Thermostat Central Heating

[pwdrive]

OK this is for any electroplumbers out there, I designed and fitted the central heating system in the house we now live in, it's wood powered.
The boiler in the woodstove is a cross flow ie: two diagonal outlets (flow and return) for the primary water heating (28mm) gravity flow to and from the HW tank (Indirectly heated ie: coil inside)
The other two diagonal outlets feed (flow and return) 8 radiators via CH pump.
I designed the system so that the radiators only come on when the HW tank has reached a mean temp of 60 degrees C, the CH pump then kicks in ie: priority hotwater.
I have a minimum temperature thermostat fitted to the CH return pipe within 6" of the boiler, it is set so that the CH pump switches off if the water temp in the stove boiler drops below a certain low level, this is to help reduce the possibility of condensation on the fire box side of the boiler if the wood contains any significant moisture.
There is a high set temp. thermostat within 6" of the boiler on the HW gravity flow pipe, this is set to about 85-90 degrees C so that if the gravity flow has not got the HW tank to its mean 60 degrees then the CH pump will kick in before the boiler starts gassing or boiling the resulting CH flow will then reduce the water temp in the boiler to a lower level ie: dump the excsesive heat.
The system is fitted with the correct size expansion pipes etc and a permenant heatsink radiator in parrallel with the HW flow and return to/from the HW tank.
The problem is the last mentioned set high temp thermostat, it is a standard bi metal Honeywell job which is retained on the pipe by its own built in springs.
The problem being that if the fire is started from cold with dry wood then the HW flow pipe beween the boiler and the HW tank can attain a high temperature very quickly, the pipe can get to almost boiling before full gravity circulation takes place, this is partly my design problem as we are limited to how high the HW tank can be physically located above the boiler and there is nothing I can do about that.
So as I was saying the flow pipe can attain near boiling temp before full gravity flow takes place, if this temp is attained quickly then the water in the boiler will start to "gass" and almost boil before the high limit temp stat kicks in, this is due to the relatively slow thermal conduction through the stats interface with the HW flow pipe and then up on through the stat to the bi metal strip.
If the fire is built up slowly from scratch then this rapid heating of the HW flow pipe is not a problem.
So in all this I need a quicker reacting (temp. transfer wise) stat that is not going to cost a dissproportionate amount of money.
Any one got any ideas?
The stat has only got to switch the current flowing through the CH pump, the pump is a standard Grundfos CH pump set to low flow (position 1) and passes less than an amp as far as I am aware.

Lawrence.

[AlanBeckett]

Lawence,
Your system sounds just like the one we had at our last house when we had a Solid Fuel Rayburn - Nouvelle I think, although it was 10 or so years ago. The gravity DHW pipes - 28mm - travelled up the wall behind the boiler and then along a long landing to the tank, with the bathroom radiator (heat leak) tapped off on the way. The Rayburn only had two tappings so the Radiator return was pumped and connected into the DHW pipe via an Injector T. Just like your system there were High and Low Stats, although I didn't bother with a tank Stat. I don't know who made the Stats except to say that the bodies were metal, they fitted onto the pipes with springs and they worked fine. Of course it's possible that the heat capacity of the Rayburn was much higher than your boiler and hence no matter how hard I tried it only heated up relatively slowly, but I tried pretty hard at times.
So, two questions:-
Does the High Stat overide the Tank Stat and run the pump?
Have to put an AVO across the High Stat to make sure it closes?
Good luck.
Alan

[pwdrive]

Hello Alan thanks for the reply, sounds like we are talking the same lingo, I put in a similar system as you described at our last place when we first bought it, it was a Stanley wood range with just the two boiler tappings for HW, I plumbed in a couple of radiators but no one would sell me an injector tee so I made my own with a short reducer set inside a standard T piece.

In the system described in my previous post the tank stat is about two thirds down from the top of the HW tank set to run the CH pump at its setting of 55 C this in practice gives about 60-65 C at the top and approx 50 C at the bottom of the HW tank.
This is overidden when the set high temp stat on the HW flow pipe at the boiler kicks in, this stat brings in the CH pump to dump the excess heat, it is also in parallel with a manual overide switch on the chimney breast which I fitted in case of total stat failure (just a little saftey feature).
The set high temp stat contacts do close but here is a thermal delay beween the temperature of the pipe it is
fitted to and the action of the switch snapping on, I have tried increasing the tension of the stats retaining springs and also the application of a heat transfer compound.
The effect I have is similar to the action of some electric kettles insofar that the kettle sometimes boils away freely before the thermal cut out switch comes into play.
The problem behind all this is that I suspect that the HW tank is to low in relation to the boiler to enact a fast gravity flow when the the fire is initally started using dry wood, there is nothing I can do about this hieght by virtue of the layout of the house and some of it's existing support walls.
The distance from the boiler tappings to the HW tank tappings is about 1 meter, the rise/fall over that meter distance is about 160 to 170 mm the only 90 degree bends are at a distance barely an inch from the HW tank tappings and as such I do not think that they are adversly impeding any gravity flow, all primary plumbing is 28 mm with no reducers.
So I guess I am needing a fast (thermally) acting stat also with these standard Honwell style stats they have quite a high hysterysis characteristic or at least it seems so to me.
The stove itself is 9Kw into the room and 9 Kw into the CH ie: the boiler has a rated output of 9Kw however many btu's that may be.
The system works very well as I have rebuilt the house to twice thermal insulation building regulations in overall thermal loss through the building.
It's just that one little problem to resolve, all primary heating circuits and radiators have the correct inhibitors added to them in the correct dose etc.

Lawrence.

[AlanBeckett]

Lawence,
Two (not very helpful) comments:-
The only Stats that I can find on a quick sweep have plastic bodies, although I would hope they're metal at the back - contact area. I tried using one of those in a greenhouse Propagator and found it was hopelessly slow.
I'll duck after I type this one - which way 'up' have you connected the cylinder coil with reference to the boiler?

A mate is dragging me off to the pub so I'll be fairly safe

Alan

[pwdrive]

Hello Alan,
It's connected the right way (I think!!) top to top bottom to bottom The HW tank is fitted with a gravity flow copper coil (28 mm) and not a pumped 22 mm coil as in oil fired honeywell "Y" plan. all the plumbing pipe stats I have seen on the net are the cheapo type I might have to knock up an electronic one and solder a sleave to take an electronic temp sensing element directly on the the HW flow pipe where the present thermostat is fitted.
As it stands now nothing really adverse happens, I tried to push the whole system to its limits by flash firing up the stove from cold with very dry wood and a good combustion air intake ie: worst case scenario, as was expected the top of the boiler started "singing" almost boiling before the stat previously described kicked in, there was a three minuit delay even though the pipe it was attatched to was well above the 85 degree C temp that the stat was set to, most of the gassing bubbled off by virtue of the physical upward gradient of the pipes and then up the expansion pipe to its overflow into the header tank.
The main reason for wanting to sort the stat issue out is that internal gassing of the boiler and external condensation on the boiler is the main reason for the eventuall demize of wood/solid fuel boilers. The external boiler condensation effect is not at issue by virtue of of the set low temp stat on the CH return pipe where it enters the boiler, that set up works fine as I have offset the temp setting by a few degrees to compensate for the reasonably large hysterysis effect in that type of stat.
The set high temp stat needs to snap in (real time, little delay) when the actuial pipe it is referenced to reaches the actual temp that the stat is supposed to operate at, to that extent I could offsett the temp setting on that stat to a lower setting to compensate for the thermal transfer delay that exists, but that tends to lead to it cutting in more frequently thus in effect it is self defeating to a certain extent in increasing the gravity flow beween the flow and return pipes differential temperatures if that makes sense.
I have seen in the past temperature sensors of the electronic type (thermocouple maybe ?) where the probe is placed on a say a power transistor heat sink and the reading is almost instantainious and direct. I think this is the kind of thing/system I will eventually need by whatever means.

Lawrence.

[AlanBeckett]

Lawrence,
I don't know what else to add, except that my HW tank was about 3m (10') above the boiler, and of course that your boiler's thermal capacity is very likely a lot lower than mine was.
I can only suggest that you make a fast acting Stat.
Good luck.
Alan

[pwdrive]

Thanks Alan, I have come to the same conclusion, I think the problem is solvable though, I will have to have a hunt around.

Lawrence.

[Michaelr]

Thanks Alan, I have come to the same conclusion, I think the problem is solvable though, I will have to have a hunt around.

Lawrence.

Hi Lawrence,

I have used this device a few times and they are very good. You could design your own controller, easy peasy if you can program a pic, failing that if you read the apps note there is an example circuit for a heater. Replace that with a relay and you have a working controller.

Mike

[Mark Hennessy]

Hi,

I've enjoyed reading through this, not least because I've just replaced the central heating boiler in this house. Mine is a completely conventional gas fired system, so not very exciting in the context of this conversation, but reading this makes me wonder: why use gravity circulation at all? I'm aware this is the way things used to be done, but pumping seems universal now. Why not convert it to fully pumped? Forgive my ignorance

For the thermostat, have you considered items like this: http://uk.farnell.com/microtherm/05en103.../dp/732540 (part number 732540 if the link fails). These are available in NO or NC, and self-resetting or manual resetting versions, and in a range of temperatures. Obviously not adjustable, but might do the job. Should be easy enough to attach to a copper pipe.

All the best,

Mark

[pwdrive]

Thanks for that Mike, that looks like the kind of thing I need, probably for me connected to a relay for final CH pump switching, I am assuming that the metal cased version would be best as this would make it relatively easy to thermally interface direct to the HW pipe.

Lawrence