condensing boiler - higher return temp == higher efficiency?

[moggie2002]

Thanks for the further replies. I'm aware of how to balance the system, and had to do this originally when a couple of rads didn't heat up. However there's little drop across the radiators, and I assumed, incorrectly it would seem in this case, that less of a drop was better. The lock shields aren't open much as it is though, but I'm going to experiment more.

Weather Compensation? Yes, great idea, and I already bought that. After all, the whole house is automated, was I going to pass up the chance for some other tech? Hardly But, it's not connected because it mysteriously disappeared after showing it to the first plumbers so we need to get another.

One poster suggested adjusting pump speed, but I think the logic+ might a variable speed danfos that it controls automatically but not positive on that.

Lowering the flow temp is always an option. The logic does a good job of maintaining the set flow temp, it's 'e' setting aims for 67 (62 measured by my sensor and at the first rad), but it can always be set lower. I did have it at 55 until the temp dropped in the last few days.

We do have a myson bypass valve fitted, as the boiler manual suggests may be necessary, which is going to impact return temp as it starts to open. I haven't checked yet what setting that's at though.

I do hear gurgling and a water trickling/pouring noise occasionally, and wondered if that was condensate.

 

[MJN]

Interesting thread Moggie - always good to see someone dabbling in domotics and integrating their central heating into it.

Do your 1-wires feed into the control loop for the heating or are they for info only? They would obviously be key to any control of the heating however your mention of somewhat rudimentary cron job scheduling makes me wonder if they are incorporated?

This is not much help now but it's a shame you've got the boiler you've got as it'll be the achilles heal of any system given the lack of external modulation control. There are many boilers out there allow proper external control (i.e. more than on/off) through either ranged demand inputs (e.g. 0-10v) or more advanced bus-based control. Whilst some of them use proprietory bus protocols there are increasingly more that support Opentherm thus more readily allowing roll-your-own solutions.

However, you've still got the makings of a really powerful control system with a network of temperature inputs and it wouldn't be too much work to introduce time-proportional-integral control of the system and weather compensation too so it'd be interesting to see how it develops if that's the way you're heading.

Mathew

 

[moggie2002]

Interesting thread Moggie - always good to see someone dabbling in domotics and integrating their central heating into it.

Absolutely

Do your 1-wires feed into the control loop for the heating or are they for info only?

They are used slightly for feedback with the intention to use them more. There are actually several networks. The 1-wire system snakes around the house over a fair distance and has 11 sensors at the moment covering the UFH, some rooms and the boiler. I'd planned to put sensors just through the ceiling but temps at the ceiling are cooler compared to elsewhere in the temperature gradient and no clear correlation, and I don't want them on the walls, so I'm leaning towards either making an RF to 1-wire bridge and making the room sensors wireless for discrete locating, or bringing up from the floor in locations not seen. Some sensors are used to control the UFH and CH demand as well as cronjobs, but this isn't handled correctly yet and I need to write a heating server that learns about the behaviour of the heating system and the house, and also that has scheduling, so that it provides a high level abstraction as well as fine grained control and querying. I can control and view the heating remotely, but later I need to hook in Asterisk so that we can phone the house to request/control heating that way.

There's a DMX network around the house for the main lighting and localised LEDs (such as LED strips for kitchen lights and elsewhere). DMX dimmers also control the UFH, with a DMX relay board for the boiler, one of the UFH zones and a NC zone valve (for the bedrooms and thinking that NO might have been better to inhibit rather than grant heat). Items such as bathroom mirrors, fan etc. will also be controlled via DMX. In the past I used some cheap chinese DMX dimmers for lighting but they have flaws, and I'm leaning towards IGBT dimmers rather than Triacs, but that means self build and I've not had time yet. The induction hob buzzes when the heating zones are dimmed, as does the CU itself, and I'm not sure if that's a result simply from the current draw pulsing or is inadequate filtering in the dimmer pack.

There's an rs485 network that daisy chains between each light switch. I haven't prototyped the light switches yet, but they'll have several buttons with LED's, support firmware updates over the network, and do pretty much what you'd expect in terms of controlling scenes, plus possibly features like party mode (where the switch lights animate), nightlight mode etc.

Each light switch socket also has one or two t&e cables so that it's possible to install conventional switches instead. I expected to need this ahead of the rs485 switches being ready, and if it's difficult to sell the house with the tech, it should be possible to back it out without too much difficulty or expense. Rather than traditional wiring, all t&e cables for lights and switches go to the loft; there's getting on for 300m of clipsal pink cat5 gone in, and hundreds of meters of t&e.

The processor is a Fit PC2 mounted on a cupboad wall with a vesa mount and solid state drive for environmental considerations. I've one with hard drive in the loft of another property that's given no probs, but an SSD here made sense. On the other side of the wall (the kitchen) is an 8" HDMI touch screen with a rudimentary interface knocked up with QML. Some PHP, a quite powerful DMX server with control language I'd written a while back, OWFS, Apache etc. make up the other bits and bobs. I aim to put some other bits onto the machine that are running elsewhere, such as asterisk, squeezecentre (or whatever the latest name is), and the music collection and I can then shutdown a hefty dell server and save some power there.

This is not much help now but it's a shame you've got the boiler you've got as it'll be the achilles heal of any system given the lack of external modulation control. There are many boilers out there allow proper external control (i.e. more than on/off) through either ranged demand inputs (e.g. 0-10v) or more advanced bus-based control. Whilst some of them use proprietory bus protocols there are increasingly more that support Opentherm thus more readily allowing roll-your-own solutions.

The Ideal does have Opentherm, but I've not investigated that fully yet, and I've no idea how complete their implementation is. Control over more bits would be great, and the behaviour of the Ideal is proving a bit frustrating because, no pun intended, it doesn't seem to be ideal.

I'm keen to get monitoring the gas usage too, either optically from the silvered zero going past, or with a hall-effect sensor to pick up the magnet (our Actaris meter is likely to have a magnet on the zero as well as the silvered zero.), because otherwise any tinkering with boiler cycles isn't that useful. The RF OWL monitor we have for electricity is not that accurate or sensitive, and I had better results when I was counting pulses and timing the intervals between flashes on an Ampy meter, so I plan to get a 2nd hand Ampy from ebay fitted by the CU for that purpose.

More on topic, with it almost zero outside and I'm still stuck with high return temps unless I briefly activate the zone valve for the bedrooms to draw in some colder water, I wonder if boilers could draw from the cold outside air to help the condensing process rather than require the return flow itself to be cold enough (but quite possibly I'm not understanding some crucial part of how that process works).

 

[MJN]

<techy stuff>

Sounds impressive, and clearly the result of a lot of hard work! Let me guess, your idea of heaven is a new build where you're involved at the outset to run the necessary cabling rather than having to retrofit it afterwards!

Each light switch socket also has one or two t&e cables so that it's possible to install conventional switches instead. I expected to need this ahead of the rs485 switches being ready, and if it's difficult to sell the house with the tech, it should be possible to back it out without too much difficulty or expense.

Indeed - always the problem in this space. I haven't dabbled nearly as much as you in home automation (although I did build an Internet-enabled cat feeder!) but you do have to be careful how others perceive it when you come to sell. My last house was flood-wired with Cat5 which I thought was a selling feature... but unfortunately only to a select few - the rest couldn't care less or almost considered it a concern!

The Ideal does have Opentherm, but I've not investigated that fully yet, and I've no idea how complete their implementation is.

Apologies; I wasn't familiar with the boiler and assumed the worst (the irony of the name as you said!)

The RF OWL monitor we have for electricity is not that accurate or sensitive, and I had better results when I was counting pulses and timing the intervals between flashes on an Ampy meter, so I plan to get a 2nd hand Ampy from ebay fitted by the CU for that purpose.

Makes sense then you can ignore the issue of power factor and can meter exactly what you'll be billed for.

More on topic, with it almost zero outside and I'm still stuck with high return temps unless I briefly activate the zone valve for the bedrooms to draw in some colder water,

Your radiators will only be able to dump a certain amount of heat into the room. There's very little you can do about this other that i) lowering your pump speed (slowing the flow through the rads hence increasing the heat loss whilst it's in there) but this may lead to inadequate flow across the system/boiler, or ii) fitting bigger radiators!

I wonder if boilers could draw from the cold outside air to help the condensing process rather than require the return flow itself to be cold enough (but quite possibly I'm not understanding some crucial part of how that process works).

I think you may have started out on the wrong foot with your thinking here. Whilst the benefits of condensing, i.e. the recovery of latent heat from the exhaust gases, do require a lower (<57C) return temperature those benefits will not be realised through any artificial temperature drop - you wouldn't recover as much heat from the exhaust gases as you'd lose from sucking in cold air.

The only way to maximise the benefits of condensation (within the boiler!) is to operate at an overall lower temperature. Hence, if the flow temperature is lowered from, say, 80C to 70C then there is a corresponding drop in return temperature to below the dew point thus condensing will occur.

However, the lower system temperature means less heat energy hence this would usually necessitate bigger radiators to still satisfy the heat requirements of your house. Furthermore, in colder weather the lower flow temperature may lead to poor performance.

That's why a well-controlled system can dynamically adjust the flow temperature to maximise condensing when possible but also giving consideration to the necessary heat demands as required. Modulation plays a role here - adjusting the heat input (flame) to the demands of the output (heat into the system/house), as do other techniques such as weather compensation (via external temperature measurement).

Condensing isn't the be all and end all, particularly if it is done at the expense of other efficiencies.

Mathew

 

[moggie2002]

I think you may have started out on the wrong foot with your thinking here. Whilst the benefits of condensing, i.e. the recovery of latent heat from the exhaust gases, do require a lower (<57C) return temperature those benefits will not be realised through any artificial temperature drop - you wouldn't recover as much heat from the exhaust gases as you'd lose from sucking in cold air.

I've not seen the innards of such a boiler dissected to formulate more informed thoughts, but I imagined the possibility of a boiler that drew cold air into itself, not into the surrounding building of course, and directed it to the part of the system that needed to be cooled to trigger the condensing process. Otherwise, perhaps a heat exchanger to cool the water return flow could be used in the boiler. Sadly I don't think I have any way to control the pump speed at the moment at least so can't see the effect of that. Almost shutting off the towel rail and a tall rad in the kitchen/living room area, which are the closest to the boiler, reduces the return a bit, but they're then significantly cooler themselves. Basically the drop across the Myson decors isn't sufficient. I do wonder what the average return flow temperature is for folks with condensing boilers.

 

[MJN]

I've not seen the innards of such a boiler dissected to formulate more informed thoughts, but I imagined the possibility of a boiler that drew cold air into itself, not into the surrounding building of course, and directed it to the part of the system that needed to be cooled to trigger the condensing process.

That's just it though - the purpose of condensing the exhaust gases is that recovery of the latent heat is transferred to the water in the return pipe hence recovering some of the heat that otherwise disappear out of the exhaust. Pulling in cold air would completely offset this pre-heat benefit!

Sadly I don't think I have any way to control the pump speed at the moment at least so can't see the effect of that.

There will be a speed control on the side (usually providing three settings).

Almost shutting off the towel rail and a tall rad in the kitchen/living room area, which are the closest to the boiler, reduces the return a bit, but they're then significantly cooler themselves. Basically the drop across the Myson decors isn't sufficient.

It sounds like you need to balance all the radiators so that they have the same temperature drop across them. (See the FAQ for guidance here) Once that is done, and the pump speed is set so as to make the drop across the boiler as close to that suggested in the manual then there's little more you can do.

I do wonder what the average return flow temperature is for folks with condensing boilers.

For a properly configured and balanced system 11C or 20C depending on the design of the boiler.

Mathew

 

[moggie2002]

It sounds like you need to balance all the radiators

Done, but I only achieved a few degrees drop on a rad. I need to check the pump again, but I think it has a variable speed grundfos rather than one with the 3 speed knob. There's always the option of lowering the flow temp a bit as someone else suggested.

The bigger issue and likely biggest win is trying to get the back of the house together so that there's more insulation and less heat loss. The house itself is quite old (1822), and substantially damp for much of the ground floor thanks to flooding in the past and re-pointing incorrectly (i.e. with modern materials). Some internal bricks literally turn into dust when you touch them. The back half is a relatively modern extension, and I wanted that gutted so we could raise the floor/ceiling, replace joists, knock some walls down and re-jig things a bit. Most walls are battened out, and I looked at aerogel as an insulator, but at well over 100 quid for a sheet of plasterboard with something like 6 or 9mm aerogel on the back, didn't go for that though it is impressive stuff. We don't have ceilings up yet in the living room/kitchen area, and despite blocking as many sources of draughts as I can, lack of insulation and inevitably still some draughts means that the CH and UFH are quite busy when it's down to the temperatures of the last few days. For me it generally feels quite warm with modest heating, but it needs much more heating for my SO

 

[MJN]

The bigger issue and likely biggest win is trying to get the back of the house together so that there's more insulation and less heat loss.

Absolutely. Adequate insulation is by far and away the most effective measure you can take with regards to maximising comfort and minimising energy consumption.

For me it generally feels quite warm with modest heating, but it needs much more heating for my SO

Tell me about it... I don't think the technology exists to solve that one I'm afraid...

Mathew