make the system inefficient and no one will be comfortable, but it seems counterintuitive, given that there are two ‘larger’ zones. I don’t have physical space for a buffer setup. Won’t the larger zones with space thermos and ODR keep the boiler from short cycling by themselves? Opinions and recommendations accepted.

Jamie Hall

Using space thermostats -- including TRVs -- and using outdoor reset correctly (that isn't always done!) and two quite opposed control philosophies.

Both of them are completely valid. Both can provide highly efficient results if designed (mostly a sizing game) and set up properly.

They are not compatible with each other.

Let's consider. First, space thermostats, including TRVs. The controlling variable is the space temperature. As this changes, it causes a thermostatic device to turn on, if the space is too warm, or off when the set point is reached. Now there are some what might be called tweaks to this -- some much more sophisticated than others -- to make the variation in temperature of the space small. The variation cannot be made zero -- but a properly designed control device can limit it to half a degree or less.

Now the control device controls something -- in this discussion, it would be a boiler which provides hot water or steam to a radiator, but it could also be electric resistance elements or a heat pump or a forced hot air furnace. Whatever it is, that heating device must be large enough to ensure that enough heat can be delivered at the coldest design temperatures (it should not be larger in most installations, and for steam it is sized differently). At warmer outside temperatures it will run less often, but if the system is properly designed this will not reduce comfort, but may not provide maximum efficiency.

The philosophy of an outdoor reset, constant circulation system -- particularly when combined with a modulating boiler or furnace -- is completely different. Instead of using the space temperature as the controlling parameter, it uses the outdoor air temperature and varies the power of the heating system so that the power being supplied, in principle, exactly matches the power required to heat the space. The circulating hot water or air temperature is varied based on that so that the radiators where they are used (forced air is a bit different) supply exactly the heat required by the space -- and no more. If this principle is combined with a modulating combustion system which can operate at low temperature, it is possible to run at a low enough temperature that the water vapour from combustion is condensed in the boiler (or furnace), increasing the efficiency of the system.

Now what happens if you add a space thermostat or a TRV to the outdoor reset system? Remember that the system is supplying just enough heat to heat the space to the desired temperature, provided that the system never stops. Now if the TRV turns off the heat supply to that space, the space will cool. When the TRV turns back on, however, there is not enough power to bring the space temperature back up. Further, the load on the boiler is now less than designed, and the boiler will either turn off or modulate down. If it can modulate down far enough, well and good -- but if it cannot, if the remaining operating zones are too small, it will turn off -- and the efficiency gain one hoped for is lost.

There are complications, of course, and most outdoor reset control systems also have a space thermostat which will bias the system up or down, within limits, to compensate for excess heat gain, such as from a sun load, or excess heat loss, such as from a high wind or cold rain. This is not the same as using that space thermostat to control the temperature directly.

Further, it should be understood that the above comments are really for smaller buildings with essentially single occupancy. In a very large building, such as perhaps an apartment house, outdoor reset can (and arguably should) be used together with space thermostats, designing on the probability that only a certain fraction of the spaces will be calling for heat at the same time. That, however, is a different matter.

Zman

I am surprised that you can't change the ODR angle. Older ODR systems used a ratio multiplier to set the angle rather than a top and bottom setpoint for the curve. They take a little more brain damage to program.

As for the engineer's opinion, like any other profession, there are all sorts... Some thoroughly understand it all and are excellent at what they do. Some are "memorizers", as long as the task at hand looks just like the textbook, they can design it. Others just stink at what they do.

In my experience, the success of an ODR controlled system depends on how well each room's heat emitters are balanced to the heat loss. If the balance is off, you will never get it to work well.

dm9321287

Thanks @Jamie Hall , your comments have had me thinking further about this for a couple of days.

If I understand correctly, you're saying the activation of the TRV will cause the room temp to fall below the target and if the ODR is correctly set, it will never catch back up again, correct?

Further, you're saying the TRV activation causes system size reduction, causing warmer return temperature and therefore potentially boiler cycling, reducing efficiency gains?

If I'm understanding correctly, that would certainly support the original engineer's throwaway comment. However, doesn't a TRV work through proportional flow rather than on/off, like a room stat? Therefore, wouldn't the TRV slow the flow and constantly adjust such that the temperature is steady, rather than drops below target and then has to catch up? Obviously a room stat would drop below target (mine works in 0.1C increments, as opposed to those that drop 0.5C or more), and then have to catch up - but would a TRV?

The second point (boiler cycling) is somewhat linked, I guess. Even if a TRV only reduces flow, it's still going to cause a slight uplift in return temperature, so could cause cycling. Unfortunately, some amount of cycling is inherent in my system, as I only have a relatively poor modulation ratio. The boiler simply can't handle very low temperatures, but then again, I'm not sure many would (i.e. put ANY heat into ambient flow temperature of 20C plus and it's very quickly going to rise to over 30C).

I feel you're definitely helpfully shedding some light on this for me, but I can't quite grasp something here. What I think my man was saying was the TRVs, by reducing flow (when above target temp), effectively make the rads smaller, and with ODR you want oversized rads, not smaller ones. I'm following so far. What I don't get though, is why that's an issue, if the room is already warm enough, other than, as you say, it can effect the efficiency through return flow temp.

I guess at the end of the day, as @zman says, it all comes down to system balance and flow temperature. If the system is perfectly balanced, and the flow temperature is set up correctly, then (a) you won't need TRVs and (b) if you have, and you set them to target temperature, they wouldn't actually ever activate (because the temperature would never exceed target, due to proper balancing and flow temp). But, and it's a big but, whose system is ever going to be perfectly set up? It can be near perfect, but there's bound to be something out of line in there, particularly if you have DHW priority because as the water kicks in, the space temp falls, therefore you need to build in slightly higher flow temp setting to catch back up (or it's going to keep falling each time there's a call for water). And then there's the windy/wet day issue, unless you have boost (which itself reduces efficiency). Or the system has a basic ODR feature where you can't adjust the angle of the curve (like mine, @Zman). And then there's insulation (@wellness). I live in a 120 year old solid wall house with some original single glazed windows. I can run the heating 24/7 and it can hold temperature, but from experience, that uses a lot of energy, which if the goal is efficiency, it most certainly isn't realised. So then you need some setback, which means a higher than exact ODR flow temp, to recover from setback.

Which is all a roundabout way of saying, perfectly balanced system and ODR, plus oversized rads, plus a good level of insulation makes for good and likely efficient constant circulation and probably negates TRVs. If any of those are sub-optimal (and that's most of the properties and systems here in the UK), then I'd bet it's never going to work properly. Am I wrong?

Thanks again for your input guys. I'd be really interested in further comments on this.

EdTheHeaterMan

I'm wondering how large this project is? ...and the law of diminishing returns. If the project is large enough a 0.01% increase in efficiency could yield thousands of dollars of savings.

Somehow I don't think this project is that big. I think it was Gill Carlson who once said, "...for a difference to be a difference it needs to make a difference." I think there is a lot of overthinking going on here. The title of the discussion says it all. ODR with constant circulation (properly designed) is the most efficient hydronic system. TRV can help in a design-flawed system. Tweak it over time (and experience) and you can't get any better than that.

I think Dan's "Dead Men Tails" this past week was spot on when it comes to engineers "Building" a sandwich. Overthinking will not make the job any better either when making a sandwich or a heating system. Stick to the basics. when it is time to replace your system then look for better controls and designs at that time.

Just an opinion

Mr.Ed

Jamie Hall

Amen, @EdTheHeaterMan . Years ago in another universe I had the acronym "KISS" hammered into me... and not by my girlfriend...

dm9321287

Yep, you're right @EdTheHeaterMan, the efficiency gains in question are likely to be minimal. I was intrigued by the throwaway comment to not use TRVs with ODR (or weather compensation, as we call it over here). Just from the comments above, there's clearly divided opinion on this. For what it's worth, I'm concluding TRVs are useful if the system isn't perfect (as you say). Mine isn't. So when I next see the engineer who shared his TRV pearl of wisdom, I'll have a little chat. You never know, he may actually know the answer. Hopefully it's more than 'because it is'.