Energy Efficient Wall Mounted Electric Radiators

[ericmark]

There was another post explaining how the charge points have a link so the DNO can control when power is taken. Seemed a little odd to me, as large cars will need 12 hours or more to recharge so not much room to select when charged.

However question as about room heaters, and to use any storage heater, one has to look at size available to install the units, be it bricks or water, they are big, and other than built with the house, bricks are hard to control. And also how much the home naturally stores matters, the last house set central heating to switch off 10 pm the house being set at 20°C at 6 am when scheduled to turn on again, rarely below 17°C but same with house before that and could cool to 12°C and did not switch off but used programmable thermostat to prevent getting too cold at night. This house also seems to store the heat.

So with a home which naturally stores the heat storage may work, but does depend on the house.

 

[Harry Bloomfield]

However, my point was related to Owain's earlier comment, in that even a 'proportional' control system is to some extent at the mercy of the the characteristics of the temperature sensor (an issue which does not really arise with things like AGC/AVC).

Temperature sensing is so easy and accurate to a fraction of a degree - a ZN423 IC (I think that's the correct number) can do that, without any correction and costs just pennies. Just three pins, +, - and temperature out.

 

[Harry Bloomfield]

However question as about room heaters, and to use any storage heater, one has to look at size available to install the units, be it bricks or water, they are big, and other than built with the house, bricks are hard to control.

I have lived with storage heaters and I have put them in. I was never impressed with them - usually they give out most of their heat when it is probably not needed, mid morning with no one home, then there is no heat left for the evening when the occupants get home from work. They worked best in offices and workshops, when heat was needed in the day.

 

[JohnW2]

Temperature sensing is so easy and accurate to a fraction of a degree - a ZN423 IC (I think that's the correct number) can do that, without any correction and costs just pennies. Just three pins, +, - and temperature out.

Apologies - my typing fingers disobeyed my brain again ... when they typed "...characteristics of the temperature sensor...", they had been told to type "...characteristics of the temperature sensing...".

Yes, electronic temperature sensors are more than accurate enough for the purpose, but sensing 'the temperature of a room' is a much more complicated matter.

Kind Regards, John

 

[Harry Bloomfield]

Yes, electronic temperature sensors are more than accurate enough for the purpose, but sensing 'the temperature of a room' is a much more complicated matter

That part can only ever be at best compromise. In a previous life (1980's) I once designed a fancy cpu based heating management system, using several ZN423 (if that was the correct number). The beauty of it, was how accurate and adaptable it was. Multiple of them could be used, to cover a large heated hall, using them as a lowest temperature sensed or as an average, all very accurate without any need for calibration. It was way ahead of anything commercially available at the time, I designed it so you programmed your desired temperatures for each part of the day, rather than CH / HW on/off, around 8x periods per day. I have only just got that facility with my latest commercial unit, with a day temperature and a night set back temperature setting.

 

[JohnW2]

That part can only ever be at best compromise.

Exactly my point. It's all very well being able to accurately measure the temperature of the air at one point within a room, but attempting to measure 'the temperature of a room' can, as you say, never be more than a compromise, as illustrated by ...

... I once designed a fancy cpu based heating management system, using several ZN423 (if that was the correct number). The beauty of it, was how accurate and adaptable it was. Multiple of them could be used, to cover a large heated hall, using them as a lowest temperature sensed or as an average, all very accurate without any need for calibration...

If the 'compromise' you use is 'the lowest temp sensed', then much of the room will be 'too hot'. If the compromise is some sort of average from all the sensors, then some parts of the room will end up too hot, and others too cold.

It goes without saying that unless one has an impractically complex heating system, it is impossible to achieve exactly the temperature one wants in multiple parts of a room (particularly a large one), no matter how many, or how accurate, sensors one has.

Kind Regards, John

 

[ericmark]

Yes, electronic temperature sensors are more than accurate enough for the purpose, but sensing 'the temperature of a room' is a much more complicated matter.

Kind Regards, John

I had a problem in last house, living room would over heat, I ended up with a rang of devices that would report room temperature, from the weather station to the pet camera including the TRV head, the temperatures reported varied that much I thought they must be faulty, however putting them together every one showed same temperature. I was the way the sun hit the bay window that caused all the problems, out of direct sun light I have recorded 8°C difference throughout the room, tried a fan to move the air, but this made it worse not better as then no cool areas in the room.

I had read the leaflets with each thermostat, where they say mount on wall opposite to radiator, however being an outside wall and having a chimney in it there seemed to be a problem and it did not seem to relate to room temperature, until it failed the wireless table mounted thermostat was an eye opener, the best place turned out to be on a tea trolley next to radiator, had to set it slightly high to compensate for being so close, but that was the best position. It was after this that I realised the TRV at side of radiator is not such a bad location specially when compensated for radiator being on by measuring water temperature.

On dull days the TRV worked A1, when set to 20°C it was 20°C. This was all at my mothers house, which turned out to be a nightmare to control temperature, however before being forced to go and live with her my house never even thought about the heating once TRV's were fitted upstairs.

Downstairs was open plan, and the Myson fan assisted radiator ensured no hot or cool spots, it just worked, so never considered how well it worked until forced to move out to look after mother.

So unlike central heating where the TRV gradually opens and closes giving a varied output, the electric panel heater is either on or off, the old 4 bar electric fire at least had 4 levels of heat, but lacked thermostatic control, in the caravan I had a fan heater modified to give out 500W or 1000W and plugged into a socket controlled by a wall thermostat, that worked well as I could select output and dropping to 500W the run time (mark) was longer so less hysteresis.

So likely two small panel heaters will work better than one big one, as one heater running 85% of the time will work better than one running 43% of the time, so second heater used to boost when first arriving home. But having one which uses solid state relays so it can switch off/on many times an hour without burning out contacts would be best option, but finding one is another story, since always used gas or oil, working out best panel heater is not some thing I got into, the oil filled radiator is only used if central heating fails, often not even then and in the main central heating fails because of no electric.

But with the air temperature being able to vary so much in a room, where to place heater and thermostat is any ones guess, I put oil filled radiator when used in centre of the room.

 

[SimonH2]

With electronic control it's trivial to run an electric panel at part load. Want 50% output, turn it one for one full cycle out of 2; want 10%, turn it on for one full cycle in 10; and so on. With a solid state zero crossing switch it would be silent both in terms of hearing it (no clicks from a relay) and electrical noise (zero crossing switching avoids any spikes).
But as pointed out, any temperature sensing is going to be something of a compromise in terms of where it's located and what it actually senses

 

[Harry Bloomfield]

With electronic control it's trivial to run an electric panel at part load. Want 50% output, turn it one for one full cycle out of 2; want 10%, turn it on for one full cycle in 10; and so on. With a solid state zero crossing switch it would be silent both in terms of hearing it (no clicks from a relay) and electrical noise (zero crossing switching avoids any spikes).

Exactly, the big surprise is that such a system is not used more, with it being so cheap to implement.

You just compare the size of the temperature error with an op-amp, then use the output to set the number of cycles it is turned on for.

I was delighted a few years ago, on buying a new toaster to find it was electronically timed, rather than mechanical bi-metal. Near perfect toast every time, rather than just the first session - it said nothing in the spec. or on the box. The only way to tell, was that it would not latch down unless it had power.

 

[JohnW2]

I was delighted a few years ago, on buying a new toaster to find it was electronically timed, rather than mechanical bi-metal. Near perfect toast every time, rather than just the first session - it said nothing in the spec. or on the box. The only way to tell, was that it would not latch down unless it had power.

My (cheap and pretty ancient) toaster will not latch down without power but, although I can't say I've ever looked too closely, I was not aware of it having any electronics. I had suspected/presumed that the latching was done by a solenoid which, itself, was controlled via some pretty crude sensor (such as a bi-metallic strip). One day (probably when it eventually dies!) I'll take it apart and have a proper look!

Kind Regards, John

 

[EFLImpudence]

Probably like a kettle - just an electro-magnet which when the timer/temperature sensor (bi-metal?) switches off the power it pops up.

 

[JohnW2]

Probably like a kettle - just an electro-magnet which when the timer/temperature sensor (bi-metal?) switches off the power it pops up.

That's what I said, other than that I called it a 'solenoid', rather than an 'electro-magnet'

Kind Regards, John

 

[winston1]

My (cheap and pretty ancient) toaster will not latch down without power but, although I can't say I've ever looked too closely, I was not aware of it having any electronics. I had suspected/presumed that the latching was done by a solenoid which, itself, was controlled via some pretty crude sensor (such as a bi-metallic strip). One day (probably when it eventually dies!) I'll take it apart and have a proper look!

Kind Regards, John

I took apart a scrap modern toaster a while back. It had a 555 timer chip inside it.

 

[Harry Bloomfield]

I took apart a scrap modern toaster a while back. It had a 555 timer chip inside it.

That would make sense. When I last looked at replacement toasters and fed up with the bi-metal trips being so variable - I looked around to see if any enterprising manufacture had developed a toaster that maybe keep an eye on the colour of the toast, rather than being time based. Finding nothing I settled on this one we have now, which obviously has some sort of electronic timer. It works surprising accurately.

 

[JohnW2]

I took apart a scrap modern toaster a while back. It had a 555 timer chip inside it.

That doesn't surprise me in relation to a 'modern' toaster, but I doubt that is the case with my ancient one. I say that because it does seem to work by sensing temperature, rather than by timing - since if I put a piece of frozen bread in it, it pops up when toasted to roughly the same extent as would be the case with non-frozen toast (with the same 'setting').