To control my central heating I’ve got a Honeywell “Sundial RF2 Pack 2” which consists of a fancy electronic programmer and radio linked wireless thermostat.

The programmer has an optimisation feature that is supposed to learn how long it takes to heat up the house so for example, as I understand it, if you set the heating to come on at 7am it should come on earlier but at varying times depending how cold it is in order to get the house up to temperature at 7am.

I’ve used it for over a year and as far as I can see all it does is come on precisely an hour before the set time and while it’s on for that first hour it says “optimising” on the display. I think I’ve had it long enough for it to do its learning but it only ever comes on an hour before and this does not vary with the temperature as I think it should.

Anyone got any experience of this model and what it does or what it should do?

On Dec 6, 8:20*pm, Murmansk wrote:
To control my central heating I’ve got a Honeywell “Sundial RF2 Pack 2” which consists of a fancy electronic programmer and radio linked wireless thermostat.

The programmer has an optimisation feature that is supposed to learn how long it takes to heat up the house so for example, as I understand it, if you set the heating to come on at 7am it should come on earlier but at varying times depending how cold it is in order to get the house up to temperature at 7am.

I’ve used it for over a year and as far as I can see all it does is come on precisely an hour before the set time and while it’s on for that first hour it says “optimising” on the display. *I think I’ve had it long enough for it to do its learning but it only ever comes on an hour before and this does not vary with the temperature as I think it should.

Anyone got any experience of this model and what it does or what it should do?

Optimum start was invented long before electronic devices came in but
was expensive then so it was only used for very large commercial
installations. Now it is cheap.
If it is working right it can knock 10%-15% off your fuelbill.

Theoretically, you tell it the occupation times of the building and it
does the rest.
So on a cold day it should start the heating up sooner than on a warm
day.
It may have optimum shut down as well. ie shuts the heating down
before you leave the building.

There used to be manual set up ones but they would be too hard for the
average user to set up ,it was only when they became "self learning"
they came on general release.

So if it is coming on at the same time regardless of weather, it is
not working.

Ba aware that most of them have a maximum length of time for the
"heating up" period and this is the fallback it reverts to if faulty.

Check the various temperature sensors are still connected.
Usually inside, outside and water temperature.

They are often thermistors which can be checked with a multimeter.
Often they will say what resistance they should read for a particular
temperature. Or maybe the handbook will.

It is usually these that go wrong.

I don't think it's quite as sophisticated as the ones you mention!

It doesn't have an outside temperature sensor or a water temp sensor. I think all it does is to learn how fast the house tends to heat up and, potentially, adjust the start time when it's a bit warmer.

BUT, because in my settings I have the thing restricted to starting one hour before the set time I think maybe it's never had a chance to demonstrate a varying start time. I can set it to start up to three hours early so maybe if I allow it to do that, I'll then see some variation in the actual start time based on the general temp of the house.

Murmansk wrote in
:

I don't think it's quite as sophisticated as the ones you mention!

It doesn't have an outside temperature sensor or a water temp sensor.
I think all it does is to learn how fast the house tends to heat up
and, potentially, adjust the start time when it's a bit warmer.

BUT, because in my settings I have the thing restricted to starting
one hour before the set time I think maybe it's never had a chance to
demonstrate a varying start time. I can set it to start up to three
hours early so maybe if I allow it to do that, I'll then see some
variation in the actual start time based on the general temp of the
house.

Some work on an assumption of something like 20 minutes per degree C. I
guess a sophisticated one would learn the time to move 1degree.

On Dec 7, 7:49*pm, Murmansk wrote:
I don't think it's quite as sophisticated as the ones you mention!

It doesn't have an outside temperature sensor or a water temp sensor. I think all it does is to learn how fast the house tends to heat up and, potentially, adjust the start time when it's a bit warmer.

Haven't come across that sort.
Maybe there should be outside sensors but they're missing?
Hard to see how it can work properly based purely on inside
temperature.

On Fri, 7 Dec 2012 23:34:30 -0800 (PST), harry wrote:

It doesn't have an outside temperature sensor or a water temp sensor.
I think all it does is to learn how fast the house tends to heat up
and, potentially, adjust the start time when it's a bit warmer.

Haven't come across that sort.
Maybe there should be outside sensors but they're missing?

Just your sheltered life Harry. Domestic optimum start thermostat/timers
just start x mins early for a y temp below the set point up to some
maximum. They aren't full commercial weather compensated control systems.

Some domestics learn how fast the house will warm up others need to be
told. But the basic target of meeting the desired set point temperature
at the desired time is still there, normally only the first set point
time has the optimum start feature.

--
Cheers
Dave.

Dave Liquorice wrote:

Domestic optimum start thermostat/timers
just start x mins early for a y temp below the set point up to some
maximum. They aren't full commercial weather compensated control systems.

Some domestics learn how fast the house will warm up others need to be
told.

I have two Heatmiser programmable stats (one for C/H and H/w, the other
for electric UFH in one room).

They are the learning variety, which slowly adjust the minutes per
degree figure throughout the year, lacking true weather compensation
they could get caught out by a cold spike, but do fairly well ...

http://adslpipe.co.uk/misc/heatmiser.png

The solid blue lines are the set temperatures throughout the day.

The orange bands are when the stat is calling for heat.

The dotted blue lines with (P) are where it's using optimum start to
reach a set temperature.

The green (H) and (M) are where manual override or hold have been used.

The [T] and [t] are the timed water periods.

Dotted black line is external temperature from nearest met office
station (not used by the stat, just for monitoring).

BUT, because in my settings I have the thing restricted to starting one hour before the set time I think maybe it's never had a chance to demonstrate a varying start time. I can set it to start up to three hours early so maybe if I allow it to do that, I'll then see some variation in the actual start time based on the general temp of the house.

I changed the settings last night to allow it come come on up to 3 hours earlier than the set start time and it came on at 2am despite my start time being 7.30am!! This just does not make sense as I'd have expected it to come on, at worst, at 4.30am!

I think I might give up on this feature and just make it come on an hour before I get up!

Andy Burns wrote:

I have two Heatmiser programmable stats (one for C/H and H/w, the other
for electric UFH in one room).

They are the learning variety, which slowly adjust the minutes per
degree figure throughout the year, lacking true weather compensation
they could get caught out by a cold spike, but do fairly well ...

http://adslpipe.co.uk/misc/heatmiser.png

The solid blue lines are the set temperatures throughout the day.

The orange bands are when the stat is calling for heat.

The dotted blue lines with (P) are where it's using optimum start to
reach a set temperature.

The green (H) and (M) are where manual override or hold have been used.

The [T] and [t] are the timed water periods.

Dotted black line is external temperature from nearest met office
station (not used by the stat, just for monitoring).

That's interesting. Did the charting software come with the
stats, or is it something you have developed?

Chris
--
Chris J Dixon Nottingham UK


Have dancing shoes, will ceilidh.

Chris J Dixon wrote:

Andy Burns wrote:

http://adslpipe.co.uk/misc/heatmiser.png

That's interesting. Did the charting software come with the
stats, or is it something you have developed?

Not written by me, but someone called Alexander Thoukydides, all I did
was bang it about a bit to make it run on centos instead of ubuntu.

http://code.google.com/p/heatmiser-wifi/

In article , Andy
Burns writes
Chris J Dixon wrote:

Andy Burns wrote:

http://adslpipe.co.uk/misc/heatmiser.png

That's interesting. Did the charting software come with the
stats, or is it something you have developed?

Not written by me, but someone called Alexander Thoukydides, all I did
was bang it about a bit to make it run on centos instead of ubuntu.

http://code.google.com/p/heatmiser-wifi/

Thanks for sharing the plot and the code link.

If I've read it correctly your system is comfortably controlling the
temp within one degree and the boiler is firing for about 20mins in a
90min heating/cooling cycle (5th - 6th Dec, approx 3degC otside). It
sort of debunks Honeywell's insistence that a boiler needs to fire 6
times an hour[1] to maintain an accurate room temperature.

It's saved me some time modelling my own system before junking the
proportional Honeywell stats and swapping to more simple on-off control.

[1] Honeywell's default firing frequency for their programmable stats,
adjustable down to 3 times per hour minimum (my chosen setting but I
wish it could be lower).
--
fred
it's a ba-na-na . . . .

fred wrote:

Andy Burns wrote:

http://adslpipe.co.uk/misc/heatmiser.png
http://code.google.com/p/heatmiser-wifi

Thanks for sharing the plot and the code link.

If I've read it correctly your system is comfortably controlling the
temp within one degree

There /is/ an option to control to within 0.5 degree, but I consider
that unnecessary, also I suppose firing more frequently for shorter
periods would be less efficient for my cast-iron lump.

and the boiler is firing for about 20mins in a
90min heating/cooling cycle (5th - 6th Dec, approx 3degC otside).

Yes, though obviously the length of the cycle varies depending on
temperature/wind, last Friday when it was below freezing all evening,
the cycle was 20mins out of every 60mins, rather than out of every 90mins.

In article , Andy
Burns writes
fred wrote:

Andy Burns wrote:

http://adslpipe.co.uk/misc/heatmiser.png
http://code.google.com/p/heatmiser-wifi

Thanks for sharing the plot and the code link.

If I've read it correctly your system is comfortably controlling the
temp within one degree

There /is/ an option to control to within 0.5 degree, but I consider
that unnecessary, also I suppose firing more frequently for shorter
periods would be less efficient for my cast-iron lump.

and the boiler is firing for about 20mins in a
90min heating/cooling cycle (5th - 6th Dec, approx 3degC otside).

Yes, though obviously the length of the cycle varies depending on
temperature/wind, last Friday when it was below freezing all evening,
the cycle was 20mins out of every 60mins, rather than out of every 90mins.

Yes, spotted that and again the 60min cycle at colder temps is useful.
I'm trying to enhance my multizone setup and was wondering if I can get
away with polling my 'slave' zones at 30min intervals (to keep demands
in sync) and it looks like I can. I'll be able to tweak down to half
degree accuracy too if needed.
--
fred
it's a ba-na-na . . . .

On Dec 8, 7:22*pm, Andy Burns wrote:
fred wrote:
Andy Burns wrote:

http://adslpipe.co.uk/misc/heatmiser.png
http://code.google.com/p/heatmiser-wifi

Thanks for sharing the plot and the code link.

If I've read it correctly your system is comfortably controlling the
temp within one degree

There /is/ an option to control to within 0.5 degree, but I consider
that unnecessary, also I suppose firing more frequently for shorter
periods would be less efficient for my cast-iron lump.

and the boiler is firing for about 20mins in a
90min heating/cooling cycle (5th - 6th Dec, approx 3degC otside).

Yes, though obviously the length of the cycle varies depending on
temperature/wind, last Friday when it was below freezing all evening,
the cycle was 20mins out of every 60mins, rather than out of every 90mins..

Doesn't that point to your boiler being grossly oversized?
Frequent firing leads to gross inefficiency in any boiler.
Ideally it should run constantly.

On Saturday, December 8, 2012 8:35:35 AM UTC, Dave Liquorice wrote:

Just your sheltered life Harry. Domestic optimum start thermostat/timers

just start x mins early for a y temp below the set point up to some

maximum. They aren't full commercial weather compensated control systems.


That's not optimum start then, but something else that the vendors have misrepresented
with some techy sounding buzz words that the purchasers don't understand..

On Sun, 9 Dec 2012 01:39:55 -0800 (PST), Onetap wrote:

Domestic optimum start thermostat/timers just start x mins early for a
y temp below the set point up to some maximum. They aren't full
commercial weather compensated control systems.

That's not optimum start then,

Whats your definition of "optimum start" then please?

--
Cheers
Dave.