What am I missing???

Forum about Opentherm, Hvac, House heating, Boilers

Moderator: hvxl

Post Reply
sdspieg
Starting Member
Starting Member
Posts: 3
Joined: Fri Nov 02, 2012 3:03 am

What am I missing???

Post by sdspieg »

I am starting to look into this issue and am kind of puzzled by the products that are currently available.
  • why are so many products with z-wave instead of regular wifi? Don't most of us already have wifi in our homes and would that not be by far better than NON-internet based protocols? Is that just a power issue? But even if so, can't z-wave also be controlled directly through tcp/ip?
  • why do we actually NEED a thermostat? If we could just put some wireless radiator thermostats (like the Danfoss living connect ones, but then NOT based on a proprietary protocol) on our radiators that could each be controlled by some open-source software - wouldn't that essentially be enough? We could then control every individual radiator through our smartphones, tablets, pcs, etc. and turn them up and down according to some scheme we could customize as we want. And no link with the HVAC would even be required
It seems to me that this need not be very expensive. I see on Alibaba that a thermostat heat sensor is about 1USD a piece; a wifi module is maybe 5USD (and I guess the others are even cheaper). I guess you'd need some valve actuator, but how expensive can that be (again on alibaba: 55usd!)? So let's say somebody could put all of this together as open source hardware for about 50$ a piece (per radiator). And then think of sthg expandable like the Node on Kickstarter and all the sensors THAT one carries - of course with Open API, Open Firmware, Open Source Code AND modular!
Wouldn't we all want to by that? We'd recoup that within a year on our electricity bills. Not to mention the overall benefits to society in energy and pollution.
So can anybody tell me what I'm missing here?!?!?
AshaiRey
Senior Member
Senior Member
Posts: 1310
Joined: Mon Feb 02, 2009 5:27 pm
Location: Netherlands
Contact:

Re: What am I missing???

Post by AshaiRey »

You are missing the CV (heater) component here.
In your case the heater must be always on (expensive) and you need some other modifications to the pipes near your heater to get it working like that.
Bram
WooTz
Starting Member
Starting Member
Posts: 31
Joined: Tue Jun 28, 2011 12:19 am

Re: What am I missing???

Post by WooTz »

a wifi chip will be more expensive, you will also need a controller, valve actuator, prototyping, designing a plastic case.....
sdspieg
Starting Member
Starting Member
Posts: 3
Joined: Fri Nov 02, 2012 3:03 am

Re: What am I missing???

Post by sdspieg »

Bedankt Bram! Still - and please indulge my ignorance - isn't a thermostat just a sort of control system that sends signals to the CV (heater) to EITHER turn it off and on OR to modulate it? I understand that all of these signals used to be transmitted in a 'wired' way AND that this also meant that people wanted to have a dedicated control device somewhere centrally in the house to 'sense' the temperature there and to then send those signals via wires to the heater instead of having to run to the CV/HVAC every time they want to change the temperature. But if you can have distributed control systems at the radiators themselves that can talk to the 'CV-ketels'/HVAC-systems - why still have a dedicated device somewhere centrally in the house? Aren't there any 'CV-ketels'/HVAC-systems that can do this in a fully digital/wireless way? And so whereby we would then NOT need a dedicated box to send the signals, but whereby those signals could be sent from the radiator sensors/actuators to the control system IN (or by) the heater and controlled by some app on your smart-phone/tablet/pc?

The reason I also ask is that it seems like a more energy-efficient, automated system right now would still set me back about €1000 (I'm talking about things like Danfoss link system, Honeywell evohome, etc). And the most expensive piece of hardware seems to be precisely those 'control boxes' that seem mostly pretty awful-looking (maybe with the exception of the Nest) AND 'closed' (proprietary etc.). To ME, all of those things look so '90s'. And we are in the 'teens' already! in car navigation terms - they want to sell me clunky TomTom boxes, whereas I just want to have Google Maps on my gsm!

I also looked on Kickstarter and couldn't find any real new 'open source' work going on in this area. But so an open-source soft- AND hardware solution, decentralized wireless sensors/actuators on the radiators, and some interface with the HVAC - all of this just over tcp/ip (whether immediately, or, if that is too energy-hungry, through the HVAC-interface). It seems to me a solution like that should be perfectly feasible for a few hundreds of euros (dollars/pounds), which a LOT of us would be able to recoup within one year. And again - think of the societal benefits in TOP of the individual benefits!

-Stephan
Digit
Global Moderator
Global Moderator
Posts: 3388
Joined: Sat Mar 25, 2006 10:23 am
Location: Netherlands
Contact:

Re: What am I missing???

Post by Digit »

You think you can save hundreds of euros per year, by being more efficient? Wow.
Good luck and I'd like to follow your project; please keep me posted on the progress, cause I think that's not possible.
AshaiRey
Senior Member
Senior Member
Posts: 1310
Joined: Mon Feb 02, 2009 5:27 pm
Location: Netherlands
Contact:

Re: What am I missing???

Post by AshaiRey »

Hi Stephan,

For me domotica means that is mainly invisible. It should work. Like for heating, my wife needs to feel comfortable with it and a nice looking box won't do that but a good functioning system does. You Set the system up and don't want to look at it any more. So, an ugly or a nice looking box, both are put away out of sight. That's how i think of it.

About the thermostats.
There thermostats with and without electronics and if they do have electronics they can send information back or they only receive control information and work independent with that infomation. As i read you post i see you are looking at the most complicated setup. You will need thermostats that shuts the radiotor off when the temperature is reached and signal it back to a control box. When the temperature gets to low is must go open again and signal the box also. The box needs to contain knowledge of all the thermostats in yous system and have logic in it to know that it can signal the heater to turn on or off. If one thermostat is going off while another is still on then the heater must stay on. However the heater can be turned to a lower capacity and reducing gas usages that way. You see that the complexity doubles with each thermostat that will be added to the system. As a bonus it will triple if you want to use modulation on the heater because that's based on the curve where the present temperature is between upper and lower temperature of that thermostat. So you need to take all curves into account to get the right amount of modulation on the heater.

Having said that then i think 1000,- for a good working, hassle free and energy reducing system is a fair price since when you doing this DIY you will surely be higher of and have lots of fustration.. But keep us informed about your progress. I lijke to learn from it too.
Bram
WooTz
Starting Member
Starting Member
Posts: 31
Joined: Tue Jun 28, 2011 12:19 am

Re: What am I missing???

Post by WooTz »

it can be easier. you need 1 main thermostat to control the boiler on off. each room a thermostat (pi) controller to control the valves. if any of valves need flow (heat) turn on main thermostat. no need to keep track which one requires. alo wifi will be too expensive, a rf12 or c1100 will do the job fine. but again i doubt you can make it cheaper. but you will learn from it :-)
sdspieg
Starting Member
Starting Member
Posts: 3
Joined: Fri Nov 02, 2012 3:03 am

Re: What am I missing???

Post by sdspieg »

Thanks for the feedback.
WooTz - why would prevent any of the radiator thermostats to send an on/off signal to the boiler? I.e. if ANY one of the thermostats senses that it's colder than the program, it sends the 'on' signal. If none do, the boiler is off. If more than one send signal, the later ones are just ignored (as the boiler is already on). And same thing to turn off the boiler - here the last 'off' would be the final one, all other ones would be ignored as long as one thermostat is still on 'on'.
And guys - my idea here is not so much to do this myself. My idea is rather to get the open source hardware community interested in this. I contacted George Yu, the guy who initiated the Node project on kickstarter (http://www.kickstarter.com/projects/108 ... of-sensors). Basically what these guy do is to start an OPEN discussion about the product (like we are starting here), get up-front funding from real people (like us), then prototype sthg, and then start scaling production based on the actual popularity. For anybody interested in this, I highly recommend Chris Anderson's (formerly of The Economist, now Chief Editor of Wired) "Makers: The New Industrial Revolution".
As to cost - I really have no idea and you may all be right. I was just surprised when I looked on alibaba what the components actually cost. Now innovation (also by companies) should of course be rewarded, but if innovation is going SOOOO slowly as in this area, it seems to me that this new open source hardware development might provide the private sector with a very useful kick in the rear end!!!
bus147
Starting Member
Starting Member
Posts: 29
Joined: Mon Jan 23, 2012 6:11 pm

Re: What am I missing???

Post by bus147 »

@sdspieg
Before anybody starts commenting about the HoneyWell EvoHome system, I'm aware off its existence but I excluded it beause of its proprietary protocol and closed nature. Topic starter wants such a system but with full control and open by design.

I think it's a great idea and I allways like crazy and wild plans. But lets step back and look at a domestic heating system for a while. Several factors must be taken into account before deciding on a heating control strategy. First the isolation level and thermal properties of the house. Second the way of heating. Third the layout. And lastly the way the house is being used.
-1- At one end of the isolation spectrum you'll find the passive houses. Upto 4 layers of glass, 20cm of isolation etc. These buildings hardly need any heating and regulations and energy prices will result in more of these buildings. They don't need such a system, the only use I see for such control is when cooling is taken into account. But that complicates things even further.
On the other end are not isolated buildings. Some of them are very difficult to isolate, for example all kinds of monumental buildings. That market is rather small as well. But as an owner of one these cash drains I do think such control system has it's uses.
-2- Nowadays more and more houses have floor and even wall heating as primary heaters. Only some radiators or convectors below large windows. Such heating system has a long reaction time and are more suited to leave at constant temperature. Houses with a low thermal mass are more suited for such desired control. More thermal mass means longer reaction time don't allow for the much differentiation (within reasonable timespans).
-3- Single level houses allow better thermal segmentation than multi-level houses. Warm air always rises. It's therefore more difficult for example to keep cool bedrooms upstairs and a warm living room downstairs with an open stair case.
-4- You're looking a thermal segmentation, if you can keep certain rooms cooler you'll save energy by not heating them. That's my understanding off your idea. Specific rooms should be at a specific temperature at a specific moment. Why heat a bedroom when nobody is there? (Remember the isolation? If you don't lose energy it doesn't cost anything to keep the room at a certain temperature) This won't work if you don't have doors, in a stair case for example. Or if kids leave doors open all the time. How great your control system is, with such disruptive forces in the house it will never work as good as it in potential can.

Are there other ways of efficiently heating your house and only heat what needs to be heated? Sure many. How about this rather conventional approach:
-fit every radiator with a thermostatic valve
-ensure that the entire system is hydraulic balanced
-install a modern energy efficient pump
-apply the above to the floor heating a well if present.
-buy a modern thermostat and depending on the amount off isolation also an outdoor temperature sensor. (better isolation = no sensor)
-place the thermostat in the warmest room and least isolated! set the thermostat to a nice level.
-adjust the thermostatic valves per room AND KEEP the doors shut!!!

The thermostat will signal the boiler that heat is required. All rooms start heating up. One after another the rooms reach their desired temperature. The thermostatic valves will close and more warm water returns. The boiler (not the thermostat) will lower it's heat output because it senses that heat is no longer being transferred by the radiators because the valves are closed. When the warmest room also reaches it's setpoint the thermostat switches the boiler off. But during this last phase the boiler was delivering a reduced heat output. One should not underestimate the sophistication off modern boilers and thermostats. They are very efficient but only when certain criteria are met.
All off the above off course wont work if the inhabitants leave the doors open and forget to set back the thermostatic valve. For example because teenage daughter think it's her birthright to walk a summer outfit 365 days a year.
For those situations (remote) controllable valves will help. Don't get me wrong I could use a few very well.

Lets look at those future valves a bit more carefully. They need to be reliable. You don't want to leave your house in the winter and upon return discover frozen radiators because the nice gizmo valve executed a 10 goto 10, or the batteries are empty or didn't receive a signal anymore because the neighbor got an RF quadcopter for christmas and jams your thermostatic valves. It might sound far fetched but it isn't. I know two hvac companies which refuse to install RF thermostats. All the interference and empty batteries were costing them so much more than the time saved by screwing one to the wall and not having to run any cable. So the only way to get this reliable is wired, in my opinion. And that's the last thing anybody wants nowadays. Everything should be wireless, the WAF is much lower if it's wired. Perhaps a few diehards here running KNX buses wont object a few more cables but not the general public.

If you do want some domotica and try to safe some energy let me suggest to following project AFTER YOUVE TAKEN CARE OF THE HEATING SYSTEM TUNE-UP described above.
For each room buy a JeeNode. Add a room node, a sensor module reporting temperature and humidity, light and wether somebody is present if you install the power hungry pir.
Next buy two DS18B20 one wire sensors per radiator. You can daisy chain them so only a single thin cable is needed. Fix them preferable just behind the valve and on the outlet. Use care when installing them to minimize errors. Dont install on plastic piping. Only clean metal surfaces and some thermal grease. ISOLATE the sensors!!! The tiniest draft is already recorded. Only some cheap isolation tube around the sensor suffices.
What can you do with all this?
If programmed correctly the JeeNodes can run for months on a couple a AA batteries.
You'll very accurate metric about each room and radiator.
Yes...
With the data you can monitor the performance of each radiator. Ideally each radiator should dissipate that much that the return water is about 20 degrees lower than the inlet. If too much water is flowing through not all energy will be dissipated by the radiator, this will show up as too small delta T's.
You can be alerted that the room temperature is above a threshold because somebody left a valve open. Or better alert your kids that they should close the valves before going to school.

Lots of uses. Not too expensive. No damage if they system fails. Can be removed when you sell the house.

My 2cents:
Would I invest in such a project? Probably not. The development costs are enormous, certification, plastic and electronics are cheap metal is not. The market probably too small. HVAC installers are very conservative.

Still an affordable,OPEN protocol, remote controllable valve... yes I would buy some.


references:
digits's blog http://blog.hekkers.net/tag/hvac/
JeeLabs http://jeelabs.org/
PDF met stappenplan voor waterzijdig inregelen verwarming.danfoss.com/PCMPDF/Inregelgi ... uik_LR.pdf
klimaatregeling.danfoss.nl/Content/4cae ... fcb5b.html
Commodore White
Starting Member
Starting Member
Posts: 14
Joined: Sun Apr 21, 2013 9:05 pm

Re: What am I missing???

Post by Commodore White »

Absolutely fascinating! Reading your post has been extremely useful. I am just about to place an order for convector rads for the house I'm building. It should be cheap to heat since it has some 200mm of insulation in walls and ceiling. Downstairs is UF heating. As you say, heat rises so my concern was to keep bedrooms cool. Nevertheless, wherever I go for information on how to size the radiators for upstairs, I'm told to ignore thermal efficiency since, if I get rads that are over spec's, they'll simply heat the room quicker before they shutdown. Bigger rads cost more and take up space - best avoided. Already I'm thinking that cooling is going to be a bigger problem than heating! Should I put lots of Insulation in the ceiling above the downstairs rooms/floors to the upstairs rooms to generate thermal isolation.

My plan at present is to have electrically operated valves on the rads controlled by my RasPi. The valves will be of the 240v type switched by a spare socket on homeeasy wall sockets - I bought a few when they were being discontinued at the local DIY store. Temperature can be measured using free standing wireless thermostats. I've bought a fistful of ds1820s to measure temperature in each room but my problem was where to put the sensors - obviously not too near the rads.

I like your idea of measuring temperature on the input and output side of each radiator using Ds1820s but was confused by you comments about the differential temperature. Clearly during the summer, the heating will be off so the differential will be zero. Are you suggesting that if heating is required a difference of more than 20deg will indicate that the rad is turned up too high. Presumably, the temp sensor is on the radiator side of the thermostatically controlled valve.

In truth, I'm beginning to think that given the amount of insulation, trying to manage heat input into upstairs bedrooms to save money is probably wasted effort and will result is very low return on my investment both financially and intellectually.

Regards, and thanks for bringing me down to earth.

Peter
Post Reply

Return to “Hvac, Heating Opentherm Forum”